>>927391 Huge: < 1kV or so: basic linear regulator using a high voltage bjt or fet. Huge: > 1kV or so: switcher with a step-up transformer and an optional voltage multiplier, regulation from primary side.
I was just wondering what projects other anons are doing with thermoelectric modules. I have a few and have been playing around with them, getting used to what they can and can't do. They are pretty fun to play around with.
Normally, the mere mention of something will spur someone to talk about it if they are doing a project or like the subject matter.
I'm designing a (bass) guitar amplifier. This is what I've come up with as far as the power amp for driving the speaker. The input stage is an op amp.
This is the first time I've actually tried designing and simulating my own circuit rather than copying some shit someone else posed online, although it is heavily derived from other amps i've studied / built before. Is it retarded?
>>927577 Well, I've used them to cool cold traps and small environmental test chambers (that was a fancy name for thing which was basically just a tiny fridge). I've also seen a project which was a non-obscene version of anon's joke. Tiny peltier cooler was attached to skin and it was used to notify the user about ... something via temperature (hot/cool/neutral).
I have an 8-bit A/D converter chip with four single-ended inputs connected to an AVR microcontroller. I'd like to attach some sensors to it and build a small weather station that I could place on a windowsill. It would measure temperature, relative humidity, illuminance and maybe wind speed.
What kind of temperature sensor would be good for this use? Would a simple NTC thermistor be sufficient for a temp. range of -30°C ... +40°C?
I'm planning to use the thermistor amplifier circuit shown on this Texas Instruments application note. It transforms the variable resistance into an analog voltage that can be fed into A/D converter. I just wonder if it would be possible to use cheap LM324 quad op-amp instead of expensive TLV2472. Pic related.
>>927666 Look up the book "Designing power amplifiers" by Bob Cordell (which may or may not be available in slightly illegal bodies of water where pirates reside). It is great for amplifier design and learning the basics, and how to go further and improve THD etc
So /ohm/, I just bought myself some 555s and I'm having no trouble using them when following instructions but would like to learn more about them Do you have any books/videos on 555s by chance? I'd very much appreciate that
we cant decide for you. some people are calculator-heads, they wanna see formulas. some people appreciate schematics, others think they're gibberish. some like cute flashing lights, others like timer applications, others want big-power stuff like Tesla coils.
>>928125 Thanks I built the most basic stuff I could find (a monostable LED timer and an astable tone generator) and I'd like to move on to slightly more advanced stuff, I thought about putting two in series to have a tone for some time whenever you press a button but other than that I have no ideas
>>927916 >LM335 Thanks for your tip. This part has built-in linearization but still can be fine-tuned with a trimmer pot. I could make a simple op-amp circuit for scaling the output signal of LM335 to fit the full input range of the ADC. For example -30..+40°C -> 0..3 V or so.
>8b external ADC instead of an AVR with internal 10b ADC Unfortunately the part that I have doesn't have any analog inputs.
ok, well an Atari Punk console is quite a popular project. every noob makes one to show their geek friends. one step above that is the Atari Punk Organ, where you make like a piano thingy. also, popular is the 3-monostable traffic light. of course, there are real-world uses for 555s, as part of bigger projects, as timers, oscillators, PWM drivers, edge detectors, etc
I am interested in trying to wire up LEDs for lighting in my barn. I do not know much about electronics but I think I have some of the equipment picked out necessary to test on a small scale what I want to do.
I am not sure If the 20 watt led will provide enough lighting but they are cheap enough that It wont bother me if I can't use them for the full project. I plan on going with a 50 watt driver but didn't have one picked out to go with this post. I figured if I went with 50 and found out that 20 watts was too little light that I could still have room to jump up to 30 watt higher lumens and not need a different driver as well. From what I understand the larger LED need heat sinks for cooling so planned on buying a few of them also but the link provided is the final pick. My general thinking was I could plug the driver into an outlet that has a on off switch and from there run wiring across exposed beams and place LED wherever I decided. I was thinking of making my own fixtures out of wood somehow or just screwing the heatsink right into a beam. Are my chosen supplies what is needed?
>>928191 You're missing an important piece of the project. You can't just stick 12v to an LED unless it's got circuits to limit the current/voltage to levels it can handle without being destroyed, and it doesn't look like those LEDs have anything like that built in. For a high power LED you need a switching current regulator; a series resistor won't work out.
>>928191 wouldn't it be easier to get these and use 120vac led bulbs. then all you need to do is run a bunch of ac outlets (or extenson cords) throughout the barn. also make it easier to change the wattage if you need more light.
So I was cleaning out my grandma's garage when u came across this. She had no idea what it was so I kept it. It has a microphone with a 10ft cable, Ethernet port (no it's not rj45), and power. When I opened it I found a speaker.
What is this device!? What does it do and how do I use it?
>>928318 Sorry I meant to say its not RJ11 port. Got my ports mixed up. I was thinking it was some sort of phone thing too but why does it have ethernet and why is the mic so long. It's really old and it has no lables so I can't tell anything else about it except for the internals
I'm just starting to learn about circuity and electronics of my own vocation, but I'm having a little trouble conceptualizing the value of an ohm.
Would the value of one ohm indicate that there is absolutely no resistance of a conductor? For example, if I were to have a 1 volt battery in a closed circuit, with 1 amp of current in that circuit, can I assume that there is no resistance, and that the ohm value of the circuit is 1?
My logic is that a 12V battery in a closed circuit with 4 ohms resistance would produce 3 amps in accordance with Ohm's law, thus, 1V with 1 ohm would produce 1 amp (no resistance).
>>928340 Maybe I'm misunderstanding voltage then. I understand in reality that real conductors have a little resistance, but for the purposes of my example, I'm just pretending my circuit is a superconductor.
