/ohm/ - electronics

I'd add w2aew YouTube's channel to the list, his explanation videos are pretty useful, personally I think his content is best of all the EE YT channels.

I'd also add KiCad to the PCB software list because its completely free and actually legit product that can easily compete with Eagle and such.

>>1045343
>add KiCad to the PCB software list
Long overdue. Also change the wording to "lay out circuit boards".

>>1045343
It's at max characters. We had a padtebin in the paste but it looks like we're back to dumping it in the op.

You could probably remove circuit wizard and replace it with KiCad because I don't know of a single person who actually uses circuit wizard.

I'd have to check out w2aew

>>1045403
We had a pastebin for the op*

I think I had a stroke when typing that.

File: Untitled.png (19KB, 95x77px) Image search: [Google]

19KB, 95x77px

Is there any resource for identifying connectors (proprietary stuff, maybe)?

I tested the two pins and they're just +-5V

Hello ohm. I am new to electronics and want to make a little machine that bruteforces keycard locks (for fun, not for anything criminal).

But I just want to know beforehand, is making such a device illegal? Also is it common for simple electronic locks to go into lockdown mode after too many incorrect attempts? I would imagine not since they don't really expect there to be any incorrect attempts in the first place. I mean it's not a keypad after all.

>>1045417

Who cares if it's legal or not?

File: why.png (43KB, 302x488px) Image search: [Google]

43KB, 302x488px

I did this, from Make Electronics

Author says it is supposed to do:

''... the LED should flash on for approximately
one second, and off for approximately one second.''

I used the exact same components he listed and a 9v battery.

However, when I connect it, I only see a really really small light that can barely be perceived on the led, then it powers off and doesnt flash again unless I disconnect the battery, wait a while, and do it again. But, as I said, the light can barely be perceived.

By this I believe It works, as everything is connected properly.

But whats going on here? And why do I have to disconnect the battery, wait, and connect again to see the light again? I am afraid that, no matter how much I study on capacitors, I just dont understand whats going on.

>>1045661
No, it should keep flashing indefinitely. Check the pinouts of your transistors.

>>1045661
> The light can barely be perceived.
IMO "Generic LED" means 2v 30mA or 3v 20mA. If that's what you used 1kΩ is too big of a resistor. Use 300Ω.

You should still be able to see the LED flash (faintly) even with the 1kΩ. I don't see anything wrong with the astable multivibrator circuit. " then it powers off and doesnt flash again unless I disconnect the battery" is the behavior of a monostable multivibrator, so you're on the right track but messed up somewhere.

File: postadsuk.com-technics-se-ch540-hi-fi-stero-system-amp-speakers-audio-amp-stereo.jpg (118KB, 1024x1024px) Image search: [Google]

118KB, 1024x1024px

Alright lads I'm probably way over my head with this one considering that im pretty much a beginner in electronics.

I have this old stereo and I've lost the remote so I want to "make" a new one. I already found some relatively easy guides on how to do this via arduino/RPi but it seems that this is only possible if I know the infrared protocol which the manufacturer has used for this particular device (not only the frequency).
After a thorough search online I couldn't find the exact protocol used for this particular remote, so this leads me to two scenarios, which I am not sure if they are at all possible:

1. Somehow decode the protocol from the receiver.
2. Set up a new receiver that would work with a different protocol.

Is either of those options possible at all?

>>1045718
Both are "possible". Neither are feasible.

Your best bet is to hope that the low-level protocol (i.e. data rate, framing) is reasonably standard and figure out the codes by trial and error.

>>1045661
It does not work as expected and the reason is that you wired it incorrectly, used wrong components or used broken components.
Post a picture of your actual build. Check your battery while you're at it.

>>1045749
yeah that's what I thought, sorry for my shit english.
After looking at a few protocols which I found for similar models, the codes for the basic operations (Power on/off, Vol.up, Vol. down) are actually the same!

http://lirc.sourceforge.net/remotes/technics/RAK-CH144WH
http://lirc.sourceforge.net/remotes/technics/RAK-CH215WH
(my remote should be RAK-CH204)
I will assume that the codes for these basic operations are the same for the missing remote as well and use one of these protocols - i dont need the other functions. Do you think there's something else that I have to change when using this different protocol in order to get the remote working?
Thanks a bunch

File: inf.png (29KB, 591x760px) Image search: [Google]

29KB, 591x760px

>>1045749
Supplementing >>1045757 with pic related. This part seems to be unique for every remote? I guess I'll try either way

>>1045758
> This part seems to be unique for every remote?
The timings (header, zero, one, ptrail, gap) are determined by measurement, and will have some degree of measurement error. You shouldn't assume that the differences are significant.

Also, this:
> pre_data_bits 25
> pre_data 0x80080A
(for the 144) and this:
> pre_data_bits 24
> pre_data 0x400405
(for the 215) are the same apart from a trailing zero, which may not actually be significant. It's entirely possible that some parts of the signal are ignored.

File: tegaki.png (3KB, 400x400px) Image search: [Google]

3KB, 400x400px

Can I use a Darlington transistor to amplify the current output of an op-amp?

I'm hoping that I can set R1 so that as the output voltage from the op-amp changes, the current through RL varies proportionally, i.e. Vout/R1 * current gain. Am I on the right track with this?

>>1046008
Is that supposed to be an NPN darlington? If you omit R1 altogether, the voltage across RL will be around 1.4V below opamp's output. Assuming resistive load, the current through it will be directly proportional to opamp's output voltage (minus that 1.4V).
Trying to accurately count in darlington's gain will just produce anal pain.

Is huge gain on an op-amp an issue if it'll never rail? I'm guessing it'll exacerbate noise problems but I don't care very much about that.

>>1046040
The actual gain will deviate from the "expected" gain and the bandwidth will go down. There can be stability problems due to layout etc. as well.
And then there are the output range and noise problems you mentioned.

>>1046040

of course it's an issue. you always wanna use the minimum gain that you can get away with, for reasons of noise, stability, and bandwidth.

>>1045753
>>1045667
>>1045663

Thanks for the answers.

While I cant post pictures (no camera) I still believe the error lies at the way I connected the capacitors.

Since I am using electrolytic, I dont know how I should connect them in this circuit desu, as in where does the positive and negative lead go?

>>1046096
Positive is on the positive side i.e. closer to the battery's positive terminal.

>>1046096
Positive end of the capacitor to transistor collector.

Any way to comment out components in LTSpice?

So I'm pretty much a newbie to electronics, only experience is a shitty "tech" class way the fuck back in junior high, and college physics in undergrad....

Anyway, say I want to (eventually) make something like an induction forge, what sort of path should I take to get there? Like, what sort of projects or specific field of electronics would lead me in that direction? What sort of background knowledge is required?

>>1046237
https://en.wikipedia.org/wiki/Power_electronics

Induction cooker is a high frequency inverter running to an air coil electromagnet (inductor).

>>1045779
thanks a lot, I will give it a try when I get all my components and set them up

>>1046156
Resistors: 1.23k -> 1.23mk or 1.23Gk, depending on whether you need short or open
Capacitors: 1.23u -> 1.23nu
Inductors: 1.23u -> 1.23nu

Well ok, I'm not sure if this works with LTSpice, but it works on PSpice, as it ignores stuff after the first multiplier character.
If possible, try to avoid changing the value too many decades, as it can cause simulation problems.

>>1046008
You only need to use a Darlington if you need more current gain than you can get from a single transistor. They're mostly used with high-power transistors which tend to have low gain (maybe as low as 10).

Also: when boosting an op-amp's output current with an emitter-follower, you'd normally take the negative feedback from the emitter so that feedback compensates the Vbe drop.

Anyone here fly quadcopters?
I've been trying to build a small racing FPV quad, accidentally burned out an ESC and I have no idea why it happened.

Is it possible to build a MITS Altair like computer around an Intel 8086 microprocessor and be able to manipulate it like you could the Intel 8080 microprocessor?

>>1046285

nope, you could only do that if the 8086 had the same opcodes for each instruction of the 8080, which is not the case.

>>1046287
I wasn't thinking software wise, but some of the hardware manipulations like single step and halt.

I'd think you could toggle programs into memory without the processors intervention.

>>1046284
what voltage were you running?
motor?
esc?
c-rating and capacity and battery brand?
and the prop size?

>>1046290
Well, 8086 has multi-master support, so entering software to memory via switches should be pretty simple to implement.
8086 has direct software support for single-stepping, so you don't really need extra hardware for it. AFAIK nothing really prevents you from implementing HW single stepping, though.

Hey /ohm/ I'm new to /ohm/ though I've floated around varoius boards for years. Anyone have cool power supply ideas?

I am planning to make a power supply to replace my very low power LM317 one, though I may make a beefier one with pass transistors though I don't understand how the pass transistors make it pass more current in the schematic. The 'new' one i am making is just gonna be one of those constant voltage chassis switching power supplies, fed directly into a buck converter, with some knobs n shit to adjust it with.

Are my ideas shitty, /ohm/?

>>1046237

Just make an induction heater

there has got to be a tutorial for absolute babbys on youtube about it

if you want to understand more about how it works build a boost converter.

when I was getting started out I made one that could get up to 1000 volts out of a 9 volt battery

>>1046279
Huh, ok. I was thinking that the collector->emitter voltage drop would be insignificant and therefore the emitter voltage would be the supply voltage. I've only really used MOSFETs before so if that's a difference between them I guess that explains it. It sounds like with a transistor the base->emitter voltage drop is fixed and would dictate the output voltage while Vce varies. Is that right?

Do I need to worry about using any resistors?

Also, unrelated question, is Film Capacitors the best family on Digikey to find a non-polarized cap around 28.4uF? The ones I see in there are almost a dollar apiece so I'm hoping there's a cheaper, more appropriate section.

anyone ever use Proteus?
it seems to be standard in the schools around here, but I've never seen it mentioned anywhere.

it's two programs in one.
one is a schematic layout with decent libraries and simulation.
the other is a pcb design suite for said layout, haven't used it.

the interface is a bit mediocre thought.
its quite structured, but there are a few redundancies too many and editing things like voltage rails are pretty counter intuitive.

File: 20160830_124723.jpg (2MB, 2988x5312px) Image search: [Google]

2MB, 2988x5312px

I guess this is the right thread to ask. Just moved into a new place and the previous owners had this monstrosity in the closet and everything wired to it. What exactly is it and is it worth anything?

