Relay switching and transistor question; keeping it quiet

[R.G.]

There is NO DC on the pots.

Hmmm. Very odd. I'll have to think on this some more.

except using a NPN with a negative supply voltage. For R1, I'm using:
R1 = Supply Voltage / ( Maximum Current Required / Minimum HFE * 1.3 )

Where supply voltage after the drop is about 12v, coil current is 12.5ma, HFE from the data sheet is 300. That yields R1 of roughly 220k (closest value) but that doesn't work, in fact the only R that does work is 0. Where am I going wrong there?

Where you're going wrong is that HFE is not fixed. It's a variable, and it drops dramatically when the transistor starts to saturate.

For a coil current of 12.5ma, I'd assume I could get a gain of 10, and use a base resistor that gave me 1.2ma into the base. The biggest resistor I'd use would be 9.1K, and I'd probably put in 4.7K or 6.8K.

Try a 6.8K if you have it. Then measure the voltage from collector to emitter on the transistor with coil current flowing. If it's not under 1V, something else is wrong.

[Enzo]

I usually grab a darlington to switch relays. Not sure I always need to, but it works for me. And I already have a couple hundred MPSA13 in my drawer.

[R.G.]

That's a very decent solution. Should work in almost all cases.

In fact, about the only cases it wouldn't work in are where there's a minimal voltage to run the relay, and the difference between a BJT saturating to under 0.5V and the inability of Darlingtons to get under about 1.4V would matter. Not very likely.

[wizard333]

I tried every value between 220k and 35 ohms, none would let the relay switch. The only way it switches is if there is no R1 at all. Is the circuit I'm using correct given a negative supply and a NPN? As show it is for a PNP and a positive supply.

Would transistor switching lower the noise assuming it was working correctly?

[Jazz P Bass]

If you are going to use that PNP circuit with an NPN transistor, remove R2.

[R.G.]

I tried every value between 220k and 35 ohms, none would let the relay switch. The only way it switches is if there is no R1 at all. Is the circuit I'm using correct given a negative supply and a NPN? As show it is for a PNP and a positive supply.

1. What circuit are you actually using? The NPN or PNP?
2. What are the relay coil ratings for voltage and resistance?
3. What is the actual value of the power supply voltage with relay off and relay on?

The comment that it won't work for any R1 tells me that there is something else going on.

Would transistor switching lower the noise assuming it was working correctly?

Depends. Metal-to-metal switching is always instantaneous with several bounces a few milliseconds apart. Transistor switching can be either very fast or slowed down where metal can't. Whether you get lower noise depends on where the noise is coming from. If it's the coil switching, yes, transistor switching offers at least a chance to quiet it. If it's the relay contacts themselves, no, no amount of dinking with the coil drive will help.

Details matter.

[wizard333]

1. What circuit are you actually using? The NPN or PNP?

As I noted a few times, NPN.

2. What are the relay coil ratings for voltage and resistance?

12V nominal, 970 ohms, they don't give release voltage on the data sheet, but as I noted elsewhere, it seems to be under 2v.

The comment that it won't work for any R1 tells me that there is something else going on.

That's what it tells me as well.

If it's the coil switching, yes, transistor switching offers at least a chance to quiet it. If it's the relay contacts themselves, no, no amount of dinking with the coil drive will help.

Is there any way to know that a-priori without hooking up the transistor switching to find out?

[R.G.]

So, on to the third question: what is the actual value of the power to the transistor and coil when the coil is on and when it's off?

Is there any way to know that a-priori without hooking up the transistor switching to find out?

Not without some good test equipment. It's simpler to hack in the transistor switching and see if it fixes things. Especially if you're already 90% of the way there with the NPN.

[wizard333]

what is the actual value of the power to the transistor and coil when the coil is on and when it's off?

Can you clarify? Do you mean the VDC? I'm not sure where you're headed. Either the diagram is right or it isn't?

[Jazz P Bass]

Hook up the device as shown & it will work.

[R.G.]

Can you clarify? Do you mean the VDC?

Yes.

I'm not sure where you're headed.

I'm headed for figuring out whether the power supply itself plays a part in how the circuit works, since we *both* think something non-obvious is going on.

Either the diagram is right or it isn't?

That's a questment worthy of some philosophical discussion on its own, but there is an implicit dependence on the power supply in all such circuits. If the power supply sags on load, then it can cause Funny Things to happen. A wizard should be very familiar with this issue.

[wizard333]

Hook up the device as shown & it will work

Except that's a positive supply and there is no "chip output"? I'm dealing with a negative supply.

DCV from the PS after the resistor/before the relay is -13v ish, drops down to -10v ish with the coil grounded. That's with no transistor in there, I took it out for now.

[Enzo]

Chip output just means a positive control voltage coming from somewhere. It happens to be a chip in wherever he cut that from. If you are going to use a transistor you are going to have to make current flow through its base from somewhere.

You want to run from a negative supply? Fine, use the exact circuit he posted, except it will have negative at the top. Replace the NPN with a PNP. Now a negative base voltage will turn it on... Oh, and reverse that diode parallel the relay coil.

[Dave H]

Except that's a positive supply and there is no "chip output"? I'm dealing with a negative supply.

DCV from the PS after the resistor/before the relay is -13v ish, drops down to -10v ish with the coil grounded. That's with no transistor in there, I took it out for now.

You want something like this then.

[Steve Conner]

Re the clicks and pops: What is the next stage that all this stuff is feeding? Does it have a path for its input bias current during the few milliseconds between "break" and "make" when the relay's changeover contact is floating? A full schematic would help answer questions like these.

Re the transistor switching: What are you driving the transistor from? Is it some logic circuit or just a footswitch contact? For a footswitch, the circuit shown by Dave H above should work fine.