If I have 0 ohms of resistance, will 1 volt produce 1 amp...? I can't divide by zero... but as resistance approaches zero, it takes very little voltage to produce one amp... Or one volt will produce an large amount of amps...
So an ohm is just defined as the impedance required to slow a current (amperage) to one coulomb per second from a 1 volt battery?
>>928336 >Would the value of one ohm indicate that there is absolutely no resistance of a conductor?
No, it would indicate that the conductor had a resistance of one ohm.
You seem to be looking at your problems backwards. The resistance of a simple circuit (containing only restive elements) never changes. A battery and almost every other power supply is a voltage source which means it gives a voltage that never changes (ideally). The current through the circuit is what changes to satisfy Ohm's law, not resistance or Voltage.
Amps are easy; just a charge (coulombs) traversing a conductor per second (current).
Voltage is the amount of potential energy needed to overcome resistance to current, and ohms are a value of that resistance? So volts and ohms are purely arbitrary values that correlate to amps via Ohm's Law? One cannot exist without the other?
If this is right, I think it's starting to make sense. I guess I was trying to define them as something tangible.
>>928432 DC, dual supply. Chink is trying to say that you can produce the required supplies by rectifying and filtering 18V AC. You can either use a transformer with a single winding and two half-wave rectifiers (requires bigger capacitors), a 2 x 18V center-tapped transformer and a diode bridge or a transformer with two separate 18V windings and two bridges.
>>928438 No, you need dual supplies, +/- 25V. In other words, 2 x 25V so that the total supply voltage is actually 50V and the ground is the mid-point (so that you have +25V, 0V and -25V). It also needs something like 2A of current at full power (to 8R load), or around 3A during peaks (or when driven to clipping).
>>928452 The amp needs to produce both positive and negative half cycles and each half cycle gets at most half of the total supply voltage. Let's say you have an 8R load and you're pumping 30W RMS sine to it. You can calculate from P=I^2R that you'd need 1.9A RMS current. Since it's sine, it means that the peak current is sqrt(2) times higher, or 2.7A. To get that much current through an 8R load you need V=IR=22V peak voltage. And that's your positive half cycle. You need -2.7A and -22V for the negative half cycle, so that the peak-to-peak output voltage is 44V.
The advantage of using a dual supply is that it's simple. If you want to use a single supply, you either need to feed the speaker via a large capacitor (since the amplifier output will be sitting at supply_voltage/2) or use two amplifiers.
Hi /ohm/. I completely forgot about /diy/ and asked a question on /g/ out of ignorance and it seems that this general is fit for my question. Could anyone verify that what this anon said >>>/g/52364029 is correct?
>>928474 It's for LEDs, so I doubt it. It's much more likely that the ground is protective earth and that you'd need two of those bricks. Try to find a kit built around some car stereo amplifier chip, for example TDA1516. They're designed for single supply operation.
as the other dude implied, your switch is leaking current, which it shouldnt. so, it's either a cheapo dimmer that does not incorporate a switch, or it's a timer switch, or a smart switch, that always passes a tiny bit of current to keep itself powered-up.
if it's not the switch, then there's some non-obvious wiring errors.
if you got skillz, and fine-tip soldereing iron, that's not too bad. the power and ground are still in place, so you just have to re-attach the 2 data lines, which you can do by running some fine wire, like 30-gauge wrapping wire, from the connector to the two resistors.
alternately, you can solder the cut-off end of a USB cable to the 4 points, and have a permanent cable hanging out of the unit. (assuming this is not the kind of device you carry in your pocket)
however, if you dont have the skillz, then you will only make it worse. since you have to ask, i assume you're in this category.
>>928529 I can't find any kits that use regular ol' DC, so what if I got two of the 5A 12VDC power supplies and used them to drive the amp? Seems clunky and dumb, but would it work? And how would I go about wiring it up? Just ground to ground, and 12VDC from one to the + terminal on the amp, and 12VDC from the other to the - terminal?
Is it counterproductive to put a heatshrinked 3034 mosfet on a heatsink? I have one pre-wrapped and I would rather have the heatsink if I had to choose between the two. The mosfet would only have 4v going through it at a time if not less.
>>928886 > It's rated at 3W 3-5V, so should I just hook it up to 5v
5V is a maximum, you dont start at the maximum ever, coz that's asking for trouble. so, you'd start low and increase towards the max while checking for temperature. if it's too hot to touch, you've gone too far.
a power LED like that will not have a resistor, and would normally use a constant current source at the appropriate rating for it, so that it doesnt go into thermal runaway.
>>928939 Alternatively I could gut it and replace them with a few 3.6v 20ma leds and just use 1/4w 100ohm resistors. That's something I've done before so I hope it's the right calculation. I don't think I'll need 2w of brightness for where I'm putting them.
>>928832 Yeah, two separate bricks is one way of doing it, but you need to have bricks which are floating. Those LED PSUs probably are, but I didn't see the ad really mention it.
The solution the Chink was referring to (18V AC -> rectifier) is the typical way to produce the dual supply for non-tiny amplifiers, but it requires a transformer and tinkering with mains voltage. Pic related. The first amp chip you mentioned can also pretty easily arranged for single supply operation (the datasheet has a schematic), but then you need to wire it by yourself.
>>928973 What would have to change to the first amp to make it single supply compatible? I'm down with some reconfiguring as long as it's not balls-hard, but I don't know much about circuit-making or anything like that.
>>928928 I don't even know what is a 3034 mosfet, but Google suggests it's actually IRLB3034, which comes in a TO220 package. Yeah, it isn't the greatest idea to have heat shrink tubing between the fet and the heatsink. How bad idea it is depends on power dissipation. Cutting the heat shrink sleeve away shouldn't be exactly challenging.
