The NE5534 is not a pin-for-pin replacement (it's single vs dual opamp), requires a dual supply, and has a 0.5mV or higher input offset.
The CA3140 has basically the same input offset specs as the LM358, requires a dual supply, and can't output more than Vcc-2v, so with a Vcc of 5v, 3v would be the maximum out voltage.
Both opamp designs are pretty old.

Even if you obtain a more modern RRIO opamp, it's going to be tough to get below ~10mV out without using a dual supply.

Are you committed to using a single 5v supply for this project?

Does the board already exist, and is that why you are looking for a pin-compatible replacement?

A bit of a bump...

So after getting V1 working and seeing what you were warning about, I now have V2 working will your suggestions. So far, much better! Thanks!

Last thing I need to do is sort out what you were talking about in terms of filtering. I have a 10nf cap alongside a 10uf cap next to the opamp. Once I hook up the MCU I will be able to see if it is less noisy and accuracy.

thanks again for the help

 

Backing up a bit - a couple of weeks ago, you mentioned:

The 38k is actually a 47 and 220 in parallel.

Well, that's not 38k either - it's 38,726.6 Ohms; closer to 39k than 38k.

I wrote a little utility the other day that I called ResCalc (seems others have beat me to that name, but it's just a Beta version anyway)... I posted it in this thread:
ResCalc - a freeware series/parallel resistor calculator for E12-E96 series
Download, unzip and run ResCalc.EXE from the ResCalc.zip attachment in the 1st post. Enter 2 to use E24 values, then type in 38000 for the value you want.
It will give you 26 options to get much closer to 38,000 Ohms than the combination you used.

The opamp should have a 0.1uF (100nF) ceramic or poly metal film and a 1uF or larger electrolytic in parallel across the opamp power supply pins.

 

Sorry, it mis-wrote that. The calc I have 38.73, not the rounded one I put

I did see your post on the rescalc, I'll download that and try. The current android one I am using is handy but a bit manual for stuff like that.

As for getting the values and gain exact, not sure if I am being lazy here but I am not too fuzzed. The output will be measured by the netduino so all I need to know is what is zero, what is 75mV and is it linear and I can do the rest in the code. My only requirement was that I got the 75mV to below 3.3 which is the max I can measure.

I also thought about putting a 22turn 4k7 (for example) trim pot between the output and the IN of the OP in series with the existing resistor. This would allow me to tweak it.
All seems pointless though because I can do it in software.

Is this a bad way to look at things?


edit
will alter the caps as well - thanks.

 

Even if you obtain a more modern RRIO opamp, it's going to be tough to get below ~10mV out without using a dual supply.

This is not *entirely* true.

Since he expects only positive output voltages, inserting a forward biased diode on the output of the op-amp and loading appropriately with a resistor to ground can produce output voltages in the 10s of microvolts or less while maintaining overall accuracy.

The max Vo will be reduced by 0.7V.

This was a favorite trick of the late Bob Pease.

 

so the feedback resistor would be post the diode? Would this then increase the output because the opamp thinks its output is 0.7V (drop across the diode) lower than it actually is?

 

As for getting the values and gain exact, not sure if I am being lazy here but I am not too fuzzed. The output will be measured by the netduino so all I need to know is what is zero, what is 75mV and is it linear and I can do the rest in the code. My only requirement was that I got the 75mV to below 3.3 which is the max I can measure.

Having a good zero-drift low-offset opamp in there will minimize the need to recalibrate it periodically. There will always be some drift, but the good ones have such a small amount that you wouldn't be able to measure it.

I also thought about putting a 22turn 4k7 (for example) trim pot between the output and the IN of the OP in series with the existing resistor. This would allow me to tweak it.

Avoid multi-turn pots. Use a small-value single turn pot in series with a much larger value (or combination of values) resistor(s).

 

so the feedback resistor would be post the diode? Would this then increase the output because the opamp thinks its output is 0.7V (drop across the diode) lower than it actually is?

Exactly. Don't forget you'll need an additional pull-down resistor also attached to the cathode of the diode to ground. Start with 1K to 10K and play around till you get a decent low Vo (with 0V Vin) without too much load on the amp.

 

thanks guys