Basic science has failed me. I've tried to make a basic

>>8078911
did you use matched transistors? with a time constant of 1s it's likely to be significant

>>8078950

I'm not entirely sure what you mean by matched transistors? They are the exact same type, everything about them is the same as far as I can tell.

For the time constant thing, I need something that oscillates at 225 seconds and another thing that oscillates at 128 seconds.

I thought of using raspberry pi's, arduino's, 555 timer chips but in the end I figured that all I really need is an astable multivibrator as pictured.

>>8078964
your circuit looks fine which means either you mixed up some components or there's a problem with the components themselves.

matched means they're made from the same silicon substrate so they have the same saturation current and thermal voltage.

Lastly, I would recommend just using a 555. it's going to cost the same and they jest werk.

>>8078974

Will try the 555 timers. Any idea what the max interval on that thing is? I figure it's less than 8 hours, hence the 128 and 225 second intervals.

>>8078980
depends on the brand but usually you can get up to a few hours.

here's a calculator to make things easy
http://www.ohmslawcalculator.com/555-astable-calculator

>>8078990
Ok I tried the calculator but the signals I can get from it are unusable for my purposes.

I basically need the following signals:
128 seconds low with a blip high, that switches a monostable circuit to run for just under a second.
The other one the same but with 225 seconds instead of 128.

This is why the astable multivibrator one was so appealing, I could just use the blip once it switches to switch a monostable circuit that runs for just under a second. I'm not sure how to do that with the 555.

>>8079007
I'm not sure I understand the waveforms you're looking for. Do you think you could post a sketch and also explain how your circuit above was supposed to achieve it?

>>8079021
I'll draw something out, hang on.

>>8078911
You're probably running one transistor in reverse because the circuit looks more 'symmetric' that way.

File: signals.png (79KB, 2042x811px) Image search: [Google]

79KB, 2042x811px

>>8079021

I hope this makes sense.

The astable multivibrator to the left infinitely switches every 128 seconds. On every switch, I want to trigger a monostable circuit that is "on" for slightly less than one second.

The other signal is the same except for 225 seconds.

>>8079060
the "Use each switch to trigger" basically means that every time the hi/low sides switch that it should trigger the monostable to the right.

>>8079060
the first signal is achievable with an astable 555, and the second likewise with a FET on the output the invert the signal.

The same applies for the last one.

>>8079080
Will try, though I still don't understand why my original multivibrator doesn't work proper. All components are exactly the same.

>>8079092
you checked that your circuit is wired correctly?

like >>8079045 suggested it's very easy to have the transistors in the wrong order which (depending on the pin config) could mean that it's in reverse active and would probably account for difference.

>>8079099
Yes, I've run through the circuit several times, including the transistors and the LED's. Heck, even the voltages and operating temperatures on the capacitors. It's really weird and discouraging.

>>8079092
Can you post a picture of your breadboard setup?

>>8079121
Lemme see what I can do. It'll be a webcam pic so I hope it'll be good enough.

>>8079110
analogue is difficult m8, I wouldn't worry about it.

there's a reason 555s are so popular and timing circuits are a pain in the arse generally.

File: bbsetup.png (318KB, 349x314px) Image search: [Google]

318KB, 349x314px

>>8079121
Here ya go. The inner two resistors are 1M, the outer ones are 1k. The rest is as described earlier.

>>8079125
>>8079080

Ok with the 555 timer I've calculated the following:
C = 100uF
R1 = 0 Ohm, but I doubt this is a healthy decision
R2 = ~3.25 kOhm (of course I won't work towards that many significant digits).

File: 1463426473745 (1).png (242KB, 349x314px) Image search: [Google]

242KB, 349x314px

>>8079154
From the picture, it looks like the left capacitor's positive end isn't in the same row as the 1 Meg

>>8079195
Thanks for the catch. I've fixed it. The reason that was the case that just before I put it away I tried different capacitors and misplaced this last one.

After the fix the problem persists though. Meaning, this wasn't the error that caused the different intervals.

>>8079200
Is the right LED really connected to the emitter of the right transistor? Looks like it goes to the base.

>>8079209
Yes, both leds are connected to the emitters.

>>8079190
I've re-computed the values:

C = 10 uF
R1 = 100 Ohm
R2 = 32467.5 kOhm = 32 MOhm + 467 kOhm + 500 Ohm

>>8079200
So this pretty lame, but if you haven't already, try switching out left components to the right and vice versa to see which component brings the problem to the other side. Transistors first, then capacitors

>>8079259
That'actually something I hadn't even considered. I just swapped components with my outside resources. All equivalent parts of course.

Result of your advice: no effect.

I feel like I might be able to solve it if I just understood the practice of electronics and not just some superficial and basic theory.

As a rule of thumb for stable operation the ratio Rb/Rc should not exceed the B of the transistors used. In your case it is 1000 and the BC337 has a rather low B at such low currents. I would replace the 1M resistors with 100K and increase the capacitors to 100µF and see if that works. A better suited transistor for low currents is the BC550C (or BC548C or BC549C).

>>8079299
Wow! That greatly helped. They're just about equally timed now. I don't need this particular circuit, however, I did need to know why the theory broke up at the values I used initially.

I'll be sure to look into transistor specifications on next builds.

Any tips on how to be more weary?

>>8079328
>about equally timed now
Don't forget that this type of capacitor has a rather large tolerance, like ±20%. If you exchange the two you might see the effect.
>why the theory broke up
I don't think it did. What looks like a symmetric circuit is in reality a two stage 'linear' amplifier with positive feedback. This needs a certain loop gain for stable operation. If it is barely larger than 1 (like with the 1M resistors and the small B transistors) you probably will not arrive at the expected behavior.

>>8079388
I suppose the biggest lesson then is that basic tutorials are just that: basic and overly simplified.

>>8079286
Did you try replacing the breadboard?