File: 20160830_124745.jpg (2MB, 2988x5312px) Image search: [Google]

2MB, 2988x5312px

>>1046486
This is the only writing on it, and I haven't been able to find anything so far. As of now it's just a giant paperweight that the trash refuses to pick up and thrift won't accept. Am I stuck with this thing?

I got the Arduino sensor shield, and its got a spot for the APC220 radio chip, but theres one too few pins. How the hell do I connect this?

>>1046486
It's a home wiring hub. Seems pretty nice, so its probably worth about 300 retail, like 50-100 if you were to sell it.

>>1046493
Thx, gonna throw it up on craiglist for 80$/OBO or so.

>>1046453
> Huh, ok. I was thinking that the collector->emitter voltage drop would be insignificant and therefore the emitter voltage would be the supply voltage.
If the transistor is conducting, Vbe will be 0.65V (1.3V for a Darlington pair). If Vc is at the +ve rail, then Vce ~= 0 (i.e. saturation) implies Ve is also at the +ve rail, which requires Vb to be 0.65V (or 1.3V) *above* the +ve rail.

This is why there's a strong preference to put the load on the high side and the transistor on the low side. Or, if the transistor must go on the high side, to use PNP or pMOS transistor (but then you need a control voltage which can drive the base/gate close to the +ve rail to turn it off).

> I've only really used MOSFETs before
It's worse with MOSFETs, as you need Vgs to be much more than 0.65V (typically 4V or more) to get saturation (Vds ~= 0).

High-side MOSFET drivers typically include a charge-pump voltage booster in order to be able to drive the gate above the +ve rail.

> It sounds like with a transistor the base->emitter voltage drop is fixed and would dictate the output voltage while Vce varies. Is that right?
Right. The base-emitter junction of a BJT is basically a diode. Vbe will be ~0.65V if the transistor is conducting. A Darlington pair is two BJTs chained emitter-to-base, so Vbe is two diode drops.

> Do I need to worry about using any resistors?
With an emitter-follower (resistive load between emitter and any fixed-voltage point), you don't need any other resistors. Emitter (and thus load) current will stabilise at the point that Vbe is 0.65V/1.3V.

>>1046453

> Also, unrelated question, is Film Capacitors the best family on Digikey to find a non-polarized cap around 28.4uF?
At that size, look for non-polarised electrolytics. I've never seen film capacitors that large. It's basically electrolytic > ceramic > film. You normally use film if you care about stability and linearity (i.e. in the signal path), ceramic if you need more capacitance than film with low ESR/ESL, and electrolytic if you just need high capacitance.

File: file.png (4KB, 235x385px) Image search: [Google]

4KB, 235x385px

>>1046541
This setup looks like it's giving me the right waveform on the diodes. It's a PNP transistor with a low side load; what are the issues with his?

>>1046566
>>1046541
NPN*

Since you mentioned that NPN transistors tend to be used with a high-side load.

>>1046566
> what are the issues with his?
None, other than the fact that the emitter will never be above the op-amp's output minus 0.65V. So if the op-amp's positive rail is also 12V, you won't get Q1's Vce below 0.65V. But at 20mA, that doesn't really matter.

Saturating the switch is a much bigger issue when you're dealing with several amps. Particularly for MOSFETs where even "logic-level" parts need Vgs of 4V or more to saturate. If you're pushing 10A and you can't get Vds below 4V, that's 40W dissipated. Even a BJT will dissipate 6.5W, or 13W for a Darlington pair, which are more common at those currents.

>>1046617
Awesome, thanks a lot.

>>1046566
You need negative drive voltage for the opamp for that to work.
Your feedback resistors are pointlessly small.
And if you want any linearity (double the voltage = double the light), you should move R17 to the ground end and take feedback from there.

>>1046806
>double the voltage = double the light
are you implying human perception of light is e en remotely linear?

I pulled apart a shattered plasma TV I had lying around. I'm able to salvage caps, transistors, etc. from most of the boards. The buffer boards don't have anything on them for me to pull, are they worth selling or just right into the trash?

>>1047179
it's not like LED light output intensity is linear with voltage anyway.

File: experimentt.jpg (42KB, 289x462px) Image search: [Google]

42KB, 289x462px

I made a couple of post regarding this experiment from the Make Electronics book for the past few days.

I finally made it and now it works properly.

Basically the LED is turning on and off. But I dont quite get how the capacitors are working here.

I get that Capacitors store energy in the form of electric field by making one side of it more positive than the other and that, in DC current, Voltage will go up to the amount of the capacitance while current will go down as Voltage approaches the max value.

I also understand that, in AC current, capacitors allow fluctuations of current, so this means that our Capacitor is moving current back and fort right? In the sense that the voltage will go up, current will go down, then voltage goes down and current goes up right?

So how exactly is this ocillator circuit working here?

Ignoring pic related circuit, if I have a simple arrangement in which theres a capacitor and a LED with a DC source, what happens when the capacitor reaches its max level? Does it release its current to the LED? or does it Store it and thats it?

>>1047209
>Voltage will go up to the amount of the capacitance

Not exactly. You could put any voltage you want across a capacitor. The only thing that'll happen is it'll blow out if it's not rated for that high of a voltage. Capacitance dictates the charge (which is the integral of current) on each plate per volt.

C=Q/V. In DC, the plates will build up a charge that opposes the movement of current. The voltage source is producing an electric field in the circuit pointing in the direction of current flow, and the charge on the positive cap plate points the other way.

In AC, the voltage oscillates and doesn't allow as significant of a charge to build up and oppose the applied voltage. No electrons are actually crossing the gap, they're just flowing into and out of the plates on each side.

It looks like this circuit has one cap build up until the voltage across it is high enough to enable the other transistor, which drains the other cap, disabling the other transistor and allowing the other cap to charge. Resistors make capacitors charge slower. t=RC, t being the time for the capacitor to charge or discharge 1/e of the way to fully charged or discharged, respectively.

> if I have a simple arrangement in which theres a capacitor and a LED with a DC source

You never really encounter a circuit with a capacitor that doesn't have a resistor. It's an edge case. The capacitor would be expected to charge instantly since R=0 and t=0. Also, I'm not sure if the forward voltage drop of the LED would be applied, since there won't be any current.

If they're in parallel, that's still a weird circuit because there's no current limiting resistor for the LED. An LED with a typical forward voltage of like 1.3V would have Vs applied across it and probably blow up.

You should download LTSpice and start simulating stuff. It's pretty simple and you can learn a lot.

>>1045326
Hey, I didn't find a small/stupid questions thread so I'll ask here.

I'm making a small lamp with G9 Lightbulbs, and I'm about to get tha housings wired. My question is- does it matter how I solder them together in terms of anodes and kathodes? To my logic it doesn't because you can put the lightbulb both ways anyway, but maybe I'm full of shit.

>>1047261

you've answered your own question: if you can insert the bulb either way, then it's not polarized, so anode and cathode dont matter. in reality, these do matter, but the bulbs include a rectifier internally to direct current in the right direction.

>>1047179
No, I'm implying that the LED light output is proportional to current.
If you want the perceived brightness to be linear then you need something fancier, but at least simple linear current drive will be closer that than a string of LEDs driven from a constant voltage source.

File: NV_1203_Marston_FIGURE1.jpg (190KB, 600x953px) Image search: [Google]

190KB, 600x953px

>>1047209
The capacitors have no other relevant voltage threshold than their maximum voltage rating. They break if you exceed it.
The capacitors in your circuit are charged and discharged via resistors (and transistors) alternatingly. When charging, the voltage across them increases while the charging current goes down. When the current is low enough, the transistors switch states. The transistor which was previously conducting turns off and vice versa. It's the transistors which make the circuit oscillate and it's the transistors which determine how low the charging current needs to be for the circuit to switch state. The capacitor values, together with the resistors, determine how long it takes to reach the threshold.
Better explanation: https://en.wikipedia.org/wiki/Multivibrator#Operation

If you connect a LED and a capacitor in parallel and feed them from a current limited source, the voltage across the capacitor rises until the LED turns on. After that any extra current is hogged by the LED and thus the capacitor won't be charged further. The current spent on charging the capacitor delays the turn-on of the LED.

>>1047265
Thanks!
I just wanted confirmation from someone with more electrical experience than high school physics class.

File: wakemeup.jpg (18KB, 507x440px) Image search: [Google]

18KB, 507x440px

Teacher is pretty shit overall and doesn't explain anything of what we have to do. The book she uses is also pretty bad and doesn't attempt to even explain the techniques to analyze these type of problems (pretty simple ones for some of you)

I am getting my ass kicked by this. Just how do I compute the Voltages between CB BE?

Any bibliography somebody could recommend me? I am willing to study anything if it gets me out of the hole.

>>1047288
Simple analysis would just assume that VBE is constant (0.6V...0.7V, depending on who you ask).
So, you'll get 20V - 0.7V across R1, which then sets the base current Ib. Simple model for the collector current is hfe * Ib and this then causes a voltage drop in R2, so that VCE is 20V - Ic*R2.
VCB is of course VCE - VBE.

File: astable.png (47KB, 1920x1152px) Image search: [Google]

47KB, 1920x1152px

>>1047209
The fundamental property of a capacitor is Q=C*V, where Q=charge, C=capacitance, V=voltage. Current I is the rate of change of charge, I=dQ/dt=C*dV/dt <=> dV/dt=I/C.

IOW, the rate of change of voltage is proportional to the current and inversely proportional to the capacitance.

See the waveforms in pic related (reposted from the previous thread). The square waveforms (red, green) are the collector voltages, the triangular waveforms (cyan, blue) are the base voltages. The other two are the total voltages across the capacitors, i.e. the difference between the two sides.

When Q1 turns on, the collector voltage drops suddenly. The voltage across C1 cannot change instantly, so the voltage on the other side of the capacitor (i.e. the base of Q2) also drops suddenly (to minus the supply voltage), forcing Q2 off. The current through R2 charges C1 until the voltage at Q2's base becomes positive, at which point Q2 turns on, pulling Q1's base down and turning it off, starting the other half of the cycle. With Q2 on and Q1 off, the current through R1 charges C1 in the opposite direction, until the voltage reaches the supply voltage. As R1 is 100x smaller than R2, this happens 100x faster than the slow charge through R2 which determines the oscillation frequency.

