Transistor output voltage

[spiffitz]

I need some schooling on transistors. The device I'm testing uses a 24v relay, triggered by an NPN transistor. The transistor checks out fine - resistance check between E, B, C, wet finger test, etc. When it's in the device and powered up, B shows ~28v, E at ~26v, and C (which goes directly to the relay) is putting out zero.

What am I missing?

 

[reese]

If I am understanding your circuit properly, the emitter should be going to ground and you should meas. 0Vdc. If your measuring 28V on the base, there is definitely a problem. Your B-E voltage (B-grd) should be around 0.7V typically. It almost sounds like what you are calling the Emitter is in fact the collector, because given what you have just described, the collector should meas. 28Vdc, under these conditions. But with 0V on the collector, it seems like the transistor is conducting, which doesn't make sense if you have 26-28V on the emitter. Check your ground connection to the emitter of the transistor. You may have an open ground or burned out coil. Have the ohmed the coil of the relay?

Reese

 

[Space Varmint]

I need some schooling on transistors. The device I'm testing uses a 24v relay, triggered by an NPN transistor. The transistor checks out fine - resistance check between E, B, C, wet finger test, etc. When it's in the device and powered up, B shows ~28v, E at ~26v, and C (which goes directly to the relay) is putting out zero.

What am I missing?

That's because the transistor is not turned on. Like Spiffitz said, you must see the .7 volts drop ot it is not biased correctly. What do you have on the base (resistors)? Anyway you can post a diagram?

 

[Hero999]

Please post a schematic.

 

[spiffitz]

Regardless of which pin is what, the pin that should power the relay is showing zero volts, and the other two show 26 and 28. After finding the base using a diode test on my MM, I did the following as found on a site:

Connect the meter leads to the unknown leads of the transistor. If the transistor is NPN, wet your finger and use it to connect the base of the transistor to the positive lead of the meter. If you get a reading, then the negative lead of the meter is connected to the emitter. If you don't, reverse the meter leads and repeat. If you still don't get a reading, then the component is burned out or is not a junction transistor.

Based on that (is he backwards?), the collector is the point that shows zero volts. One pin of the relay has power all the time, the unpowered side goes to this point on the transistor. The relay is good, I have tested it off-circuit.

Sorry, I don't have a diagram for it, all I can do is take a picture or make a crude drawing. I have emailed the company for a schematic.

 

[Space Varmint]

Regardless of which pin is what, the pin that should power the relay is showing zero volts, and the other two show 26 and 28. After finding the base using a diode test on my MM, I did the following as found on a site:

Based on that (is he backwards?), the collector is the point that shows zero volts. One pin of the relay has power all the time, the unpowered side goes to this point on the transistor. The relay is good, I have tested it off-circuit.

Sorry, I don't have a diagram for it, all I can do is take a picture or make a crude drawing. I have emailed the company for a schematic.

Oh, I see what your saying. Your trying to fix something. Yeah, it sounds to me like the transistor is blown. Not uncommon. Is there a diode across the relay coil? If not, then relays are know to blow transistors from the reverse collapsing field of the coil when it is de-enegized. They call it "back EMF".

 

[spiffitz]

Is there a diode across the relay coil?

There isn't a diode per se, but there is a transistor there which I assume is being used as a diode. Only two of the leads are connected.

If the transistor tests OK off-circuit, it's still a likely possibility that it is bad on-circuit?

 

[reese]

There isn't a diode per se, but there is a transistor there which I assume is being used as a diode. Only two of the leads are connected.

If the transistor tests OK off-circuit, it's still a likely possibility that it is bad on-circuit?

It is possible that the transistor is defective, even if it measures good. Although it is rare, I have seen it. It sure would help if you could provide a schematic for us to look at. But it sound like it is being used as a simple switch for the relay.
Have you tried measuring resistance from ground/chassis to the emitter of the transistor? You should meas. 0 ohms. If you are meas. 28V on the base to ground, than there is a problem.
Let me ask a question here. Why do you have a system controlling a transistor with 28V driving a 12V relay? Are you sure about these meas.? If the relay is good than you may have a cracked solder joint on the collector of the transistor or an open trace. Ohm out all connection points to the transistor to verify you have connectivity.