>>928976 See the amplifier chip's datasheet. www.ti.com/lit/ds/symlink/lm1875.pdf
The second page shows the dual supply configuration, which should be pretty close what the Chinaman used. Page 3 shows the single supply version. The principal changes are that the pin 1 is now fed from a voltage divider (R1/R2/C2), output has a large capacitor (C6) and the filter capacitors from the negative side are removed. The feedback thing (R6/R5/C3) is also using different component values, but there's a pretty good chance that the original values would also work.
>>928982 Alright man, thanks for the help! I'm trying to make a mono amp using this spare 60W speaker I've got, and this seemed to be the cheapest solution. I'll probably end up getting a car monoblock and a 12VDC adaptor though lol
I'm not sure if this is the correct thread for this so apologies in advance if it's not.
I recently managed to get an electric scooter for dirt cheap. I've seen some videos of people with the same model who have tricked them out to got faster by changing the batteries out/ adapting the motor somehow. Is this something a complete novice could manage to do with out destroying it or should I just find someone with more know how than me to do it?
>>929054 Based on my limited knowledge of such things, if you do enough research and ask around some more, this sounds like a pretty decent first-ish project. Maybe some fab work involved, but with such a simple concept (scooters in general) to begin with, I don't see how it could end up being that complex.
With that in mind, going 60 mph on one of those things sounds fucking terrifying. But hell, go for it, bro. You only YOLO once.
i want to power a circuit via batteries. the microcontroller i use needs at least 2.3V @ 8 Mhz (pic related). since there are some more components Vcc>3V will do (tested it)
now i don't want to use 3 batteries, because thats just too clunky. two batteries work fine, but only if they are nearly fully charged and i don't want to switch them when they aren't even fully discharged. i want to use the full capacity.
so what i would love to use is a boost converter that works from 1,8V-3V and provides around 3V-5V. also i would like to avoid an IC and use a discrete ciruit instead for interest sake.
i know the basic theory about boost converters but haven't tested one myself yet. will a typical circuit do the trick? what do i have to take into account, considering my requirements?
When you connect a microcontroller pin to something via a resistor in series, then how come the thing you've connected it to still "see" the voltage?
Like with an Arduino - it's 5v out of each pin so you can use it to control something with a 5v logic level. But you usually have to put a current regulating resistor in series (like with an LED), so how does the voltage drop of the resistor not affect the voltage seen?
Is it something to do with the impedance of the IO ports?
>>929133 > will a typical circuit do the trick? Uh, what's a "typical" circuit?
In addition to the basic inductor-switch-diode-capacitor, you need an oscillator, and some kind of feedback to maintain a constant output voltage.
> what do i have to take into account, considering my requirements? Given that you're presumably talking about very low power, there aren't really many constraints. Most of the design constraints relate to efficiency.
The basic design requirements is: don't exceed the maximum voltage, current or power dissipation of any component.
The capacitor's energy capacity needs to be such that each pulse should constitute a small fraction of the capacitor's total energy capacity, so that you don't get noticeable ripple.
You need to keep the inductor current below its saturation current.
You can operate either in continuous mode (inductor current never reaches zero) or discontinuous mode (let inductor current reach zero before starting each cycle). The latter means that you only get switching losses when turning off, not when turning on.
The lower the switching time, the lower the switching losses. When the switch is fully conducting (saturation), there's negligible voltage across it thus negligible power dissipation. When it's off, there's no current and thus no power dissipation. In between (when the switch is in the linear region), there's power dissipation. Transistors (either BJTs or FETs) turn off faster if you actively pull the base/gate to ground than if you just disconnect it.
The higher the switching frequency, the higher the switching losses, but the smaller the required inductor (the inductor needs to be able to store one cycle's worth of energy for discontinuous mode, more for continuous mode). At high frequencies, you may need multiple output capacitors (i.e. something with low parasitic inductance to handle the high frequency components).
The energy dissipation of the inductor and capacitor will be determined by their parasitic resistance. Particularly for the capacitor, you need to average the square rather than squaring the average (nearly all of the power dissipation will occur in the charging spike at turn-off).
>>929775 >>929782 >>929783 Mk, just a quick update. Figured I'd take y'all's advice and rigged up the shittiest connector on the face of the Earth to see if charging using my dedicated charger would be faster, and it definitely is.
But I thought you guys may get a kick out of the shit job I did.
Pic shows the new charger, the shitty connector I made, the perilous way its connected to the battery, and the rating of the charger that the battery came with the 6+ years ago when I bought the rc truck.
>>929833 Well, a quick calculation suggests that ½mm thick heat shrink tube would increase the thermal resistance about 20K/W, which is about the same as for the tiniest clip-on heatsinks (20x20x20mm). That is, the heat shrink tube alone would turn your infinitely large heat sink to an equivalent of the tiniest heat sinks. To put it another way, the MOSFET with heat shrink tube would be capable of handling around 5...7W on an infinite heat sink and 2...3W with those tiny heat sinks. And that 2...3W would correspond to 35...42A current at the nominal Rds(on) @ Vgs=10V.
Does anyone know cheap /diy/ friendly custom MCPCB supplier? Or is MCPCBs even worth it? I found some random article that says that FR4 with plated vias would perform just as well, but that seems bit suspect.
I'm working first time with high-power LEDs (maybe upto 20W total) so that's why I'm bit clueless...
Got a receiver like this. Same model but not mine.
Overall it sounds pretty good I think, but this is my first fancy audio system.
Sometimes, the left channel will cut out entirely unless I set it to mono, and when I have headphones on I can tell the right channel is significantly stronger.