>>1047288
Vbe is clamped to 0.65V for a BJT. So there's 20-0.65=19.35V across R1 => Ib=19.35/470e3=41uA.

If the transistor was saturated (Vce=0), you'd have Ic=20V/3.3k=6mA which would require a gain (beta) of 6mA/41uA=146.

The data sheet for the TN2219A gives beta as anywhere between 35 and 300. At beta=35, Ic=35*Ib=1.4mA, which would drop 1.4mA*3.3K=4.7V across R2, giving Vce=20-4.7=15.3V. At beta=300, the transistor would saturate (Vce=0).

Which is why circuits which use a transistor as a linear amplifier always use some kind of negative feedback, so that the behaviour of the circuit doesn't vary in direct proportion to beta.

If the transistor is used as a switch, you determine the base current which is sufficient to saturate the transistor at the minimum specified beta.

File: 1472170036834[1].jpg (287KB, 2048x1536px) Image search: [Google]

287KB, 2048x1536px

Currently I'm trying to sort out an electronic scooter I just purchased.

200w (4000 rpm) Engine.

24V battery (upgraded from 10ah to 13ah, but it still only runs for an hour)

It can take me up hills barely, So I was thinking of upgrading to a more power-consistent Li-Ion. I believe they have a high average potential as they are used. They are expensive as fuck though, I need to attach some rated at 3.7 V and 5AH in parallel and series in order to get:

28.8V and 10ah Li-ion.

Can anyone tell me if 10ah is gonna be enough or will it be barely different from the lead-acid 13ah.

Another thing I was gonna do was add some super capacitors in series with the batteries in order to put out a temporarily boosted power (only by 2V or something). Will this by any chance fuck up my motor / Li-ion batteries?

>>1047370
Changing the batteries won't make the motor more powerful.

Increasing the voltage will likely just burn out the motor and/or the controller.

File: fuse.jpg (1MB, 2560x1920px) Image search: [Google]

1MB, 2560x1920px

im relatively new to practical electronic stuff and im trying to repair some old PC components and laptops which i collect, so i hope you dont mind my retarded questions
i have 2 compaq aero 8000 handheld PCs which, aside from occasional flashing LEDs, were completely dead when i bought them
yesterday i was testing stuff on their motherboards with a multimeter and i found the problem (which is strangely same on both boards): an smd fuse has blown (pic related, circled in red). i just jumped it with a small metal clamp and both of them booted just fine
one of them let out a smoke from the area where the LEDs are, it worked after that though, i dont know if i shorted something (they were disassembled so maybe something touched together)
anyway here are my questions:
1. what could be the cause of the blown fuse? on pic related, right above the blown fuse is the battery connector, does it have something to do with battery condition - cause both have empty, equally fucked battery (10+ years of age), or maybe someone tried to remove it while running?
2. if i was to replace the fuse, how would i know what kind to use? only marking here is the "50" but i dont know what that is. could i measure current with the multimeter and then estimate how much should the fuse be rated for?
3. is it safe enough to just run without the fuse, shorting the two contacts? i would just like to test them and see if i can charge the batteries, because if one board turns out to be not working, i can just take fuse from it and replace it on the other board

File: 2016-09-01 14.33.40.jpg (1MB, 2560x1920px) Image search: [Google]

1MB, 2560x1920px

>>1047459
also here is the whole board, if someone is interested or specialized in this kind of hardware

>>1047457
Youre wrong. Il be able to squeeze out a little more power because this thing uses 8A, which is already most of my Lead-Acid's usaable power.

The Li-ions will provide me with a slightly higher average voltage for longer. I need like 500$ worth of li ion to make this a decent bike :/

>>1047459
The powersections are usually the first to go on any sort of electronics, so says my tutor.

You can run without the fuse if you are confident enough that it wont blow again.

As for why it blew, it could be a electronic issue, though I doubt that, these things happen, fuses are delicate remember, they are afterall very thin and delicate pieces of wire.

Anyone used an APC220 on Windows 10? I think my chips are both good, because I can read and write to them, but I can't detect any serial transmissions through the monitor.

>>1047471
any tips on how should i measure for possible replacement (if i even can)?

>>1047484
Circuit diagram would be your best bet. Alternatively follow the circuit to another component and see if you can find out its max ratings. Should be easy if its a transistor or anything.

File: cap.png (432KB, 690x417px) Image search: [Google]

432KB, 690x417px

Could someone help me with pic related? I believe it's a capacitor but not 100% sure. I'm trying to see if I can repair this old vintage car stereo that's completely silent and this was what I first saw.

Is it a cap? If yes, what kind? Bipolar? Those markings mean 0.47uf50V right?

The cap blew because it's sitting right against the internal light bulb.

>>1047498
Yeah, sure, it looks like 0.47u 50V capacitor. Probably a mylar (polyester) capacitor, but I doubt the material is critical.

>>1047497
thanks, ill try the second approach because as with 90% of old electronics (especially laptops), there is no circuit diagram publicly available

>>1047500
Thanks, will pick one up tomorrow and see if the stereo wakes up.

>>1047498
>Doesnt even know what a capacitor looks like.

Thats good bruh keep learning, we were all there at one point.

>>1047539
Well I was pretty sure, just needed confirmation. I mostly mess around with quite a bit newer electronics so wasn't sure if it was some weird old component that just looked like a cap.

File: Screenshot_20160901-233028.png (251KB, 1280x720px) Image search: [Google]

251KB, 1280x720px

Guys is a ksc1845 a good replacement transistor for a 2sc1313R

It is used in a old phonopreamp. And I add the part of the scheme where it's used in. I found some articles where they used it as a replacement in reel to reel amplifiers. Will it work here?

>>1047550
why dont you compare datasheets?

File: 46480417_614.jpg (38KB, 614x460px) Image search: [Google]

38KB, 614x460px

Is the Tektronix T912 oscilloscope worth picking up if I only plan to use it with Arduino and ARM microcontrollers?

If it would be good for that application, what's the max you would pay for one of these? Dude wants over $100 but I am going to try to haggle a good price.

I've been looking into digital scopes and they appear to be a shit, so I'm seriously considering grabbing this unless it's not a good unit.

>>1047568
>application
a good old analog scope won't capture data so if you want to use it to analyze data and shit then no you will struggle.
if you want to look at e.g. pwm pulse width or duty shit like that then sure go for it.
digital scopes are complicated, if you want one you need to seriously do your fucking homework or what you need/want.

>>1047573
Interesting. Thanks for the heads up. I shall read moar

>>1047568
You're probably paying extra for "portability".
Like the other anon said, it depends on your use whether it's good or not.
On the plus side is that it does not suffer from aliasing (stuff looks correct at all sweep rates) and that the screen update rate can be higher.

Hi there /ohm/, relative noob to electronics here.

I'm working on a little electronics project here I'm hoping to manufacture a few of and start a little business, and I'm fairly confident in the design. I've got split-out parts of it functioning on breadboards and I'm pretty comfortable with my design in the parts that I can't immediately breadboard.

I'm interested in finding a firm or a freelancer who is more professionally versed in electronic engineering and PCB design in general, and I'd like to know how to find this particular kind of contact.

I've found a few instances of large engineering companies that appear prepared to take on large-scale clients, but I figure there's got to be something a bit more low-key (and that isn't tens of thousands of dollars for 10 hours) where I can just have my design reviewed and get a "yes"/"no" from a professional. Perhaps an agreement for some extra hours making little adjustments to improve the functionality of it..?

I'm wondering if y'all would know what or who I should be searching for in particular to get this kind of reassurance that my project is on-track and sensible, or if I should just suck it up and trust in my design and existing testing before I dump some cash into producing a few of them.

Freelance EEs? Hobbiest-level PCB fab houses with consulting services? Any advice on where I ought to look for a professional review would be helpful.

If it helps, I'm in the Portland, Oregon area.

Thanks!

>>1047652
You can try smaller (down to one-person) engineering offices. If even that is too expensive, then nearby college's bulletin board or w/e.

If you're going to sell your stuff to consumers, there are the emission tests and whatnot if you want to (try to) cover your ass. They aren't cheap, even less so if you have to repeat them due to shitty design.

When I am performing simple operations on circuits like Ohms law, and they give me resistors in kOhms. Do I just perform the operation as in, for example:

V = 20 A * 200K

Or do I need to always do it as 20A * 200,000 ?

>>1047670
200k = 200,000 = 200E+3

of course the magnitude is important! 200k does not equal 200!

>>1047670
Generally, current will be in mA and resistance in kOhm so they will cancel out.

>>1047568

If you are doing digital logic get a logic analyzer. They offer higher bandwidth, way more channels, and are cheaper but only will work for digital signals

If you still want an oscilloscope why not get something that isn't considered an antique?

> If you still want an oscilloscope why not get something that isn't considered an antique?

Coz going without food for 2 months would probably be unhealthy.

>>1047652
You could do something like CrowdSupply but they take a cut.

Got a local makerspace? They might have someone willing to provide advice.

complete and utter noob here. I want to make a vibrator (sex toy) that has both a vibrating motor as well as a rotating motor, and I want to be able to scroll through preset patters with a button. I assume this is something that would need a plc? Can anyone suggest some reading towards accomplishing this? Nothing too in-depth, just looking to get enough info to complete this job

>>1047710
Mostly because I think it looks kind of badass. But thanks for that info, I will look into logic analyzers.

>>1047755
Different anon here. A 'real' logic analyzer is probably overkill, and personally I don't find them that useful compared to a regular digital scope. Something like a Bus Pirate will probably serve you well starting out.

>>1047561
I can't find a complete Datasheet for the 2SC1312R

>>1047774
Search without R then, it should be just the gain category.
And it's not like you need to match every single parameter for it to work decently.

>>1047670
>4 million volts
Anon what are you doing I don't think you should be messing around with 4mil volts

I doubt you even have an 80MW supply

>>1047782
I have a Cassettedeck with 6 2SC1312 in it. I'll try the KSC1845 in the Cassettedeck and I will see what the results are. The cassettedeck is noisy as fuck at the moment, so a good test subject :)!

File: IMG_0474.jpg (2MB, 3264x2448px) Image search: [Google]

2MB, 3264x2448px

My key fob is fucked up. The unlock button (Top right) keeps getting stuck in the down position, and it won't actually work until it's released plus none of the other buttons work while it's stuck down
Think this is fixable by cleaning or some shit, or will I have to try and replace it wholesale?
I'm not great at soldering small shit, plus any electrical components I have to order online and wait as there are no local suppliers.