 

[reese]

I need some schooling on transistors. The device I'm testing uses a 24v relay, triggered by an NPN transistor. The transistor checks out fine - resistance check between E, B, C, wet finger test, etc. When it's in the device and powered up, B shows ~28v, E at ~26v, and C (which goes directly to the relay) is putting out zero.

What am I missing?

All right. Sorry, I am trying to keep track of too many threads here. You don't have a 12V relay. Sorry about that.

 

[Space Varmint]

There isn't a diode per se, but there is a transistor there which I assume is being used as a diode. Only two of the leads are connected.

If the transistor tests OK off-circuit, it's still a likely possibility that it is bad on-circuit?

Yeah, that's a good observation. That is probably what they are doing. Not the best method though. I would just replace them both, but use a real diode this time.

 

[reese]

Regardless of which pin is what, the pin that should power the relay is showing zero volts, and the other two show 26 and 28. After finding the base using a diode test on my MM, I did the following as found on a site:

Based on that (is he backwards?), the collector is the point that shows zero volts. One pin of the relay has power all the time, the unpowered side goes to this point on the transistor. The relay is good, I have tested it off-circuit.

Sorry, I don't have a diagram for it, all I can do is take a picture or make a crude drawing. I have emailed the company for a schematic.

Try ohming out the point of the relay that has power all the time to the collector of the transistor. You should meas. the coil resistance, what ever that is. This will verify the pin that has power is indeed the high side of the coil and not one of the contacts that switches power to the load.
Also, try meas. both E-B and C-B junctions. If you read .7V on both emitter and collector to the base, with the positive lead on the base, than the part is an NPN. It would make more sense to suggest what you have is a PNP part supplying 28V to the coil (emitter side high and collector to coil). That is why you need to verify the powered pin on the relay.
What is the mfg part number on the relay?

 

[spiffitz]

Try ohming out the point of the relay that has power all the time to the collector of the transistor. You should meas. the coil resistance, what ever that is.

Yes, it does return the same value.

Also, try meas. both E-B and C-B junctions. If you read .7V on both emitter and collector to the base, with the positive lead on the base, than the part is an NPN.

I did that when the transistor was out and they read .645v, close enough?

Have you tried measuring resistance from ground/chassis to the emitter of the transistor? You should meas. 0 ohms. If you are meas. 28V on the base to ground, than there is a problem.

I wasn't too clear on the first post. When I meant by 26v on the emitter, that was the + led on the relay, and the - lead on the emitter. Following the emitter ground it ends at a dual-color LED, at one of the "thin" leads (anode?).

I'm working on a schematic. The middle of the board is obscured by a large AC/DC relay and I will add the rest of the schematics tonight once I get that relay out of the way.

**broken link removed**

Could a bad LED be a possibility? The LED is green when there is no heat, and should turn red when there is heat. When I tested it using the diode function on my MM I think I only got voltage between one side lead and center lead but I wasn't sure of the correct way to test a bi-color LED. It is soldered back in, but I can remove it if someone suggests a better way to test it.

 

[audioguru]

Your schematic shows N-channel FETs, not NPN transistors.
The LEDs and the FET do not connect to ground so they will not work.

 

[spiffitz]

Oops, right you are. Better? This is not complete, just what I know without pulling stuff off of the board. Basically, "middle Q" (forgot to label them) has no voltage coming out of the collector. With meter + on the relay +, meter - on base shows 28v, and meter - on emitter shows 26v.
**broken link removed**

 

[audioguru]

The relay's coil is not connected to a positive supply (+28V?) so it and the middle transistor won't work.

The LEDs do not have a connection to ground so they and the middle transistor won't work.

Connect the meter's - lead to ground then tell us the voltages at the base, emitter and collector of the middle transistor when it is supposed to be turned on.