Check: Simple Circuits

[Devil_H@ck]

I've just put together some simple circuits. I'd very much apreciate it if someone could them me if they will work, especially the 1st, which I put together myself. The others are pretty much standard design.

I tried to simulate the first one, but SIMetrix wouldn't let me. All it has to do is: if there's a signal on one of the input signals the relay has to switch on. Signals on the diodes will be at least 9V, with a maximum of 12V (maybe 18V, but the chance is pretty small).

Ok, so here it is:

An externally hosted image should be here but it was not working when we last tested it.

Here are the 2 proven designs, they should work, but I figured I might as well post them to make sure I didn't make any stupid mistake.

Power supply:

An externally hosted image should be here but it was not working when we last tested it.

On/off signal with push switch:

An externally hosted image should be here but it was not working when we last tested it.

The out signal is going to be one of the inputs on the first circuit.

 

[Devil_H@ck]

Ah, c'mon, sure there must be someone who can tell me if it will work by just looking at it? Half of you guys are EEs.

 

[Frank Berry]

If you put the relay coil in series with the collector rather than the emitter, the relay will pull more reliably with input voltages that are less than the relay's coil voltage. You might also want to put a 100k resistor in parallel with a .1uf capacitor from the transistor's base to ground to make the circuit less prone to false triggering due to noise.

 

[Devil_H@ck]

If you put the relay coil in series with the collector rather than the emitter, the relay will pull more reliably with input voltages that are less than the relay's coil voltage. You might also want to put a 100k resistor in parallel with a .1uf capacitor from the transistor's base to ground to make the circuit less prone to false triggering due to noise.

Thanks . One question about the resistor & cap to earth though. Is this really necessary? There will either be a signal or no signal at all on the diodes, nothing in between.

*Edit: Ah well, nevermind, I managed to added it to the schematic and I only had to make the PCB 25 mils longer .

 

[Frank Berry]

Yes. The resistor and cap are necessary. Without them, the base of the transistor will see an extremely high impedance when none of the diodes are conducting. The circuit will be very sensitive to RF interference and other "noises."

 

[Devil_H@ck]

Yes. The resistor and cap are necessary. Without them, the base of the transistor will see an extremely high impedance when none of the diodes are conducting. The circuit will be very sensitive to RF interference and other "noises."

Ok, thanks for the info!

 

[Devil_H@ck]

Hm, OK, I got kind of a problem. I made the circuit about 2 weeks ago, but hadn't tested it yet. Yesterday, I decided to do so and the thing has kind of a very annoying flaw. It only turns on, I can't turn of the relay, no matter what (well, OK, shutting down the power helps ).

Why is that? Here's the schematic I used:

An externally hosted image should be here but it was not working when we last tested it.

Should the cap & resistor be in front of the 10K resistor or is it something else?

Oh, btw, I used a 47nF cap instead of a 100nF cap, but I can't imagine that that's causing the problem. I even took out the cap and tested again and the result was the same: it only turn on, not off.

Ah, and you can find the PCB layout on my website: http://members.lycos.nl/anthonyvh/index.php?page=pwrsw

 

[Elso Kwak]

Darlington

Hm, OK, I got kind of a problem. I made the circuit about 2 weeks ago, but hadn't tested it yet. Yesterday, I decided to do so and the thing has kind of a very annoying flaw. It only turns on, I can't turn of the relay, no matter what (well, OK, shutting down the power helps ).

Hi, Use two transistors Darlington connected!

 

[Devil_H@ck]

2 darlingtons in series in front of the relay coil? Or 1 in front & 1 behind?

 

[Elso Kwak]

2 darlingtons in series in front of the relay coil? Or 1 in front & 1 behind?

Two transistors Darlington connected before the relais or one Darlington transistor at the same position.

 

[Devil_H@ck]

Ok, I'll try that. Why is it that it can't turn off? Is it because the "high" current is keeping the transistor open?

