Building and Testing a STM Circuit - Issues with Gain

[tigigi]

Hi,

I'm building a STM, and this is the circuit I use :

http://www.geocities.com/spm_stm/Simple_STM_Rev_C.PDF


I'm right now testing the circuit I build. I build a small part (U3A), and then test it, and build another small part (U3B) and test, then connect the two and test it again. This is the way I build it and test it. However, I found that I can't get the gain to be 1.

For U3A and U3B, I get around 0.08/2.25 gain for each part. While I conncet U3A and U3B, the gain I get is even smaller. I tried to change R2 to be bigger (it original is R2=10k, R1=10k), like 20k, 100k, 10M, the gain becomes a bit bigger but is still much smaller than 1. I checked all the frequencies I could get from 50 to 5M, and can never get gain = 1.

I connect the ground to the ground of oscilloscope and signal generator. Apply voltage to 4 and 11 by power supply.

I didn't have any experience about this, and don't know what's wrong with this. I would like to know what would be the possible way to solve this.

Thanks for your help. I appreciate it.

 

[NoTime]

The circuit computes to the gain stated.
Did you verify the DC voltages on the op-amp pins are correct?

Did you put the bypass capacitors close to the power pins of the op-amp?
It may be oscillating.
The frequency likely to be higher than the scope can detect.
You can check with an RF probe.

Try hooking scope to where signal generator is connected to circuit with scope probe set to the high impedance setting.
The display amplitude should be the same as amplitude with scope probe moved to op-amp output.
This should eliminate scope calibration and signal generator loading problems as well as some potential ground loop issues.

 

[berkeman]

When debugging opamp circuits, I like to check that the "virtual ground" is working correctly. Put one scope probe on the - input of the first opamp U3a, and verify that even when you have a 1Vpp input signal, the - input stays at ground potential (to match the + input via the feedback path). If it does not match, then either the IC is bad, or there is a wiring mistake.

Are you building this circuit by soldering parts onto a pre-made PCB, or hand-wiring it in some way? It is generally good practice to tie off the other devices in the package (U3c and U3d) while working with some of the devices in the package, but not tieing them off would increase IC current consumption and noise in this case, and would not change the gain of U3a and U3b in-circuit.

BTW, there is a circuit design error at u2d. Quiz Question -- what is the error, and how can it be fixed? Hint -- we covered this error in the thread here in the EE forum on "Bad Circuits".


https://www.physicsforums.com/showthread.php?t=178516


.

 

[tigigi]

Thank you for replying.

I use the power supply instead of the battery since that's what I have right now. So actually I'm using AC voltage now. The power supply I have is one way, not two way, so I tune it to be 9 volt, connect ground to 11(-v), connect + to 4 (+v). I also tried to change the voltage from 1 volt to 15 volt, but not really helps.

I just connect everything on a bread board right now, so I could remove and reconnect anything quite easily.

I looked the bad circuit page, but found that I'm clueless about what's the connection between that and my circuit. Actually, my EE background is just like freshman or high school level, and don't have any concept about op-amp. But since I need to make the STM work by the end of the year, I checked out sth to just understand what my circuit means. Thank you for providing the reference book, and I'll check it out in the library.

BTW, you know how the STM circuit works first, so you figure out there's an error in the circuit ? Is it right ?

Thanks again.

 

[berkeman]

Thank you for replying.

I use the power supply instead of the battery since that's what I have right now. So actually I'm using AC voltage now. The power supply I have is one way, not two way, so I tune it to be 9 volt, connect ground to 11(-v), connect + to 4 (+v). I also tried to change the voltage from 1 volt to 15 volt, but not really helps.

I just connect everything on a bread board right now, so I could remove and reconnect anything quite easily.

The power supply that you are using could be part of the problem. You need to use split supplies that are referenced to ground (not a floating 2-port battery or 2-port floating power supply), and they need to be wide enough to accommodate the TL074 input and output voltage range. Check out the TL074 datasheet to learn more about the requirements of the power supply.

I looked the bad circuit page, but found that I'm clueless about what's the connection between that and my circuit. Actually, my EE background is just like freshman or high school level, and don't have any concept about op-amp. But since I need to make the STM work by the end of the year, I checked out sth to just understand what my circuit means. Thank you for providing the reference book, and I'll check it out in the library.

BTW, you know how the STM circuit works first, so you figure out there's an error in the circuit ? Is it right ?

Thanks again.

I don't know much about how STM circuits work, but the error that I flagged is a common one. Check out post #39 in that Bad Circuits thread to see the start of the circuit discussion that gets to the error that I flagged in the STM circuit diagram.

 

[tigigi]

The power supply that you are using could be part of the problem. You need to use split supplies that are referenced to ground (not a floating 2-port battery or 2-port floating power supply), and they need to be wide enough to accommodate the TL074 input and output voltage range. Check out the TL074 datasheet to learn more about the requirements of the power supply.

I checked sth about 2 port battery, but not that sure, so I'd like to confirm that if the battery we use in daily life, like in alarm clock, etc, 2 port batteries ?? I double checked the power supply I use is " hp, 115V 2A, 230V 1AT, 6284A DC POWER SUPPLY, hewlett packard " It has +, - and ground, but I'm not sure if that means it has reference ground.

I don't know much about how STM circuits work, but the error that I flagged is a common one. Check out post #39 in that Bad Circuits thread to see the start of the circuit discussion that gets to the error that I flagged in the STM circuit diagram.

I'll go check that out. Thank you so much !

 

[tigigi]

I checked sth about 2 port battery, but not that sure, so I'd like to confirm that if the battery we use in daily life, like in alarm clock, etc, 2 port batteries ?? I double checked the power supply I use is " hp, 115V 2A, 230V 1AT, 6284A DC POWER SUPPLY, hewlett packard " It has +, - and ground, but I'm not sure if that means it has reference ground.

 

[berkeman]

I checked sth about 2 port battery, but not that sure, so I'd like to confirm that if the battery we use in daily life, like in alarm clock, etc, 2 port batteries ?? I double checked the power supply I use is " hp, 115V 2A, 230V 1AT, 6284A DC POWER SUPPLY, hewlett packard " It has +, - and ground, but I'm not sure if that means it has reference ground.

No the HP 6284A is a single power supply: http://www.4testequipment.com/HP_6284A.htm

If you want to make split supplies with it, you need to use two resistors (like 1kOhm) between the supplies and connect the center point to the ground terminal on the supply. Be sure to add bypass capacitors from both power supply rails to ground.

 

[tigigi]

No the HP 6284A is a single power supply: http://www.4testequipment.com/HP_6284A.htm

If you want to make split supplies with it, you need to use two resistors (like 1kOhm) between the supplies and connect the center point to the ground terminal on the supply. Be sure to add bypass capacitors from both power supply rails to ground.

Thank you, but I don't really understand. The supply has one +, one -, one ground. Do you mean connecting 1 resistor to ground and +, the other to ground and - ? and connect 2 capacitors, 1 to ground and +, and the other to ground and - ? Thanks again !

 

[tigigi]

I checked the datasheet for TL074 that :
supply volt V : +-18 V
input volt Vin : +-15 V
differential input volt Vid : +-30 V
power dissipation Ptot : 680 mW
output short-circuit duration : infinite

 

[berkeman]

Yes, what you are doing is biasing the + and - outputs of the power supply, so that they are symmetric about ground. Connect a 1k resistor from + to ground, an a 1k resistor from - to ground. Then put capacitors across each of the resistors (use something like 22uF polar capacitors of a high enough voltage rating, and be careful to check the polarity as you put them on. If the polarity of the caps is reversed, they could pop rather violently.