I was thinking of replacing all the capacitors, but I've never done something like that before. I found this guy http://www.vintageaudiobliss.com/2012/02/akai-aa-1010-stereo-receiver.html has done it, but doesn't list the capacitors you need. Is it as simple as reading the voltage and capacitance ratings and then buying capacitors of the same values, then soldering them in place?
I'm in a microfabrication lab this semester and we get to make any IC we want for a final project. I'm not sure what application I should design mine for. At first I was thinking about making a fizz buzz IC for the novelty, but I might want something more interesting but still relatively simple.
>>930223 >Is it as simple as reading the voltage and capacitance ratings and then buying capacitors of the same values, then soldering them in place?
Yes. That said you most likely don't need to.
Give the switches a clean with some of this. Not too much. You don't want it everywhere. Just squirt a little in the switch and push / turn it a bit. I wouldn't put it in the mains switch. Switched headphone sockets can need some on occasions but it sounds like your fault is before that.
Other than that check for dry joints which you want to do anyhow in something of this age.
i've heard multisim can do microcontroller simulation, but since everything in multisim is complicated and unstable i can only assume that is too. can i easily use that feature to have code toggle outputs based on inputs during the schematic simulation?
i'm not doing anything technical, i'm just trying to use multisim to understand switching power supplies better and that requires implementing a PID loop to control output voltage. which i'd like to do in a few lines of C instead of simulating some bullshit voltage controlled pwm circuit if possible.
>>930285 741 555 7805 7474 sampling gate with drivers time to digital converter or a part of it 1b processor flash or folding ADC fast test pattern generator electrometer amplifier wide-range hall sensor "high voltage" charge pump with integrated capacitors temperature sensor
How do i make LED lamp (only need one) for 12V AC current.
stereo lamp (filament bulb) burnt out and i want to replace it with LED, but i looked everywhere and every site gives this large robust circuits. I need something small and compact so it can fit in receiver
>>930360 Oh you. No you don't. Just keep the rectifier (diode bridge) part outside the tight spaces. That is, put the rectifier somewhere near the transformer where there's space for it and wire the 12V DC output to your LEDs. Also, since you know your bulbs are 12V, why not use 12V bulbs?
>>930368 because i live in gods forsaken land (croatia) and i just went to our larger store (chipoteka) and i asked for 12 volt light bulbs and they only have 12V DC light bulbs. I guess i will have to order overseas again
>>930377 cont. I have just returned from my trip to store. Bought AC to DC converter (220V 50Hz -> 12V 600mA) Will that converter run or burn my gramophone (12V 100mA). also how can i know which is negative and which is positive. One has straight line like this ___________ The other has straight line ____ and broken line like - - - - - -
>>930447 Use the continuity measurement on a multimeter and see where the wires go inside the gramophone, see of you can find any input filtering capacitors as that will reveal which wire is positive and which is negative.
Is there a model number for this gramophone? One of us might be able to find some schematics.
complete beginner here: how would i go about making a simple contraption that would have single knob to shift the pitch and would just splice between your speakers/headphones and the audio output (pc/phone)
>>930360 > How do i make LED lamp (only need one) for 12V AC current. Put a diode and resistor in series with the LED. You don't need a bridge rectifier (half-wave rectification just means that you get 50/60 Hz flicker rather than 100/120 Hz). The diode is there because LEDs are typically limited to ~5 V reverse voltage.
your circuit is wrong. your pot is essentially in series with two diodes to ground, which makes it super touchy and likely to burn up. you want to put the load on the emitter, and wire it up like this instead.
>>930631 >your circuit is wrong. your pot is essentially in series with two diodes to ground, which makes it super touchy and likely to burn up. I don't understand. I realise 100ohms is probably too low but when the pot is maxxed it's about 1mA. Should i pad with a 330ohm maybe.
Also why the load on the emitter? So the lamp can use the base current too? Or to add more resistance from pot to ground?
you should just try it instead of asking questions. it's faster and you learn better.
having the load on the collector essentially makes it an on-off circuit. as soon as you apply enough current to turn on the transistors, the lamp goes on full power. by putting it on the emitter, you can raise the lamp voltage incrementally from 0 to about +12 so you have a range of brightness.
>>930637 >having the load on the collector essentially makes it an on-off circuit. as soon as you apply enough current to turn on the transistors, the lamp goes on full power. by putting it on the emitter, you can raise the lamp voltage incrementally from 0 to about +12 so you have a range of brightness.
true, but so what? the difference is in the transfer function between input current and output current. with an emitter load, it's a nice straight line, with a collector load, it's more like Al Gore's hockey stick graph.
>>930692 maybe if you are dumping the load voltage onto the base instead of working out the current range to drive the load properly. load on the emitter is just another voltage drop to factor in calculating your base current you don't need to worry about a the voltage drop of the load on a collector, figure your base current wrt the be drop which is pretty much always .7 or 1.4 or something reliable and boom done. no sweat.
sometimes you can't drive the base voltage up to overcome the load drop either, say you want to switch something with a uC that supplys 5v only? 12v load? tough shit.
load on the collector is better but in theory there is no difference between the two methods.
>>930627 > What are the key transistor parameters I should understand? h[FE] (i.e. current gain).
The 2SC2625 has a minimum current gain of 10. That's not very high, but it's not atypical for a power transistor.
A 12V 21W lamp will draw close to 2A of current. With a gain of 10, you're talking about 200 mA of base current. With an 11-point-something volt drop across the base resistor, that's around 2W dissipated in the resistor.
In this particular application, you'd be better off using an emitter-follower, i.e. putting the load between the emitter and ground. That way, the base voltage will rise as load current increases, meaning the amount of voltage you need to drop decreases as current increases, which will limit the power dissipation in the base resistor. It will also make the load current less dependent upon the transistor's gain, which varies with temperature.