Like fuck I'm paying the dealership $100 for a new one though

>>1047857
Bruh you dont have to bother with fixing this fiddley shit. Just shoot it with an emp blaster everytime you wanna open the door.

>>1047867
I doubt that runs off a button cell battery and fits easily in one's pocket

File: SMD-4-4-1.5-Tactile-Tiny-Button-Switch.jpg (24KB, 600x496px) Image search: [Google]

24KB, 600x496px

>>1047857

spray it with some alcohol and then brush with an old toothbrush. let it dry then repeat.

the sure-fire solution is to replace the switch, but those very thin ones are not so easy to find. here's one place i found after a quick search: http://www.electrodragon.com/product/smd-441-5-tactile-tiny-button-switch/

is there a way to continue a circuit through a rotating moving part?

>>1047952
Thanks, will give that a shot

File: 144799-8147024.jpg (21KB, 600x600px) Image search: [Google]

21KB, 600x600px

>>1047965
Slip rings (several ways), transformer which allows rotation, optical power/data transfer, radio-based solutions.

Anyone got experience with wirewrapping? I have an old Rotel amp that makes quite a loud pop when powered on and after opening it up i noticed that most connections between boards were done with wraps. Do wirewraps deteriorate with time? The amp is from 1982 so I'm anyway going to do a partial recap of it and also open up the volume pot and do a proper cleaning on it.

If wraps are prone to suffer with time I could just add some solder onto the joints while I do all the other repairs.

>>1047983
All old amp's pop when startup

>>1048002
But this one has gotten significantly worse over the past months. I'm fine with a small pop when powering on but it's now at the point where it's first a very loud pop, followed by a random crackle for a few second and then multiple pops before it calms down to normal. Sounds like someone is inserting speaker wires while the amp is playing.

>>1047774
Looks like 2SC1312 is a standard NPN with high beta. You could try the BC547C as a starting point.

>>1048018
Start by replacing the electrolytic caps. I'm assuming it's a solid-state amp?

>>1048021
I thought tube amps generally didn't pop on power-on.

>>1047744
The preset stuff will force you to use some kind of micro controller. You've a long road ahead of you, friend.

>>1048028
Do they still sell Arduino Nanos? Cause one of those would actually be perfect.

>>1048026
They usually do if they have standby switches.

>>1047978
thanks, that opened up a lot towards the googling

File: 20160902_165554.jpg (3MB, 4128x2322px) Image search: [Google]

3MB, 4128x2322px

Hey /diy/nosaurs. I have a neon sign that is acting up. It was flicking on and off rhythmically about .5 seconds on, .5 seconds off. Figured it was a capacitor so I opened up the power supply and looky there! Mr. Capacitor leaked all the magic fluid that makes him work. So I need to order some new ones.

So my question to you is, are these AC or DC capacitors? How can I tell?

File: AlwaysSunnyPhillyStars.png (410KB, 616x344px) Image search: [Google]

410KB, 616x344px

>>1048054
see that white stripe down the sides? that indicates the negative side, they're meant for dc use. most bad caps people encounter will be cylindrical electrolytics. there are bipolar ones but they will be clearly labeled. just try to match the voltage and capacitance ratings. it's okay to go higher on either or both but if you do the capacitor tends to be larger and you'll have a harder time fitting it in the space that you have.

>>1048021
You're assuming correctly. I started desoldering the main power caps and they have 3 leads, are they maybe 2in1 caps that share an electrode?

>>1048060

Thanks famalam. Going to order some 200V 100uF caps now on Chinebay.

>>1048066
post pics of them

>>1048060

One more question, I found some 150uF caps on ebay (200V). Can I use these with no problems?

Just got an arduino recently and gonna start fucking round with it.
I have a fair amount of programming experience but I'm just unsure about one thing in the Arduino IDE.

If I want to use the shiftout method to sent a byte to a shift register, I know the method can take arguments for the pin to output on, the clock pin, the bit order and the data.
My question is what format can the data take?
Can I just write it as a string of 8 binary digits like 00110111?

>>1048071
As long as they're through hole electrolytic caps then yeah.
Capacitor plague is the biggest load of shite that ever happened.
Bunch of corrupt wankers covering each others tracks

>>1048080
Capacitor plague? No comprede amigo.

Is this about the shitty caps that came out years ago from a copied Japanese design?

>>1048082
Yup
https://en.wikipedia.org/wiki/Capacitor_plague#Responsibility

>>1048071
they may be fatter and or taller than the originals, other than that you should be fine

>>1048085
Interesting. I have heard about it but I never actually researched it.

>>1048086
Yeah, I'm not sure of the size of these, but I'm going to give it a go. Worst case scenario I'm out $6.

File: IMG_20160902_224809.jpg (2MB, 3024x4032px) Image search: [Google]

2MB, 3024x4032px

>>1048069
Two of these

>>1048094
Can't really get a sense of scale there but it's probably a dummy pin.
It's there either to stop large caps from pulling the barrels out of the through holes or to stop retards putting them in the wrong polarity

>>1048096
Quite likely, two of the pins seem to be connected to the same plane on the pcb with only one pin going to a separate pad.

The amp is a rotel ra-840 (1st gen) btw.

>>1048094
So two of the leads on that type should be shorted together. This is done to reduce parasitic inductance in the capacitor and thus reduce noise. This is not really a popular package any more, so getting replacements will be tough. You can use regular two-leaded caps, but they will not perform as well as these did in their prime.

>>1048103
But replacing them with standard two lead electrolytics will still have the amp working, right? I'm way far from an audiophile, just want my amp upgraded to caps from this century.

>>1048106
Yeah definitely.

File: 1446603727731.png (19KB, 998x549px) Image search: [Google]

19KB, 998x549px

>get a cheap audio dac ic off ebay
>it can't be that hard to implement i2s, can it?
welp

>>1048076
you can write it as 0b00110111 or 55 or 0x37. it all compiles to the same thing.

File: New-Black-Push-Button-Mini-Switch-6A-10A-110V-250V-KCD1-101-2Pin-Snap-in-On.jpg (118KB, 800x800px) Image search: [Google]

118KB, 800x800px

Noob question: how do I properly turn off a circuit with an ATTiny on it? Should I just stick a picrelated somewhere around the battery?

>>1048425
Yeah, a switch is a very simple and reliable way to turn things on and off.

Alternatively, you can play with your processor's power save modes, use the shutdown features of your other ICs and so on, but it's much more tedious way to achieve the same. You can also do it with your own hardware or use a power management IC with "power button" functionality, but again, it's a more tedious option than a simple switch.
The good thing with these more complex arrangements is that you can turn your gadget on and off automatically, if needed.

>>1048432
I'll just use a switch then, thanks!

File: pepepcb.jpg (205KB, 1099x633px) Image search: [Google]

205KB, 1099x633px

Do you think people at the PCB factory will say anything when I send this to be produced?

>>1048229
>i2's

What is it?

>>1048451
They'll tip their fedoras to your 1337 meme skillz.

>>1048451
https://www.reddit.com/r/mylittlepony/comments/1ut74g/in_about_15_minutes_princess_luna_is_lifting_off/
https://www.reddit.com/r/mylittlepony/comments/2ry7u0/when_this_post_is_30_minutes_old_princess/
If this guy didn't lose his job (not that he would have had any reason to) you'll do just fine

File: TzJSDRA[1].jpg (1000KB, 4147x3110px) Image search: [Google]

1000KB, 4147x3110px

>>1048464

>>1048451
They will wonder why most of your component labels are fucked up, and then send the finished product to you without returning payment.

>>1048471
But anon, you just need it to solder the right component in the right place :^)

>>1048452
3 wire digital audio bus. clock, data, and a word clock for word length agnosticism. it's very simple, but evidently not simple enough for me.

>>1048483
With IC's you just need to plug wires into the correct places. How could you fuck this up??

I mistakenly got this op-amp off Digikey without noticing that it's a two-circuit, single supply.

https://www.fairchildsemi.com/datasheets/LM/LM258A.pdf

The datasheet says "operation from split supplies is possible" but I don't see how from the block diagram. Can I use this op-amp as a regular single circuit two-sided supply op-amp somehow?

>>1048491

Just connect the gnd pin to the negative supply, tada split supply opamp.

>>1048494
Great, thank you.

>>1048487
presumably the fault is somewhere in my beautiful niggercode

http://hastebin.com/ofivikenul.cpp

i don't expect anyone to actually read it but you seemed curious

>>1048497
>http://hastebin.com/ofivikenul.cpp

Is this not a automatic DAC system?

>>1048491
remember to ground unused pins.

>>1048506
not sure exactly what you mean. it's just an atmega and a tda1543 (plus datasheet circuitry). i'm repurposing it for a digitally controlled power supply, not audio like it was meant for.

>>1048506

Bad idea with unused opamps. Ground +in, conect out to -in.

>>1048507
meant for >>1048504

>>1048497
Can't say I really thought out how it is supposed to work, but
- why is your for loop running 65 times?
- line 30 seems to produce an unsymmetrical wc pulse
- does line 34 really happen at correct time?

In general, why all this state machine like playing with i? If it does not work, why not start with something really straightforward? You know, like using two for loops (one for the left and one for the right channel), making them run 16 times, toggling clock manually high and low during each iteration and toggling wc between the loops?

>>1048507
Youre using a built in DAC. Was that output you showed my meant to be analogue?

File: Total circuit.png (21KB, 1256x631px) Image search: [Google]

21KB, 1256x631px

Its for an electronic scooter. Any problems ?

>>1048612
the bit clock toggles twice for every data bit, so 16*2*2=64. the 65th loop maintains the timing on the last clock cycle. the wc line is driven low before the LSB is written so it is offset but not asymmetrical. i've verified the timing, i'm just fucking up the data bits at some point (probably not an issue with line 34 because i'm testing the left channel, and bad timing there would only impact the lesser bits).

>>1048625
yes, the graph is an integer 10b adc readout of the dac output. i was too lazy to scale it.

i should be clear i'm not seeking help, i vaguely know what's wrong. i just wanted to share my failure.

>>1048648
It's a mess. What are those transistors supposed to be? NPN? Mosfets? Where's ground/reference point and where are those switches going? What is this shit supposed to do?

>>1048625
Huh? He has an external I2S DAC. ADC is apparently internal.

