I have a thermopile which produces a very small voltage (about 1-4 mV as far as I can tell). I would like to amplify this up to about 5 volts if possible.

From what I have read, an OpAmp shoule be suitable for this. I have a 741 OpAmp but I can't seem to get it to do anything with my small voltage, although I'm not entirely sure I have it wired correctly as I have no experience with OpAmps. The 741 has pins for Input Bias among other things and I'm not sure what this does.

I have it wired similarly to this:

With the + from the thermopile connected to the non-inverting input of the OpAmp but nothing connected to the inverting input, and a voltmeter as the load to see what voltage comes out of it.

Any advice on wiring the 741 OpAmp, or other ways of amplifying the voltage is appreciated.

Thanks,
Jet

 

An opamp has two inputs and your diagram has only one so there is a problem. Your input voltages are the same order of magnitude as the offset voltages and the input bias currents. You might have to research the instrumentation amplifier. It would probably be a good idea to get some ice-time with a modern part before you proceed.

 

Work through this first.
http://www.allaboutcircuits.com/vol_3/chpt_8/index.html

You should be looking for

The difference between inverting and non inverting inputs

Op amp supply arrangements.
The 741 is mean to have a positive and negative supply, balanced about zero (earth). Your circuit cannot work with a 741 because of this.
As PB says, there are more modern op amps that can work from a single supply.

Feedback arrangements to give the required gain (of the order of 1000) reliably.

Feedback arrangements to limit the op amp frequency response to prevent unwanted oscillation. The output from a thermocouple will vary slowly so you don't need a high speed op amp and need to limit the response to perhaps 500 Hz.

 

Thanks for the help so far. I have a couple more questions, however.

I had a look at instrumentation amplifier ICs but they don't seem to be nearly as easy to get hold of as OpAmps, so I'll stick with an OpAmp for now.

I now have a LM358N OpAmp which works fine on a single power supply, but I would like to make sure I understand what I'm doing before I go exposing my (rather expensive) thermopile to it.

Here is my circuit diagram:
The voltage source in the red circle is the thermopile, the green circle is the ground. Am I on the right track?

My worry is that this will try to put current through my thermopile. I put a voltmeter across where the thermopile will go and got a reading of about 30mV. Could this damage my thermopile?

Thanks,
Jet

 

An LM358 is a really old design, and probably leaks badly out the input terminals. If you think you need really sensitive amplification, then look at a modern FET input op amp. I could recommend an OPA129 with only 30 fA leakage, but TI seems to be hiding them lately. You can get the OPA124, though, with 1 pA leakage.

 

Will the leakage cause problems for the Thermopile? I don't mind if the amplification isn't 100% accurate or noiseless, I just don't want to damage the Thermopile.
I can't find anywhere to source an OPA129 or OPA124 from. Would a TL072CP do?

 

Some other FET input opamps:
LF353, TL071, TL072, TL080, TL081, TL082, LT1007, LT1057
The LF353 is the oldest one of the bunch I think. It's made by a number of chipmakers. I think TI came out with it first.
The TLs have been out for years, too.
The LT's are made by Linear Technology, and are pretty recent. The LT1007 is single, 1057 dual.

 

Sure, the TL072 should work fine.

 

Sure, the TL072 should work fine.

But it won't drop into the schematic he posted, because the input common mode range doesn't include ground, and the output won't get anywhere near the negative rail. He'll need a negative power supply in addition to the positive supply.
If he wants to stick to a single supply, and can't make the LM358 work, the easiest part to find would probably be a rail-to-rail I/O op amp, even though he probably doesn't need the inclusion of the positive rail.
I doubt the bias current of the LM358 would be a problem. It certainly won't damage a thermopile, which is just a stack of thermocouples - very low resistance devices. I think the biggest problem would be input offset voltage, which Papabravo has already mentioned.

 

Thanks Ron, it works with the LM358. What is the problem posed by the input offset voltage? It seems to be working ok to me but I could be missing something.

Edit: It seems there are a lot more OpAmps for supplies with positive and negative than there are for positive/ground. Is there a way to make a positive/negative supply?

 

Thanks Ron, it works with the LM358. What is the problem posed by the input offset voltage? It seems to be working ok to me but I could be missing something.

Edit: It seems there are a lot more OpAmps for supplies with positive and negative than there are for positive/ground. Is there a way to make a positive/negative supply?

As Papabravo said, the input offset voltage may be more than your maximum signal voltage. When you apply gain, input offset is indistinguishable from the output of your thermopile, and it can add to it or subtract from it. Or, it might be zero.
You can find out what the offset is at the amplifier output by shorting out your thermopile. Any change from that value is signal from the thermopile.
You can buy op amps with essentially zero input offset voltage.