 

[amb]

Are you sure you're using a simple (rather than a latching-type relay)? Also, since the input voltage is high (9-12V) and the transistor will turn on at 0.6V or so, you might want to change the 100K ohm resistor to a much lower value (something like 1K ohm, so that it forms a voltage divider with the other 10K resistor to drop the input voltage.

-Ti

 

[Devil_H@ck]

Are you sure you're using a simple (rather than a latching-type relay)? Also, since the input voltage is high (9-12V) and the transistor will turn on at 0.6V or so, you might want to change the 100K ohm resistor to a much lower value (something like 1K ohm, so that it forms a voltage divider with the other 10K resistor to drop the input voltage.

-Ti

It's a simple relay, I've used them at least 5 times before and they are just plain, normal relays. Nice ones too .

The input voltage is 12V. I'll try changing both resistors, but would there be any difference? Zero volt is zero volt, the problem isn't turning on, it's turning off. Even if the trigger voltage were 200 volts, once there isn't any signal at the trigger input, it's 0 volts, no matter what. Right?

I'll go find myself a darlington tomorrow & then try changing resistors if it still doesn't work .

 

[HDTVman]

Do you know that the base is at 0 volts? Have you measured it? The lower value resistor from base to grd will insure that the base gets pulled to grd even if there is a small amount of leakage in the transistor and remember, caps take time to discharge. Oh ya. (Have you tried the darlington yet? oops I didn't check the time stamp. sorry)

BZ

 

[Devil_H@ck]

I got 2 BC517s today, according to the storekeeper they are 1A darlingtons. I wanted BC317s, but he didn't have those and BC517 seems to resemble it close enough .

I'll try it tonight.

* Edit:

Ok, I looked the thing it up:
High current (max. 500 mA)
Low voltage (max. 30 V)
Very high DC current gain (min. 30000)

Should be fine.

 

[Devil_H@ck]

Ok, it works perfect! Thanks Elso!

Hm, well, almost. There's one weird malfunction kind of thing, but it isn't that important. When I switch the relay on & then off and then pull the power chord so that there isn't any power left, the relay closes for a very short time. Not that this only happens when it's in the off state & the power chord is removed from the power supply, when turn it on/off it works just fine. What could be the reason of this?

 

[amb]

Are you still using the 100K resistors or did you already change them to 1K?

-Ti

 

[Devil_H@ck]

Ah, yes, that's probably the problem. Let's see if I get it correctly:
- The relay turns on, the 47nF cap charges
- The relay turns off, the 47nF cap starts to discharge very slowly through the 100k resistor
- The power is removed, the 47nF cap tries to keep the voltage it was fed by at the base of the darlington, thus turning it on. In turn the relay turns on and drains all the power left in the supply's caps.

Hm, actually, rereading it, it seems wrong, since the cap would do the same when I just turn off the relay, right? So, please, tell me, why is it that it turns back on for a short time (untill the supply's caps are empty)?

 

[FredM]

Hi - Lets look at the theory - the transistor will turn off when the voltage on its base is less than Vbe (about 1.4V with a darlington) -

Provided the EN signals are all below (0.6V+Vbe), the time taken for the 100n b-gnd capacitor to discharge will be negligable - the voltage on this capacitor will not charge to more than Vbe anyway (it will be clamped) so it only needs to discharge a small amount for the transistor to turn off. with t=rc, 100n*100k=0.01 (10ms) but I would expect the time to be more like 500us given that rapid discharge occurs first, and that only a few mV drop in Vbe would be required.

I can see no mechanism for the transistor to be turned-on briefly as the supply drops - except if some spurious signals are coming from the EN inputs.

An externally hosted image should be here but it was not working when we last tested it.

 

[Devil_H@ck]

Hm, ok then. It doesn't really matter, because it only happens when the power plug is pulled. The relay is used to conduct the AC to the torriods in an amp, so there won't be any more AC to conduct.

But thanks for the explanation, I always forget about the transistor voltage drops.