File: Total circuit.png (24KB, 1256x631px) Image search: [Google]

24KB, 1256x631px

>>1048654

>>1048656
What's the cap used for? Why is it not connected to the battery during "use"?

>>1048656
Also put some damn resistors in, for the love of all that's nice in transistors

>>1048656
I'll assume +battery refers to the potential at leftmost transistor's emitter and that the ground is motor's -.

The leftmost transistor can't be turned on with +battery voltage.
For the capacitor to do anything it has to be comically big.
You'll risk burning the capacitor charge transistor if you have no base resistor.
Your current paths waste 3 diode drops.

>>1048666
The cap supplies extra power to the motor for periods of up to 3 minutes. More power when going up hill.
Its just a theory that it will scale hills better with a temporary boost of power.

RE the wiring, it needs to be done that way when in use it cannot charge otherwise short circuit.

>>1048667
The charge controller draws a certain current, so its resistance is variable depending on the throttle. I just need the transistors to act as a switch.

The transistors are power transistors specced for this current and voltage they should be expecting.

>>1048677
>I'll assume +battery refers to the potential at leftmost transistor's emitter and that the ground is motor's -.

Its connected to a high resistance line from the battery, so its a voltage of very small current, but enough to turn on the transistor.

>The leftmost transistor can't be turned on with +battery voltage.

it can its npn low base current.

>For the capacitor to do anything it has to be comically big.

Its gonna be a few super caps in parallel. Total of 1500 F, 2.5 volts (though il draw like 5-10W from it)

>You'll risk burning the capacitor charge transistor if you have no base resistor.

Yes I didnt include the resistor for the switches, but there will mostly certainly be some. I will use very tiny current to turn it on.

>Your current paths waste 3 diode drops.
which ones?

Oh boy, you have no idea what you're doing
>The cap will do nothing and last even less
>Base current will destroy the transistors
>Voltage is completely up the ass in the whole thing

>>1048733
What is that supposed to mean?

File: RoboArm-Output.png (778KB, 2000x1482px) Image search: [Google]

778KB, 2000x1482px

Anyone experienced with EMG amplifiers here?

I just populated the PCB for the attached circuit but I can't seem to sense any muscle activity. I have a feeling it is because my reference voltage (V-REF) is not set up correctly...
With that said can anyone comment on the circuit?

>>1048753
Mate there is too much going on here.

Have you tried prodding the various points with a multimeter

>>1048753
There's very little gain after the instrument amplifier and if you increase the gain of U1 to compensate, it just saturates due to DC voltage difference at its inputs.
It would be a good idea to connect V-REF to your body as well.

>>1048451

Legend ink gets printed after solder mask is cured, so enjoy soldering those pads with ink sputtered all over them.

File: outpins.png (395KB, 920x1290px) Image search: [Google]

395KB, 920x1290px

Is there any functional difference between two types of pin sockets in picrelated?
(besides being able to break off bottom ones)

>>1048844
You definately want to go for the smaller ones. Big ones you may not have enough room to attach (one day)

>>1048844
Machine pins are 'precision' and are a sign of higher quality but doubt the price is worth it. You may be able to stick them in a breadboard whereas the other kind is usually too wide to fit.

File: mpin6ra-dscn3999.jpg (35KB, 800x600px) Image search: [Google]

35KB, 800x600px

>>1048844
I find it a lot more difficult to get male pins to fit into the circular sockets in the bottom picture. Male header pins are squarish not circular. I don't know why people pay so much more for something that doesn't work as well.

File: Break-Away-Header-Male-2.54-mm-Around-1.jpg (17KB, 600x290px) Image search: [Google]

17KB, 600x290px

>>1048861
Yeah, generally you shouldn't cram square pin headers to them. You use the circular sockets with components, certain kinds of cable and shit like pic related.

File: ir_repeater.png (53KB, 1122x629px) Image search: [Google]

53KB, 1122x629px

i don't know what i'm doing

>>1048903
The unregulated voltage will blow your shit to boomtown

>>1048911
I don't think the coin cell battery will have enough power even for the poptown.

>>1048912
Not boomtown as in "you-stuck-this-chip-to-mains" boomtown, boomtown as in "your chip is fucked and it does not work"

>>1048914
There are a lot of projects online with ATTiny85s connected straight to CR2025/CR2032 cells. They seem to work pretty well.

>>1048762
My mistake was jumping straight into the circuit from simulation, should have physically build it up block by block and tested it. That's what I'm doing now...

>>1048768
Righto, there is definitely saturation happening somewhere from what my oscilloscope tells me. Now to find where it is...

>>1048903
your circuit is fine but ir has a broad application span so it's easy to get receivers and leds that won't work so well together. be sure you get the right parts. the attiny timer module should get you very close to whatever carrier frequency your receivers want.

File: Wiring.jpg (251KB, 2558x1284px) Image search: [Google]

251KB, 2558x1284px

I am making a 3d printer with AC 230V 750W heated bed. Usually people just use a Solid state relay and that is it. But I have read that if SSR fails, it shorts, which means that bed heater would keep heating. And that could damage a printer, a heater or even set something on fire. My idea was to use extra power relay that would turn on when the print starts and turn off when it ends. It also turns off if temperature starts to rise uncontrollably (if SSR fails). And if power relay fails it would stay open. Pic related. I will still have to figure out how to set the firmware that it would control the power relay, but i think it can be done.
I am not an electrician so my question is: will this work? did I forget something? should I consider some other properties of relays that might cause problems?

Manual switch with fuse: http://www.ebay.com/itm/10A-250V-Inlet-Module-Plug-Fuse-Switch-Male-Power-Socket-3-Pin-IEC320-C14-/252163480447
Power relay: http://www.ebay.com/itm/OMRON-ELECTRONIC-COMPONENTS-G5RL-1E-HR-5DC-RELAY-PCB-SPCO-5VDC-/271384923720?hash=item3f2fcd4e48:g:ta8AAOxyDEVSa74j
SSR: http://www.ebay.com/itm/Fotek-SSR-40-DA-Solid-State-Relay-/271897072354
Heater: http://www.ebay.com/itm/30X30CM-750W-220V-Silicone-Heater-3D-Printer-Heated-Bed-Pad-w-3M-Thermistor-/281642076373?hash=item41932d08d5:g:4FUAAOSw7FRWWa6X

File: 1466480467128.png (21KB, 414x390px) Image search: [Google]

21KB, 414x390px

>>1048939
drop the SSR entirely. the mechanical relay is inherently safe as you say, and it won't dissipate as much power as the SSR will at 750W.

you need special driver circuitry for larger mechanical relays. SSR triacs are easy to trigger, but now you're looking at an inductive load that can draw a fair bit of current. if your control board doesn't have these built in, you'll need a flyback diode and a small driver transistor (to-92 package is fine).

>>1048941
by the way if you're programming some of this shit yourself be sure to include hysteresis in your bed temperature control so you don't get relay chatter

File: TSOP-1738.jpg (21KB, 276x182px) Image search: [Google]

21KB, 276x182px

>>1048937
Not sure about the LED, but the receiver I'll be using is the same as the one in the target device. So everything should work alright, hopefully.

>>1048923
Your component values make me wonder how strong input signal you're expecting. EMG is typically weak, much less than 1mV, from smaller muscles.

>>1048946
with the leds just keep their wavelength and view angle in mind. having the wavelength a bit off is fine but if you have too broad a view angle it'll kill your transmission distance, as i know from experience ;_;

interestingly you can have different carrier frequencies on tx and rx as long as you control the leds' pwm frequency. not that you'd need that.

File: SCR_1473004498.png (99KB, 551x268px) Image search: [Google]

99KB, 551x268px

And you're tellin' me I can get three of these bad boys for 3.30$ on OSH Park? That's a sick deal.

>>1048941
I am using Ramps 1.4 board with Arduino mega 2560. 12V output for heated bed is rated for 11A (as it is meant do power DC heater) and that would control SSR. I wanted to drive power relay with i/o pin on arduino (5V). Now I found out that power relay I choose draws too much power. Arduino is limited to 40mA per pin. I found these on ebay: http://www.ebay.com/itm/New-One-1-Channel-Isolated-5V-Relay-Module-With-Optocoupler-For-Arduino-/131209937830?hash=item1e8cb8c3a6:g:MBgAAMXQyFFTkpRo It is supposed to be designed for arduino with only 20mA current. And it looks like it has that circuit you were talking about (what I see from the pictures. There is no schematics provided). It is very cheap and I don't know how reliable are these chinese stuff?

I wanted to use SSR because there would be a lot of switching on and off. Marlin firmware has this option:
//#define SLOW_PWM_HEATERS // PWM with very low frequency (roughly 0.125Hz=8s) and minimum state time of approximately 1s useful for heaters driven by a relay
If I understand this correctly It checks for temperature every 8 seconds and if it is too low it gives power. So relay could be switched every 8s. On 10h long print that could be 4500 switches. And that every day for 2 years is 3 million cycles. That I think is too much for power relay (I did overestimate this numbers just to be safe).

Keep in mind that I am very new to this, so any advice is more than welcome.

>>1048844

the bottom ones are much nicer

I keep a lot on hand for making perfboards. you can snap them down to any size easily and they mate with these >>1048861 or the straight variant really easily.

I'm gonna ask some real dumb questions right now, please bear with me

My vibrator takes 2 AA batteries (1.3v), that means the motor would be 2.6v right?

An arduino nano puts out 5v, does that mean that I could have 2x 2.5v motors running off it? There's going to be 3 buttons as input to the arduino, will the addition overdraw the power supply?

>>1049003
the chinese relay board you linked will work fine. it may have dubious isolation in the pcb layout and its relay may have a shorter life than one you've handpicked but there's not many other corners there for them to cut.

you make a good point with the switching frequency which i didn't think about enough. SSRs are definitely at an advantage there as they can switch each mains cycle with no meaningful additional losses or fatigue. depending on how much thermal mass is in the heat pad you may be fine decreasing the switching frequency significantly. you can evaluate that with either a simple heat transfer calculation or by just wiring up the hot bed and measuring it.

you could use a thermal fuse thermally bonded to the pad, which would blow if the SSR fails closed. i've never used one though so i can't give advice there.

i don't believe your proposed solution would work very well without a latchup detection circuit to automate shutoff, because prints can take hours and in the meanwhile your thermal pad is burning up due to the failure.

