Dammit! Stop right there. WHAT 386 ?? That's the first time you've mentioned a 386. Where's the schematic????????????????
Yes, and it will continue to be baffling until you use some rigor in your posting. You mention power amplifier, current draw increases dramatically, plenty of AC, frequency, a cap, brighter There is not a specific given for any of that. For example, there's no description and no diagram of how you measured "plenty of AC" or what "plenty" means. You said you were going to post a schematic. Do it.
When exasperation reaches the level my wife can show, I know I'm in trouble. ;-)
This is what I'm using currently:
formatting link
I cobbled this together quickly, FWIW:
formatting link
The two rails couple the layout box on the left to the engine on the right. The AC is injected where shown, from the LM386 circuit. The DC is applied to the left of the inductor.
At the engine, the AC is isolated from the motor via the inductor, and coupled to the rectifier through the cap. The lamp is then illuminated by the derived DC.
The problem is that a load across the motor -- which draws more current through the rails, of course -- also results in more current drawn from the signal generating circuit (the LM386). That's based on direct AC current measurement where the signal's injected on the left side. I don't understand that.
I'd imagine it has to do with the overall circuit tuning. When the motor is sped up by increasing the DC voltage applied the rail, the lamp extinguishes until the motor comes up to speed, reducing DC current draw. The lamp gradually brightens again as the motor current draw ramps down.
However, the lamp does not brighten fully to where it was when the motor was not runnning (no DC applied to rails). I noticed that adjusting the frequency of the AC upwards -- from 20 kHz to 25 or more
-- resulted in a brighter bulb again. My untutored inference is that the DC voltage and/or the additional current draw is changing the tuning of the circuit.
Thanks for bearing with me on this. It's been WAY too long since I've been in this stuff, and I was never knowledgeable about motors and how they interact with circuits.
Pardon the wrong symbols, BTW. I just d'loaded TinyCAD quickly to do this, and the libraries seemed incomplete. Is there a better open- source app for this?
What is the the power source for the 386? What is the signal generator feeding its input? Why are you using a bridge rectifier between C1 and the lamp? Get rid of it.
You can draw in ascii. Examples: ---[R1]--- ---[C1]--- ---|
1) this is done all the time. Call up the hobby shop and ask them what they do.
2) Aren't there model railroad newsgroups or websites where you can talk directly to people who have done this?
3) Put AC on the track. Use the + half to drive the train and the negative half to drive the light.
On the test bench I just have a variable power supply on the LM386; the circuit's working well with 9VDC. Likewise, I have a signal generator that's letting me vary the frequency.
The bridge is not needed with the cap isolating rail DC from the lamp, true enough. It's there because we're forward-looking to some on-board electronics. We can't use the rail DC because that can vary from several volts down to nothing.
I'm heading over to the shop, and we'll lean down to the cap and see how things work. I'm just baffled by the way the circuit draws a quarter amp from the amp circuit when the motor is under load. I think that quarter amp is killing the signal, then, so on the one hand I think "I need to deliver more current on the amp's output!" but on the other I just wish I knew why a motor under load is causing the circuit to load the AC so much.
spamme0: There's no in-between commercial system that does this. Trains either run simply on DC with no rail-delivered action such as light control, OR they're digitally controlled engines with a constant power applied to the tracks. My fabricator is an expert in the O scale trains, and no one's doing this. (3) not possible. among other things, trains need to go in reverse. :-)
Off to the shop to play again. Meanwhile, the technical question is this: why does a loading mother change the tuning of the reactive components in this schematic, such that the AC source is insanely loaded and the AC effectively gets shunted to nothing?
What tests have you done to verify that the circuit is working well? Working well means it is delivering a healthy signal at the output *while under load*.
Is the ground from the sig gen connected to the 386 ground and to the track ground?
The bridge is dropping ~ .7 volts on each side of the AC. That's one reason to get rid of it.
Next: what is the frequency measured at the rails with the motor load and without the motor load? If the frequency drops, the inductive reactance drops, and the current draw goes up. What is the p-p measurement at the output of the 386 and at the rails when the motor is loading the rails?
Drawing a quarter amp from your 386 is way too much - it may have killed the chip. Double check your inductors to verify they are 2 milihenry, not 2 microhenry. Either you don't have 20 kHz through the inductor or it is not 2 mH. You can't draw 250 mA from your circuit, unless the frequency is a lot lower than 20 kHz, or the inductor is lower than 2 mH.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.