Hey guys, I built a circuit using a STM32F0 microcontroller, and using its internal DAC, I am outputting an analog signal that is 0 volts to 3.3 volts. This goes into an Op-Amp, where it is amplified up to a 0 volts to 15 volt signal.

The problem I am experiencing is I am only getting up to 13.3 volt output and I can't figure out why.

Here is my circuit:

These are the part #'s I am using:
Op-Amp: OPA171AIDR
Darlington-transistor: MJD44E3T4G
Rectifier: ES1B-E3/5AT (not shown in schematic, but the input power goes through this rectifier first, then into the 15V rail).
R5: This is actually a 2.8K resistor (I switched it to 2.8K because I had another 2.8K resistor in the design to reduce component count).
D5, D6: BAV101-GS18
D4 (16V zener): SMAZ16-13-F

I measured the power input voltage to be 15.2 volts, and after the rectifier it was 14.6 volts. So in this case, I would expect the darlington transistor to at least be outputting up to 14.6 volts, but like mentioned above I'm only getting 13.3 volts.

And speaking of this rectifier... is there a rectifier I could use here that won't drop the voltage as much? Do I just search for a rectifier with a lower forward voltage?

I took a few more voltage measurements:

• Input voltage of pin 7 at OPA171 is 14.6 volts.
• 14.6 volts on both sides of R6.
• If I reduce the input voltage going into the op-amp, the output seems to be accurate. It seems to be clipping the top at the 13.3 volts even if I continue to increase the input voltage going into the op-amp.

Any ideas what may be going on here? Any help or advice is greatly appreciated, thanks!

 

You will lose 1-2V because it is a Darlington transistor.

 

You will lose 1-2V because it is a Darlington transistor.
View attachment 174006

Thanks for the reply! Ok this is making sense. So ideally, I need the voltage going into the collector to be 2 volts higher than the highest expected output... e.g. 17 volts if I want to generate an output all the way up to 15 volts?

And with that said, my current rectifier has a forward voltage of 0.9 volts. So if I swap it with this (0.3 to 0.6 forward voltage depending on current draw):
SL310A-TP

Then this would get me up to 13.6v to 13.9v, keeping everything else exactly the same?

 

Something like that. R4 and the Zener might have a significant effect as well.

 

I would eliminate the rectifier (not shown) R2 and the zener. That will get you 15.2 volts at pin 7 and maximum output then is 15.2 volts. Why do you need the transistor? What is the load at the analog output?
SG

 

I would eliminate the rectifier (not shown) R2 and the zener. That will get you 15.2 volts at pin 7 and maximum output then is 15.2 volts. Why do you need the transistor? What is the load at the analog output?
SG

I was considering removing the rectifier and just putting a jumper/short in its place. I couldn't tell if I should or not. I have it there for polarity protection.... but a power adapter with a barrel plug is used, so I don't see an accidental reverse polarity situation occurring? Is there any other reason to leave it?

There is a lot of external interference present, and the zener suppresses transients from damaging the OPA171.

The output needs to be able to support up to 1 amp max, that is why I was using the darlington transistor.

 

There is a lot of external interference present, and the zener suppresses transients from damaging the OPA171.

The output needs to be able to support up to 1 amp max, that is why I was using the darlington transistor.

In that case you can leave the zener and R2 that will still get you 15.2 volts at R6.
OK on the transistor.
Can you raise the supply voltage?
SG

 

In that case you can leave the zener and R2 that will still get you 15.2 volts at R6.
OK on the transistor.
Can you raise the supply voltage?
SG

Not easily... although it sounds like that's probably the best solution. So I'm using a 15VDC power adapter that inexpensively. There really isn't any power adapters that aren't obscure up to about 24VDC. But the Op-amp has a maximum DC input of 18 volts, and that's with no headroom. So I would need to make a step down voltage regulator to 17 volts. So that's the reason why I was trying to get the 15V regulator to work initially since its common and inexpensive.

 

But the Op-amp has a maximum DC input of 18 volts, and that's with no headroom.

Using a single ended supply the max voltage is actually 36 volts.
What about using a laptop supply/charger, they are usually around 19 vdc, just a thought.
SG

 

Would there be any downside in me removing the power input rectifier? I'm not worried about polarity being accidentally swapped. Could there be an issue with the analog output being tied to the 15V with the diode?

 

Would there be any downside in me removing the power input rectifier?

No. That was what I was suggesting earlier. According to your readings that should get you 15.2 volts at pin7
I don't see any issue leaving the diodes in the analog out .
SG

 

Thanks again for all of the help.
So here is a question... is there a better transistor that I could use here? Something that has a smaller Collector-Emitter saturation voltage? Or does this definitely need to be a darlington, and I probably won't find anything with less voltage drop?

For instance... unless I am reading it incorrectly, this one seems to have a bit less Vce (saturation) than the current one I'm using, so this would yield a bit more voltage output?
https://www.digikey.com/product-det.../NJVNJD35N04T4G/NJVNJD35N04T4GOSCT-ND/9087318

 

Something that has a smaller Collector-Emitter saturation voltage?

Probably not. The most you can get out of the chip is 15.2 volts with a 15.2 volt supply. Besides the Vce saturation there is also the base-emitter voltage drop which is usually about 1.2 volts for a darlington. The most you will probably get out is 14 volts. The only way to get higher is to increase the supply voltage.
SG

 

Your output transistor is in the emitter follower configuration and cannot saturate, the worst configuration if you heed the highest possible output voltage. If you change to a PNP transistor things improve, but this changes the input stage of the circuit and can have stability problems depending on both the design and the construction.

Note that while this circuit can source over 1 A, it cannot sink it. If there is any energy storage in the load (capacitive or inductive components), it will affect the fidelity of the output waveform.

ak

 

You will get much better help if you describe what the load is, and the project in general.

 

You can reduce the voltage drop to one Vbe voltage if you use a Sziklai pair in place of the Darlington (below):