>>1049044
i always forget something in my post. you can *significantly* increase the life of your SSR by adequately heatsinking it. it'll be dissipating at least 4W, and heat is what usually kills them.

>>1049040

Yes if they are in series

Your arduino puts out 5 volts but it is very low power because it will be at low current. This is typical of digital systems. You need to build a simple power amplifier with some kind of switching transistor if you want to control something like a small motor.

Another thing you need to understand is how an arduino will manage power: it will never destroy itself. Really you don't need to worry about it. If you provide an atmega328p a supply that is near the minimum amount of current it can run off of, this will effect the output current.

>>1049040
pls b in boston

>>1049040
you might want to elaborate more on how exactly you want to combine your sex toy with the arduino

the arduino won't run off two AAs at all so what's your new power source?

>>1049040
> My vibrator takes 2 AA batteries (1.3v), that means the motor would be 2.6v right?
Primary (non-rechargeable) cells are 1.5V, so unless it has a non-replaceable secondary (rechargeable) cell the motor will almost certainly handle 3V.

But motors are limited by current more than voltage. You can generally run a motor well above its nominal voltage provided that it's unloaded, as it won't draw much current until you start to load it.

A spinning motor acts like a generator. The generated voltage ("back EMF") is proportional to the rotation speed and opposes the supply voltage. So for a given supply voltage, the faster a motor spins, the less current it draws.

> An arduino nano puts out 5v, does that mean that I could have 2x 2.5v motors running off it?
You can't drive a motor directly from a logic output.

If you're driving them from a microcontroller, you'd typically put them in parallel and use PWM to control the power. That way, you're not limited to 2.5V (and you can control the motors individually). Motors are particularly amenable to be driven via PWM, as they're inductive loads.

Don't forget the flyback diodes.

> There's going to be 3 buttons as input to the arduino, will the addition overdraw the power supply?
Buttons don't matter. But the PSU needs to be able to power the arduino and the motors.

>>1049048
thank you

>>1049062
the idea is a sex toy with two motors, one providing vibration, one providing rotation. the rotation part would be pressed up against the skin, so it would be low speed high strength. the arduino would be for presets

I figured it would use lithium cells for power

>>1049064
thank you, appreciate it

>>1049066
2 lithium cells should be good. it sounds like you're talking about something bigger like a hitachi so a 5V wall wart could be preferable. if you haven't picked out your higher torque motor yet you shouldn't settle on a voltage.

i think the electrical aspect is going to be the least of your worries though because i have no idea how you'd rig a rotating head onto an already existing product.

I'm about to graduate with a bachelors in electrical engineering but I felt like the classes taught me maybe 20% of what I know. I learned most of what I know from side projects.

Anyone else feel like their degree is nothing more than an expensive piece of paper?

>>1049071
i'm a meche and i've only ever used knowledge from maybe two classes i took in college. all the degree was good for was getting me a job.

people argue that college provides foundational knowledge but personally i think that's just a bullshit justification. i got more use out of a single highschool CAD class i took than i did out of my 4 years in college.

>>1049068
I'd be concerned about skin stretching or tearing as well. Vibration shouldn't have more than a few mm displacement which skin can stretch to easily but depending on the size of the head and rotation speed, it could get messy.

Please post pics

>>1049044
>>1049045
Heater has a termistor built in. Or I could mount one on the bed separately. So the temperature of the bed is always measured. And firmware already has some safety features. For example if temperature reaches too high temperature or if it is sending power to the heated bed but the temperature doesn't change it also knows something is wrong. And stops the print. This all works fine when bed is powered with 12v DC from the board. But if SSR shorts it doesn't matter if it knows that temperature too high it has no way of stopping it. That is why I would change the firmware so if any of the safety features gets triggered it turns off power relay and therefore cuts power from SSR.

Yes I was planing to use heat sink.

>>1049075
Went to a shit college.

>>1049080
alright it sounds like you know what you're doing. your drawing said the mechanical relay would toggle only at power on and off. as long as you're monitoring the thermistor as well for emergency shutoff you'll be good.

if i were you i'd look into thermal fuses more though. that seems much more convenient than a double relay setup.

>>1049081
to the contrary, i'd say i got a great deal. i graduate with no debt and i now have the same job every other meche has, at one of the large aerospace corporations.

>>1049086
>to the contrary, i'd say i got a great deal. i graduate with no debt and i now have the same job every other meche has, at one of the large aerospace corporations.

How does that dispute you went to a shit college?

>>1049095
i'm disputing the premise of your post which is that the quality of a college is in any way related to the quality of the education it provides and not just how it impacts your lifetime earning potential.

unless you're the type of social savant that can benefit from "networking" in college (i suspect more people say they've done this than have actually done it), there's really no reason to prefer more expensive schools. after you've finished your first job nobody cares where you went to school, so all you carry with you is your student loan debt.

no offense but i'm not going to argue this topic further. it's too subjective, and on the internet everyone is making six figures and went to MIT so it's hard to have an honest discussion.

>>1049103
>quality of a college is in any way related to the quality of the education

How is it not related? Keep on making up excuses for going to a shit school. Working at a large aerospace corporation as a mechanical engineer is not an accomplishment. Those are the dumping grounds for the mediocre. There is no reason to be upset at having a boring, middle class life.

>>1049106
oh you're just trolling now, nevermind. try /tv/.

>>1049103
There is no reason to be upset, but don't live in self-delusions.

>>1049103
I doubt you did any "networking", so what else does a college provide? Education, and clearly your college was lacking in that department, or at least the engineering school. Face it, you didn't go to a very good school, which is fine. Clearly you got a job, but don't expect to be successfully as an engineer.

>>1049109
>trolling

Yes, I said something that you don't like, so I must be trolling. This is the mind of someone who belongs right where they are, mediocrity.

>>1049109
Someone is awfully insecure resorting to HE IS TROLLING, HE MUST BE.

>>1049103
What school did you go to then? You don't care about it, then it will be easy to disclose.

>>1049103
>after you've finished your first job nobody cares where you went to school, so all you carry with you is your student loan debt.

Exactly what a poorfag would say. Keep on limiting yourself and continue the cycle.

>>1049113
all a college offers is a first job. if you're putting yourself into tens of thousands in debt, you're going to an objectivelt worse school than one which you can graduate from with no debt at all.

most engineers, myself included, haven't done any serious networking. that's the only added benefit the ivy schools offer.

>>1049116
>>1049118
>>1049121
oh well, i guess no board is safe from the edgy college students that really truly are going to be CEO of spacex one day.

>>1049109
Nah, you seem like the kind of person that would go to /tv/, delusional.

>>1049123
>edgy

I got a place for you /b/. You seem infantile by calling everyone "edgy" or a "troll".

>>1049123
>going to be CEO of spacex one day.

At least they have a dream. You are a white collar desk jockey.

>>1049123
>all a college offers is a first job

Talk about edgy.

>>1049129
if you're honestly just a student i'd warn you to temper that dream. it'll only lead to depression. there's a less than 1% chance that any single graduate is going to manage anything better than middle management at age 40, and you need to look long and hard at whether you're in that top percentile.

>>1049123
>if you're putting yourself into tens of thousands in debt, you're going to an objectivelt worse school than one which you can graduate from with no debt at all.

Unless, that college gives you more opportunity, which could in the long term get you money than the other guy.

>most engineers
Don't generalize. Just because you or your friend didn't, does not mean it is true. You are sure closed-minded.

>>1049135
"opportunity" is easy to promise and hard to quantify. let me know if your life is really the success story you're implying it is, because i'll gladly be proven wrong.

>>1049134
Wew, this is projecting. Just because you are a failure, doesn't mean you should drag others down. Just because the bill of sale failed you, does not mean it will be true for other.

File: Clipboard01.png (3KB, 420x494px) Image search: [Google]

3KB, 420x494px

>>1049066
> two motors, one providing vibration, one providing rotation
In which case, you probably want each one on a separate output, so that they can be controlled independently.

If you're driving a motor of 4V or less from 5V logic, pic related is a fairly typical drive circuit.

Rsense lets you monitor the motor current (e.g. via an ADC); that might be useful for the rotation motor (which will experience variable load), but probably isn't useful for the vibration motor (which is free running).

The transistor is needed to boost the current, the diode is needed to prevent the motor from frying the transistor due to flyback transients.

>>1049137
You don't think that the more prestigious colleges have a better network with the alumni and have more handshake deals? You pretend to be some wizened sage, but you are naive.

>>1049137
I implied nothing. That is your doing. I don't have to prove you wrong, because you only speak in absolutes and very condescending manner. You are your own undoing. Plus you fall easily for bait and come back for more. Sadly, you don't have a very discerning or logical mind.

>>1049123
https://en.wikipedia.org/wiki/Confirmation_bias

Maybe you should read an actual book every once in a while.

>>1049146
>Maybe you should an actual book every once in a while.

Right back at ya.

>>1049142
>>1049143
>>1049144
>>1049146
alright guys good talk but as far as i can tell i'm talking to college students thinking in terms of edge case successes rather than what 99% of us can actually expect. as i've said, if any of you are actually in a position higher than middle management please let me know and i'll admit i'm wrong. otherwise i very much doubt that you're anything more than starry eyed students that bought a little too hard into the "you can be whatever you want" meme.

>>1049149
>college students

Once again, having to come up with excuses.

>>1049149
>"you can be whatever you want" meme.

And you fell for the college meme. See I can use your logic too! And it as effective.

>>1049113
The quality of education your university provides is important; however, graduating from a mediocre school doesn't mean you will never be a great engineer.

Confining yourself to the idea that "only an institution can teach XYZ" limits yourself to the same level of mediocrity you accused him of. I can find resources that teach just about everything that you'd learn in my major. Hell, reading the Cambridge series on electrical engineering would probably do that by itself. And I can do that without a university.

The only asset a uni gives you that you can't get easily is access to professors and a certificate proving you can do it. But again, don't get me wrong. A quality education does hell it's just not the end of your career if you don't graduate from an Ivy league uni.

>>1049149
I rather see a person dare to dream, then a cynical asshole who is perfectly fine with mediocrity. One day you will realize you are old and have accomplished nothing meaningful. You are a hostile for a reason, so maybe you know this to be true, but want to live in safety.

>>1049155
I agree, but he is said his education was lacking.

>i've only ever used knowledge from maybe two classes i took in college

Doesn't even sound like he went to a school that taught properly.