Another option is to use a MOSFET instead of a bipolar transistor.
in the city, i see old or broken electronics on the street pretty much every garbage day as i walk around; printers, stereos, TVs, VCRs, washing machines, electric gardening tools, etc. a thrift shop will have stuff which is usually ok or half-broken, or missing pieces, very cheap. there's also a repair shop nearby: the stuff that would cost too much to repair, they put outside in the lane.
>>930765 >In this particular application, you'd be better off using an emitter-follower, i.e. putting the load between the emitter and ground. That way, the base voltage will rise as load current increases, meaning the amount of voltage you need to drop decreases as current increases, which will limit the power dissipation in the base resistor. It will also make the load current less dependent upon the transistor's gain, which varies with temperature.
That's a lot to get my head around. Any good online guides to help? Transistors are mad.
>>930765 I changed the circuit so the bulb is on the emitter. It's definitely better - the resistors didn't get hot anymore.
The bulb was a little dim so I played with the resistors. I found that when I took the base resistor off the first transistor, the bulb stayed on.
So that stage didn't seem necessary so I took it out. Now it looks like this and seems to work perfectly. The pot (of which I'm only using the 2 legs) seems fine. I measured the base current and it's only about 63mA tops.
Previously I measured the bulb to draw about 1.6A now in this circuit the whole thing draws about 1.5A.
Is this setup advisable? The power transistor gets damn hot too. still. Is that normal for this application?
>>930986 Does the load on the emitter make a votage divider with the base resistor? Or would it be one anyway? Does a transistor base have an appreciable impedance or is it essentially a wire to ground?
>>931004 The inductive load would be where it says Load, it looks like 1 diode would do the job here too but i believe if you use one at each Source - Drain it would distribute the power equally. Maybe the diodes are just added next to each FET so you know which one you should get more easily since most power misfets have a diode between S- D integrated already.
>>931000 I just experimented with this. Seems like a resistor on the base and on the emitter does act like a voltage divider. When I short the emitter to ground though, I still get some voltage at the base, so the path from base to ground (via emitter) must also have some inherent resistance, right?
removing the first transistor is not advisable. a pot isnt meant to carry any appreciable current, and 60mA is way too much. if you put back the first transistor, you can reduce that current by about 100 times.
> The power transistor gets damn hot too. still. Is that normal for this application?
it's normal in this kind of setup. at half brightness, the transistor is gonna dissipate about 10 watts. if that's not acceptable, you have to go the PWM route, like the circuit shown at >>930971
>>931033 >I'll see if I can get the Darlington pair setup working
well, using the darlington pair means you lose about 1.4V, so if you start with 12V, you end up with 10.6V on the lamp. but, if you start with 13.4V, then you end up with 12V, so you dont lose any brightness at all.
the 'how do i bias an NPN transistor' explanation takes about 5 minutes in person, and a week thru text. so just know this: emitter voltage is base voltage minus 1 diode drop. since you have 2 transistors in series, it's 2 diode drops. the collector drop is of no concern to us (in this configuration) coz the emitter is king (in this configuration).
>>931063 >the 'how do i bias an NPN transistor' explanation takes about 5 minutes in person, and a week thru text Ain't that the truth. If i just had someone to ask in person.
>. so just know this: emitter voltage is base voltage minus 1 diode drop. since you have 2 transistors in series, it's 2 diode drops. the collector drop is of no concern to us (in this configuration) coz the emitter is king (in this configuration).
Ok I see that now. I thought the transistor was just modulating between short and open crcuit between the collector and emitter - thus letting you go between 0 and the supply voltage,
the collector just deliver what the emitter asks for, or dies trying. so, putting 8V at the base means 6.6V at the emitter. the collector drops 5.4V in this case, or in general, it drops whatever is necessary to provide what the emitter has ordered.
>>931073 Collector current (Ic) is proportional to base current (Ib). Emitter current is the sum of the two.
Base current is proportional to .. well, unless it's limited in some way, it isn't proportional to anything. It's roughly exponential with base-emitter voltage. In practice, this means that if Vbe is below 0.65V, Ib will be negligible; once Vbe goes over 0.65V, Ib will be whatever it can get.
Putting the load between the emitter and ground (emitter-follower) means that the base voltage will be 0.65V above the emitter voltage (or conversely, the emitter voltage will be 0.65V below the base voltage).
If you try to increase Vbe above that figure, Ib increases, so Ic increases, so Ie increases, so Ve increases, so Vbe reduces. If you try to reduce Vbe, Ib decreases, Ic decreases, Ie decreases, Ve decreases and thus Vbe increases. IOW, you have a negative feedback loop which tries to maintain Vbe at 0.65V.
>>927666 The other anons don't know what they're talking about. Topology-wise, that looks fine. Keeping the power amp in the feedback loop like you're doing will greatly help to reduce distortion. If it it simulates OK, I'd go ahead and try it out. Be mindful of heatsinking.
>>930704 Are you an idiot? Using the transistor as an emitter follower is probably the best way to do *variable lamp brightness* without resorting to pwm. You have half of the answer yourself - the base emitter drop is .7v. Therefore, the voltage across the lamp is ALWAYS the base voltage minus the base-emitter diode voltage. The transistor's gain will attempt to make this as true as it possibly can.
As for how linearly the voltage across the lamp tracks the brightness of the lamp: I don't know.
I frequently see circuits that use gate drivers following logic outputs to control power fets. Are these really necessary? As I understand it modern mosfets, even for high current applications, still only need logic-level voltages at extremely low currents to switch fully on.
>>931324 Not all MOSFETs are logic level devices, as logic level drive is a compromise. Also, while you need only minimal amount of current to keep a MOSFET turned on (or off), you'll need lots of current to make them switch fast.
>>931334 It works, it's unlikely to be the limiting factor in that design and it's cheap and easy to get. The primary problem with it is the limited supply voltage range, but the same applies to the most opamps. And if he's unhappy with it, he can always just drop another opamp in its place.
If he wants to improve the circuit, he has plenty of more interesting options to check. Like that level shifter / drive stage and frequency compensation. Short circuit protection would be nice to have in an experimental amp, too.
>>931333 There are three leads coming out of the device. Just open the lid, it shouldn't be rocket science to figure out which one is positive and which is negative. Heck, you could even try measuring continuity against the case with a multimeter
>>930070 This is amazing. I'd love to make something like this. I've recently been reading some books on basic electronics as was planning on making some effects pedals, but was only thinking today that maybe making modular stuff might suit me a lot better as I don't play guitar but am into synths and general sound design. Do you have any advice on where a good place to start would be for learning about electronics for sound generating and processing, preferably euro rack stuff? I have a few pedal kits to build and basic soldering skills.
digikey's product table for all of the transistor categories has a column titled "max power". is that the device's maximum rated dissipation, or the maximum Vds*Id? it doesn't say anywhere, including a datasheet i looked at.
>>927806 Are peltier coolers a practical method of cooling test tubes by 10-20C or am I better off sticking with an ice bath? Obviously the coolers would allow much more fine tuned control, which would be useful.
>>931324 1. Not all MOSFETs are "logic level". 2. A "logic level" FET typically needs 5V Vgs; if you're using 3.3V logic, it's unlikely to be enough. 3. Even 5V isn't enough to fully saturate most "logic level" FETs; you need more if you want Rds to be close to its minimum value. 4. Large FETs have significant capacitance. If you want them to switch fast (in order to minimise the time spent in the linear region where both Rds and Ids and thus I^2*R are significant) you need to be able to charge and discharge that capacitance fast. A decent driver will push/pull *several amps* to/from the gate in order to achieve fast switching. 5. If you're driving a push-pull pair of FETs, the high-side FET will have its source above ground, maybe even well above the logic level. The gate needs to be 5V+ above that, so you definitely need a level shifter (high-side drivers often also include a charge pump to generate a gate voltage above the supply voltage).
>>931474 Yes. It's a bit cumbersome but in the past I have used audacity to generate waveforms and export them as mp3's, then once transferred to a phone just connect the headphone jack to the circuit (careful not to short it).
I used this method to test a small audio amplifier I was building, if you are feeling confident you can use the audio output of your PC directly using a program called SigJenny.
I want to make an amplifier using a TDA2030. I have this toroidal transformer. It's rated for 60VA and 50Hz. From what I have read, it can run on 60Hz just fine.
It has a wire for the neutral, one for 110V and one for 220V, then two for 12V. I also have some rectifier diode bridges and a few large capacitors.
How do I make it so that the amplifier can run on either 220 or 120v? Is there such thing as a capacitor calculator to determine the optimum capacitance? I assume 35V capacitors will work just fine, right?
>>931478 >SigJenny Sorry I meant to say SweepGen, SigJenny hasn't been updated in years and give an error on my PC.
When using a PC soundcard as a signal generation be sure to exercise caution, one wrong move could seriously fuck up the card. That's why I suggest using an old cell phone first just to make sure the circuit works correctly beforehand.
A good start would be making up a custom aux lead with some 1k resistors in series with each channel.
>>931483 > It's rated for 60VA and 50Hz. From what I have read, it can run on 60Hz just fine. Yes, running a 50hz transformer on 60hz is fine. You can't run a 60hz transformer on 50hz however.
>How do I make it so that the amplifier can run on either 220 or 120v? Using a switch like they do on some PC power supplies, make sure its correctly set before powering it on. A metal switch lever must be earthed/grounded for safety. >pic related
>Is there such thing as a capacitor calculator to determine the optimum capacitance? Rule of thumb is about 2000uF per amp of DC load, for an audio amp you may want to increase this a little.
> I assume 35V capacitors will work just fine, right? For 12vac that gives just under 17vdc when rectified and smoothed, so yes 35v capacitors will be fine.
Read up on this guide about transformers and reciters+smoothing caps, you'll need to derate it for DC loads. With a full wave bridge rectifier and smoothing cap that 60va transformer will suddenly become a 37.2 watt transformer.
Hmmmm, if that transformer is not suitable for that then I bet get another, because it's heavy as fuck and not easy to mount to a chassis. If I made the amplifier single ended I'd need a voltage divider to bias the input stage, right?
>>931503 You mean that potentiometer in the image? If so that's just for volume control, you connect one channel of your audio source to "Vi" and the other end to circuit ground (the power supply negative in this case).
If you're feeling confident you could always add a couple more secondary turns around the core to get more voltage out, you'll want about 22vac for a 30vdc supply.
>pic The top voltages are for split supply whilst the bottom is single ended, y axis shows max output power compared to supply voltage.
I have 5 or 6 fax modem boards (ISA / PCI) which i want to use up somehow. I have an idea to connect one to telephone line and use it as answering device, but that's only one card. Do you have any more ideas what cool things can be done with them as they are?
I consider to cut their edge connectors and use them for my custom expansion boards, but want to double check if I'm not missing out on something.