File: Can't Handle.gif (3MB, 319x239px) Image search: [Google]

3MB, 319x239px

>>1049149
>This fuck is still fighting with low-tier bait, proving he is an idiot too

>>1049155
>Confining yourself to the idea that "only an institution can teach XYZ" limits

Never said that. But he put himself on that path by going to college.

>>1049161
>I was only pretending to be retarded!

>>1049159
I feel like it might be a bit of an exaggeration. After you graduate, especially after you have been in industry for a while, it's hard to pinpoint "this class taught me that". You just know what you know and he might be taking for granted what he learned in college.

Same with this guy.
>>1049071
Practical hands on experience is different from class theory work. Both are important, but I'll bet hands on experience feels more tangible.

>>1049164
(You)

Even if the arguments are retarded, he is retard to try to argue with them.

>>1049164
>H-He must be retarded, I-I can't be wrong!

>>1049171
>Waah nobody will help me with my Arduino project!

>>1049171
Why don't you add something constructive to the general then?
Your post is closer to a shitpost than our discussion is.

>>1049140

thanks, so Rsense is useful because it allows you to provide more current when the rotation motor is under higher load to compensate?

something else that occurred to me- rotation motor will need to change directions rapidly, how can this be done?

>>1049186
Why would it need to? It should be able to rotate freely over an infinite range of motion, assuming you're thinking of it as having a 360 degree range.

File: Clipboard01.png (4KB, 662x437px) Image search: [Google]

4KB, 662x437px

>>1049186
> so Rsense is useful because it allows you to provide more current when the rotation motor is under higher load to compensate?
Well, it allows you to measure the current. From that you can estimate load and change the voltage accordingly.

> something else that occurred to me- rotation motor will need to change directions rapidly, how can this be done?
H-bridge. Pic related.

This does make it a bit harder to sense the motor current, though.

>>1049198
As for driving an H-bridge:

For "forward" motion, you'd hold PWM_B low and apply PWM on PWM_A. For "reverse" motion, you'd either hold PWM_B high and apply inverted PWM on PWM_A, or pull PWM_A low and apply non-inverted PWM to PWM_B.

You now have two Vbe drops in the circuit, so with a 5V supply the peak motor voltage (100% duty cycle) will be ~3.6V.

Use of emitter-follower (common-collector) mode is simple (no resistors needed) but less than ideal, particularly at higher currents, due to Vce being no less than Vbe (i.e. you drop ~0.65V across each transistor, dissipating 650mW for every amp of current).

If that's an issue, you can just swap the transistors (so that the PNP is on the high side, NPN on the low side) and add resistors between the logic output and base to limit the base current. The resistor value should be just under 4.3V*beta[min]/current, where beta[min] is the stated minimum value of the transistor's current gain (beta or h[fe]).

Or you can replace the transistors with logic-level MOSFETs (again, nMOS on the low side, pMOS on the high side), which don't need resistors.

If you need the motor voltage to be above the logic voltage, things start getting more complicated.

I want to try a 240V to 12V transformer i salvaged from a UPS.
Can I use a light bulb in series as a resistor/fuse in case something goes wrong?

>>1049214
I'd suggest using an actual fuse.

>>1049210

did some googling and came across this guy
https://www.amazon.com/Qunqi-Controller-Module-Stepper-Arduino/dp/B014KMHSW6/ref=sr_1_1?ie=UTF8&qid=1473025325&sr=8-1&keywords=l298n

looks like what you're describing but all packaged up for me. 12v of power total too which should be sufficient for 2 motors as well as the arduino. The description specifies that it's intended for stepper motors though, which is fine for the rotating motor. think I can just attach a normal non-stepper DC motor for the vibration motor?

>>1049222
Are the motors going to be in series? If they're not you don't need 12V, much less 46 which that seems to provide.

>>1049214
Yes you can.

>>1049218
Lightbulbs are more useful, because if the testpiece is doing what it's supposed to, the bulb acts like a wire, and if it goes dead-short, the bulb only lets half an amp through, plus it sinks almost all the energy inside itself.

Their resistance is dependent on temperature, and they generate temperature, thus their behaviour in series with a load is distinctly non-linear WRT current.

>>1049222
btw r u a qt

>>1048999
>OSH Park

[x] Genuflect

>>1049222
> The description specifies that it's intended for stepper motors though, which is fine for the rotating motor.
It'll probably work, but it's significantly more complex (and expensive) than using a normal DC motor.

> think I can just attach a normal non-stepper DC motor for the vibration motor?
You can't drive a normal DC motor from a stepper motor driver, and a stepper motor won't be fast enough for the vibration motor.

>>1049273
I'm not but my qt3.14 gf I'm gonna use it on is

>>1049315
after further research it looks like it can handle either 1 step motor or 2 dc motors, so I'll be doing the latter

my current shopping list is:
- arduino nano
- l298n
- 3‑5V high torque motor for the rotator
- some other 5V motor for the vibrator
- analog to digital converter/sensor for load on the rotator motor. any suggestions for a part on this?

anything else?

>>1049357
An L298 is probably overkill. An L293 or SN754410 would suffice if you want a monolithic H-bridge IC (the L298 can do 2.5A/channel, the others "only" 1A/channel).

But those parts exist to do something you don't need: to drive motors whose supply voltage exceeds the logic voltage (46V for the 298, 36V for the other two). Without that requirement, a H-bridge is just four transistors and (maybe) four resistors (i.e. about 50 cents rather than a $7 module).

> - analog to digital converter/sensor for load on the rotator motor. any suggestions for a part on this?
It's built into the arduino.

How much money do I need to start playing with electronics (hobby for a start)? I already have a soldering station and a multimeter (continuity tester doesn't work I think, it's making noise as soon as I select it). At what point do you get oscilloscope?
If I want to be employable is it possible to learn everything by yourself and still get hired?
How much time does it take to git gud?

>>1049086
Thanks for your help. I think it will work. Of course I will test it before I put it in a printer. I might also include thermal fuse as they are quite cheap.

>>1049153
>>1049154
>>1049156
>>1049161
None of you did what he asked you to do essentially confirming his point quite well.
>I rather see a person dare to dream, then a cynical asshole who is perfectly fine with mediocrity.
That has nothing to do with the college/university you went to. That's entirely up to your talent/intelligence and willpower. If you're stupid you could be going to fucking MIT.
A good school helps, but real opportunities are made/achieved by you. Your school won't just magically give you some AAA-tier position where bitches suck your dick all day.

File: 816273707455427430.jpg (231KB, 600x600px) Image search: [Google]

231KB, 600x600px

I made this power supply kit recently. But it wont work. And im pretty sure i made all the connections right. If anybody could point out what i did wrong it'd be much appreciated.

File: 20160904_105425.jpg (1MB, 2656x1494px) Image search: [Google]

1MB, 2656x1494px

>>1049764
Here's my kit

File: 20160904_105449.jpg (905KB, 2656x1494px) Image search: [Google]

905KB, 2656x1494px

>>1049766
And the bottom

>>1049768
even though the traces are coated i'd trim any of the long leads that overlap traces that they're not supposed to contact.

File: ir_repeater_pcb.png (153KB, 826x889px) Image search: [Google]

153KB, 826x889px

How's this for a babby's first circuit and PCB? (not pictured: some cheapo phone battery with a JST connector soldered on)

I still don't know what I'm doing, but I am sure I'm breaking some rules of sane PCB design.

>>1049679
thanks this is good info, I'll probably go with the l293, saves me some space in the device too

> It's built into the arduino.
gotcha thanks, would that be one of the A interfaces?

for reasons, I want to make a PCB for just a couple of SMD parts (maybe 10-15)

is there a convinient way to make one on my own or is there a place where i can turn? don't really wanna do it on a perfboard

>>1049789
The decoupling cap should be as close as possible to your chips, but it shouldn't matter

>>1049806
Depends on your idea of convenient.
Try googling "toner transfer method" and "photoresist method" for diy PCBs. There are other options too, but those are the most common ones.
For PCB factories... OHSpark was mentioned ITT. There are other similar services too, or you can contact the factory of choice directly.

>>1049807
Yeah, I tried putting it closer, but the whole design fell apart (because I'm trying to keep all this on one layer for no particular reason).
Perhaps I'll try again, but will definitely keep this in mind for future designs.

>>1049816
Should have been OSH Park.

>>1049817
Use a capacitor with 0.2" pin spacing and you can put it right next to the controller. Or you could one track between its legs.
The other anon is right, though - it doesn't really matter in your case.

>>1049821
>Use a capacitor with 0.2" pin spacing
Yep, that worked. Thanks!

>>1049816
>>1049819
thanks

OSH Park looks good actually, whats the catch?

>>1049691
> If I want to be employable is it possible to learn everything by yourself and still get hired?
Employable as what? Technician-level stuff maybe, but you're not going to get an engineering job without either relevant qualifications or experience or both.

> How much time does it take to git gud?
Engineering is mostly theory. You could get a product into production without a workshop or tools. Manufacturing is outsourced, even of prototypes. Testing for any kind of regulatory compliance is outsourced.

>>1049828
Never used it, but the typical catches are long delivery times and limitations for board and hole sizes, buildup, etc.
There was also some other similar service where the board quality was supposed to be lower. I don't remember its name, though.

>>1049828

It's relatively expensive. $5 per square inch gets you 3 copies. Meanwhile, Elecrow offers 10 copies of a ~4in^2 (5x5cm) board for under $10, or a ~16in^2 (10x10cm) for $16. The same boards from OSH would be $20 and $80, respectively, and you'd still only get 3 copies.

Upside is that it's a USA fab house, so time to your door should be (theoretically) a fair bit shorter. I believe OSH also allows arbitrary board profiles; the aforementioned Elecrow pricing does not allow any custom routing, v-grooving, or lines of holes.

That isn't a problem for me, since I have a CNC router, but still.

>>1049836
dirtypcbs.com probably

>>1049768
ehh... you forgot to solder the trimpot.

also bottom third leftmost lead looks pretty bad, are you sure it has contact?
centre lead on the pot also looks pretty bad.

like the other anon said, trim the excess leads off.

>>1049834
>Employable as what? Technician-level stuff maybe, but you're not going to get an engineering job without either relevant qualifications or experience or both.
Figured as much, thanks.