>>928064 Overdriven LEDs (like the ones you find in a typical retail LED light bulb) burn out so you have to buy new bulbs. If you drive them properly, according to the original TI spec sheet, they might get a little dimmer after 50 years. I've never seen one in an indicator application burn out. Never. (Barring massive surges and/or lightning strikes)
>>931685 >a few caps and resistors and you can make sine waves
yeah, very very shitty sine waves. the usual purpose of a function generator is to test if your setup can produce sound without distortion. if your test signal is already 20% distorted, what the fuck is that good for?
you can muck around with discreet wien-bridge oscillators, colpitts oscillators, or Hartley oscillators, or could you stop wasting your time and get a dedicated sine-wave generating chip like the AD9833.
>>931883 Just get something that is rated way way higher than you need. Either that or wind your own. Inductive devices are not very complex. I sometimes wind my own transformers and inductors. For a solenoid you just need to add a spring and a permanent magnet.
What's the best way to model the loading effects of a 32 ohm speaker in Multisim?
Does a 32 ohm resistor work? I read somewhere you need a resistor and an inductor which kind of makes sense but I'm not sure what values I'd need. Maybe an inductor alone is enough but again, I don't know what value to pick. Since the frequency of music keeps changing the inductance value would need to be varied constantly to keep the impedance the same.
if you use an official crimping tool, the wire will be loose, or not as tight as it should be. if it's pliers or vice grips, you can tighten it as much as you want. also, you can use solder so size wont matter. this is what i do coz i dont trust pliers, and aint gonna pay $50 for a proper crimper.
>>931936 I use ebay for most stuff except caps. It's generally cheaper than most other places. Don't use it for capacitors though, or at least for electrolytics. Almost every capacitor lot is all really shitty cheap Chinese cap brands like Chang and Jakec. Don't bother. Other passive components like resistors are super cheap, especially if you order from China though they'll probably take a month to arrive and even if QC is bad it doesn't matter it's just a resistor. If you need 1%'s for whatever reason best go with a US seller though. They come in lots ranging from 100 to over 1000 usually for less than $10. You can get a lot of different ICs too though again I'd stick to US sellers to lessen the odds of getting fakes. Depending on the chip they'll usually be in lots of 10. I think I got 20x TL072's for $5 and 10x LM3915's for $7.
Other than ebay, Digikey is the place to go for most other parts. Use them for capacitors and stuff that's hard to find on ebay or other sites. Digikey will be more expensive but they have a huge selection and it's not cheap Chinese shit. Downsides are the site is more caters to enterprise needs rather than hobbyists, prices are higher, and shipping is usually expensive too. Also, shipping isn't calculated at checkout, you will get an estimate but all your shit needs to be gathered and weighed before you are charged for shipping, you may pay more than the estimate.
Hobbyists are fucked for electronics suppliers, most options are shit.
>>931430 I don't know about your test tube coolers and their requirements, but generally speaking, when you say things like > the coolers would allow much more fine tuned control, which would be useful better prepare for disappointment and/or lots of work. Point being, while the Peltier coolers themselves aren't difficult to control, the thermal contact from the coolers to the stuff inside your test tubes is very likely what determines the end result - in addition to the heat losses. At least your (stirred?) ice bath has about as good thermal contact with the test tube walls as you can reasonably expect.
>>931936 There is a small independent hobby shop 4 minutes from me that sells individual components as well as starter kits they make themselves. I suggest you visit there because its really convenient not having delivery delays
>>931929 You have discovered the inherent difficulty in reproducing sound. For non critical applications, a simple resistor will do. It's pretty difficult to accurately model a speaker coil's impedance and even more difficult to design an amplifier which can continuously match that impedance. There are few amps which are capable of doing so, but there are a very large number of amps which can reproduce sound very well despite not even considering variable speaker impedance.
...and if you're driving a speaker directly from a TL072, you have many more sources of much worse distortion that are far easier to correct than trying to compensate for load impedance variations.
Arduino fag here but I guess this applies universaly. The flourescent lights I have seem to fuck up the serial connection I have going between PC and microcontroller-board upon switching them on or off. Is there an easy way to fix this or another connection method I should try?
>>932184 Because it was a fun little project I thought of. Plus, it seemed like the next logical step after making it on the breadboard and writing a class to drive it. It makes it easier to plug into a breadboard... and just about anything, should I put female headers on it.
I'm not really attuned to the culture here, but I figured this is one of the best places to ask:
I'm interested in doing a robotics project. I have no experience with it (besides however far my CS education can carry me through the programming aspect), but I'm really eager to dumb a decent amount of money into a project.
Where is a good place to go to learn about applied robotics? Servos, actuation, control structures, etc. as well as actually constructing something physical that I can start to experiment with?
/g/ here, just trying to build myself a little generator. For fun. I have a cheapy 3 phase RC outrunner motor that probably gives out 12V if you push it. I want to make a full wave 6 diode rectifier, but my electronics knowledge is pretty primitive once you stop talking in amps and volts and go to ohms and stuff.
Basically, what diodes should I use for this? I'll have the thing geared to hell and back so I'm sure I'll hit the RPM/voltage I want, but I'm clueless. I know I'll need a schottky diode, but that's all I can say. I don't want to go overkill because this thing's not gonna be putting out too much power to begin with, I don't want to loose too much. Thanks very much in advance.
>>932601 Worth saying, I just found a pile of in5404's lying around. Just tested, I'm loosing 0.3V at 1.6v, so they're not too shit. Still, I wouldn't mind knowing what I should actually be using. Is it as simple as "What's the highest voltage and current you'll be using, well go get that one then"?
>>932601 You do not need a shottky diode, right now you're just trying to rectify the sinusoidal alternating current. Any general purpose diode should do. >>932605 diodes always have a 0.3-0.7 Volt drop, that's how they work.
By the way the outputs will not be DC+ and DC- they will all be DC around some voltage level probably about VPeak-Vdiode offset by the phases of the motor.
The figure should show what the rectified output should look like, i suggest getting an oscilloscope and verifying this before we continue to the capacitor bank, full bridge rectifier and then voltage regulator.
>>932637 Oh right, so you need power and a signal to tell it to go.
Before I read this, I was just upstairs trying combinations of wiring on my power supply. Most of them drew about 100mA without it doing anything (to my eye). But one combination caused a short circuit. I haven't damaged it have I?
>>932653 uhm yeah that sounds correct. Maybe you can find a wiring diagram on google? My list of signals to try would be: A 3.3-5 volt DC "on" signal. kind of like just a square wave with only 1 square. A low frequency sinusoid. ~3 volts or so and beyond that you might need PWM (pulse width modulation) which would require some kind of signal generator or a microcontroller
>>932678 I'm not that guy. I just saw the posts and thought I'd se a wave editor to simulate three phase rectification. The amplitude seems to have just over doubled. What's the supposed mathematical amplitude (with the source sinusoids being amplitude 1)?
>>932683 Ok so I just added the sines of 120 and 240 degrees (which were equal in magnitude, opposite in sign) and the total is 1.732051x, which for the amplitude of 0.3 I used for the source waves, shouldn't that give a peak of 0.519615?
>>932601 Voltage and current ratings need to be sufficient. Switching time needs to be significantly lower than the AC period (that might be an issue if you're running it particularly fast, but probably not). It's probably worth using Schottky diodes for the lower voltage drop.
I posted a while ago about the cheap chinese benchtop power supply that died after 5 minutes.
They sent me another one, free of charge and I finally got around to having a look inside the dead one. I don't really know what I'm doing (otherwise I wouldn't have bought the cheapest powersupply I could find) so I just took a few variable resistors, led displays, the fan and some transistors. Didn't bother touching the capacitors apart from discharging them with the 5w resistor that was elsewhere on the board.
Also ended up breaking my multimeter and fixing it, kek.
Well guys, I got it soldered and it's wonderful. Gives out 8v easily, 12 without too much more effort. I'd love to see it under an oscilloscope but I don't get back to work for a week. I won't have a stand for a couple days, but till then let's work on this capacitor thing.
>>be me >>Get new arduino kit >>look up parts >>hmmm this is a weird cap... Not labeled..... What? >>look everywhere to find out the caps stats >>give up hope >>after a month see a picture of the device >>mfw its not a cap at all! >>fml Pic reated
Has anyone here had any dealings with Hantek? I bought one of their cheap USB oscilloscopes and it worked well for a few weeks, then all of a sudden it stopped rendering voltages above 5v (its limit is 35v officially).
I sent them an email over a week ago and still no response.
>>933127 In a delta configuration, each winding produces a differential voltage between two of the three terminals. In a star (Y, wye) configuration, each winding produces a differential voltage between the common point (which may or may not have an external connection) and one of the three terminals.
In an isolated network, "zero volts" or "ground" is whatever you choose it to be.
E.g. that alternator/rectifier circuit will probably end up with the negative side of the rectifier being designated "ground". Even if the alternator has an exposed common terminal, it will appear as a ripple voltage relative to "ground".
>>933160 They have okay products but terrible customer service. I bought one of those too. The software is really really really shitty. Luckily, they have public apis so I actually wrote my own software for it.
>>933038 >I am also interested in this. So far I've only cannibalized them for the audio transformer, which seems quite wasteful. Well, the older modems I have are based on generic processors (mine have Intel 80188 running from external ROM), so they can be remade into somewhat generic microcontroller boards. Then somewhere there must be some small RAM chips and glue logic to the ISA bus. Newer ones often run integrated controllers and are often beyond repurposing.
>What would be the basic set up of an answering machine? >Just a PC running some software -> phone line?
Yes, essentially this is the setup. There are fax modems
>It would be wicked if I could get a raspberry pi to do it instead of a full fledged computer...
I'm making a tiny micro-controller based project for my "Embedded systems design" class. What I'm doing is in a tiny amount of space and requires the combined weight of the components to be as light as possible.
Currently I'm using a 2 layer PCB and an ATTiny85 microcontroller.
Anyway, my question. Will a combination of 2 x CR2303 cells be ok to power the microcontroller aswell as a few LEDs?
They are 3 volt lithium cells, combined would make (theoretically) 6 volts, which is above what you would safely consider CMOS supply (5.5 volts).
But obviously the cells aren't gonna provide perfect 6 volts, and the voltage would drop further when they provide some amount of current (current approximation is ~200mA so would last about an hour at ~220 AH rating).
Has anyone done something like this before? Will I fry my uC? What can be done to stop that from happening?
i'm pretty sure you're gonna have the opposite problem: too low a voltage instead of too high. because they're so tiny, CR2032s have a very high output resistance, so at 200mA the voltage will be close to nothing.
if the thing has a sleep mode, then in that case the voltage may be as high as 6.4V, but a 5.1V zener across the cells will knock it down to a safe level.
anyway, CR2032s are a bad idea for this kind of processor. if the specs allow it you can get very tiny LiPos about 1inch by 3/4 inch, which put out 3.7V to 4.2V. they have way more capacity.
>>933519 So when one is peaking, the ones with a lower voltage act as virtual grounds? But then aren't you only getting the voltage of the peak relative the the next highest wave? Still confusing. I'd have thought that, if single phase alternators have 2 connectors, then a 3 phase one would have 4.
All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.
This is a 4chan archive - all of the shown content originated from that site. This means that 4Archive shows their content, archived. If you need information for a Poster - contact them.
If a post contains personal/copyrighted/illegal content, then use the post's [Report] link! If a post is not removed within 24h contact me at firstname.lastname@example.org with the post's information.