>>1049834
>>Engineering is mostly theory. You could get a product into production without a workshop or tools. Manufacturing is outsourced, even of prototypes. Testing for any kind of regulatory compliance is outsourced.
Yeah, because it's cheaper I assume. But kinda feels bad to not make something with your own hands, well I guess I will tinker with electronics for a start and see.

File: pcb-001.jpg (2MB, 3324x1660px) Image search: [Google]

2MB, 3324x1660px

>>1049789
PCB traces should not be at right angles. That is why you will notice a lot them come in at 45.

>>1049896
Someone needs to debug or at least verify the first prototypes and it's kinda difficult to outsource - unless you outsource the design as well. And it's not really that unusual that the designed puts the first prototype together.

>>1049898
I actually realized that not too long ago. But my five minute research has concluded that 90 degree angles are as valid as 45 degree ones.

Source: http://electronics.stackexchange.com/a/226693

>>1049905
Just good rule of practice. 45 angles will decreased on length, generally speaking, which will lessen on changes of untended noise.

>>1049905
The best reason to avoid using 90 degree corners is so people don't pick up on the fact you used them.
Personally I don't use them because I think they are ugly.

>>1049906
>>1049910
Well, Fritzing doesn't have the "make all angles 45 degrees" button and redrawing them by hand would be painful, so I'll just keep flying my "Right Angle -- Only Angle" flag.

>>1049913
Just for future reference. More experience people will question you why the 90, solely based on "that is the way it is done".

>>1049903
> Someone needs to debug or at least verify the first prototypes
For complete products, verification boils down to "plug it in and check that it works". The customer will know whether it works or not without test equipment.

If you're doing iterative design, where you need to create and analyse prototypes until you get it right then, yeah, you're going to need a workshop with test equipment.

> and it's kinda difficult to outsource - unless you outsource the design as well.
The company that does your compliance tests will be quite happy to do any other tests you pay them to. That may well turn out cheaper than setting up your own workshop, particularly for stuff like complex logic at GHz+ speeds.

> And it's not really that unusual that the designed puts the first prototype together.
Modern high-density circuits are almost impossible to construct by hand. You certainly aren't going to be soldering BGAs with a soldering iron.

>>1049910
Haha. I was asked to review a PCB a while ago and I said nothing about the angles. Another reviewer then complained about that.

>>1049764
>>1049766
>>1049768

Dam that looks sexy. I believe thats a transformer (step down, hopefully) Or could it be a rectifier?

File: 20160905_212155.jpg (1MB, 1886x1060px) Image search: [Google]

1MB, 1886x1060px

>>1049764
>>1049766
>>1049768
>>1049772
>>1049892
>>1050077
I forgot the most important part. The schematic.

>>1050111
Oh you bought a lego set power supply.
>not building your own
Get good bruh

>>1049906
also sometimes enchant can get trapped in the corner of a 90deg bend and fuck you up a bit

File: Photo 6-09-16, 18 11 25.jpg (1MB, 3264x2448px) Image search: [Google]

1MB, 3264x2448px

I'm a little stuck here.
I made myself a shitty "power supply" from parts from china, and have both an AC and DC (half wave rectified) outputs. The box uses a triac/diac circuit.

When I measure the voltage across the AC and DC outputs I measured the proper wall voltage (a bit less for the rectified DC) at any knob position (normal), but when I plugged an incandescent lamp into its AC outlet, it only used 2 Watts and didn't light up. Is the rectifier causing this problem? Can anyone tell me how to fix this?

>>1050168
You read a voltage, but cant get a light to work.

Either the light is broken or you dont have a voltage / current.

>>1050170
There is power going into the bulb (2W). There is a voltage. The lamp also worked fine when tested. The problem (unsurprisingly) is within the power supply.

>>1049922
>If you're doing iterative design
Well, that's how people usually get their designs to the state where they can be finally turned to actual products.
>The company that does your compliance tests will be quite happy to do any other tests you pay them to
This works pretty badly with general prototyping, where the design might have outright errors or where the design changes a lot. It might make sense to pay someone else to perform the tests requiring special equipment (or rent the equipment), but if you need to request someone else to hold a multimeter for you, that gets very old (and expensive) very fast.
There are plenty of products which can be designed with relatively cheap equipment, particularly when you compare the equipment prices to engineer's salaries.
>You certainly aren't going to be soldering BGAs with a soldering iron.
No, but not all products need BGAs and it's not like there aren't reasonably priced options for prototyping purposes. And if your end product will be full of BGAs crammed on a tiny PCB, it isn't that unusual to do the initial prototyping with bigger devices.

>>1050111
So, what's wrong with it? What happens when you plug it in? Have you measured any voltages, like what's the voltage at D1's cathode?

>>1050171
2W seems way too weak for a light bulb. Normal house bulbs are rated at between 50-100 W.

What should I be looking for in a really strong 5v DC motor? Ideally low rpm

>>1050290
>What should I be looking for in a really strong 5v DC motor? Ideally low rpm
Get one with a gear reduction. There's thousands of them on aliexpress, many under $20.
Try to get one with a planetary gear reduction, where the output shaft is coaxial with the input shaft.
The gear reductions that have off-center output shafts use smaller gears and don't last as long.

My [unrelated] question:
I make a lot of arduino-level stuff using perfboard, and I usually use screw terminals to connect any wires.
I can use a piece of tape on each wire to say what it is, but what is a good way to label what each screw terminal is?
I can't think of any way that I like,,,,, that is fairly quick and easy, and effective.

>>1050179
> Well, that's how people usually get their designs to the state where they can be finally turned to actual products.
Maybe for analogue; most digital stuff ought to work first time.

>>1050299
Sharpie no?

File: 20160906_173648.jpg (1MB, 2322x4128px) Image search: [Google]

1MB, 2322x4128px

>>1048060
>>1048080
>>1048085
>>1048086

Thanks guys!

File: image.jpg (54KB, 413x395px) Image search: [Google]

54KB, 413x395px

>>1050317
>most digital stuff ought work the first time
maybe the digital guys i've worked with are terrible, but i'm keking hard right now

Say I have a 300 rpm motor. Is there any benefit between using PWM to lower the speed to 200 rpm, versus buying a natively 200 rpm motor? Would the native 200 one have more torque?

How much did microcontrollers cost in the 90s and early to mid 2000s? I recently started working at a company where one of their products was made with an 8-bit MCU and the rest use 32-bit MCUs. I asked one of my coworkers and he said that it was intended to be a low cost model, and at the time when it was made (mid 2000s) it was significantly cheaper to use an 8-bit MCU, so even though the original software was written for a 32-bit, they rewrote it for the 8-bit to lower the cost. He also said going forward they're using all 32-bit ARM now, even in low cost models.

Was the price difference really that big back then? Or is is just the the small difference becomes a big deal when dealing with volume? I looked up both the one they regularly use and the 8-bit one and currently the difference in price is slightly under a dollar.

>>1050548
That's not right. Think about what you're doing again, and think about what tool you're trying to use.

>>1050590

enlighten me?

>>1050548
For any given motor, gearing to a lower speed increases torque.

PWM effectively reduces the supply voltage. It allows you to change the speed dynamically. It also allows you to use a supply voltage above that for which the motor was designed (e.g. driving a 3V motor from a 5V supply).

Note that voltage ratings aren't particularly meaningful for a motor. They may indicate the maximum voltage which can be applied to a stalled motor without exceeding its maximum current, or the maximum which can be applied to a "reasonably"-loaded motor without exceeding its maximum current.

The relevant parameters for a motor are the winding resistance, maximum current, torque constant Kt (in Newton-metres per amp), and the back-EMF constant either Ke (in volt-seconds i.e. volts per radian per second) or Kv (the reciprocal of Ke, usually given in RPM per volt). Maximum voltage is just maximum current times winding resistance.

Manufacturers and vendors targeting "serious" use (e.g. RC vehicles) will normally list all of those parameters. Otherwise, you at least need the maximum current, winding resistance is easily measured, Ke is relatively easy to measure, Kt can be assumed to be equal to Ke (Kt/Ke is the ratio of mechanical power to electrical power, which has to be less than one).

idk

File: 40kV Probe Meter 2.jpg (21KB, 511x640px) Image search: [Google]

21KB, 511x640px

Can anyone think of an reasonably accurate way to calibrate one of these?

The most accurate thing I have access to is a fluke 21 series meter.

File: multimeter_21.jpg (13KB, 278x375px) Image search: [Google]

13KB, 278x375px

>>1050731
>Pic related

Any EEs here try doing contract work?
I've had two contracts in the past, one making an enclosure for a museum and the other was for a fashion designer.

I enjoy it more than a 9-5. Any suggestions for getting more contract work?

>>1050399
>Sharpie no?
I've done that and there's not a lot of room to write, even when there's space on the perfboard. And permanent markers are not colorfast (they fade eventually) and paint markers don't draw that fine....

I've tried a couple different ways already and didn't like any of them.
There's nothing that really sticks out as a good idea, so far... :[

>>1050734

build something that outputs 1kv measured from the fluke and then use that to tune it to 1kv

new thread bitches

>>1050981
>>1050981
>>1050981
>>1050981

Do any of you have experience with voltage inverters? I'm using a TC7660S with 12V input and it's only outputting -7V in my design. If I isolate it it gets all the way down to -9V. I've got a 10uF polarized cap on it just like it asks for. Why is it not outputting the right voltage?

>>1050299
Just carve one or two letter(s) under every connection into the perfboard maybe?

You can use a separate table for the letters if you want. For stuff like G/-/+/D+/D- you won't really need to really.

>>1048693
If you just want a switch, use a MOSFET, not a transistor.
>>1048666
Capacitor in parallel with battery will supply fuck all energy, there's not enough change in voltage, and you'd also need a massive cap to do anything.
>>1048751
Base to emitter of a transistor is just a diode, without a current limiting resistor you'll get huge current.

What is the circuit supposed to actually do?

>>1048916
That's fine then. Swapping to an ATTiny85V would let you run for longer, the regular 85 is only rated down to 2.7V where there's still a decent amount of energy left in the cell.

>>1049764
Get your multimeter and start probing. First, check you're getting ~12V AC out of H1, then check you get ~16V DC across C3. If you've got that, your transformer and rectifier is working. Then check the regulator, you should see 1.2V difference between pin 2 and 1 on the LM317, and you should be able to change the voltage across C1/C2 by adjusting RP1.

>>1050168
Post schematic, and inside pics.