Signal distortion by op-amp.

Thread Starter

yuesu jin

Joined Aug 22, 2019
20
Hi, all,
Currently I am using a LM675 amplifier to amplify DC-20Hz small signal. The reason why I use LM675 is that I need a large current power amplifier as drive circuit. The schematic diagram is the standard application on the datasheet, and the signal input is a 10Hz, 100mV sin wave. Power supply is +12v DC power. When I increase the frequency to 30 Hz, the signal is pretty good, but when I lower the frequency below 20 Hz, signal is heavily distorted. I see a current source using LM675 on the data sheet, I think this amplifier could work in very low frequency even in DC regime, couldn't it? Thanks a lot!!!!
question.png
upload_2019-9-16_20-21-1.png
 

danadak

Joined Mar 10, 2018
4,057
Your .1 uF bypasses at 20 Hz or ~ 80 K ohms, basically not even present.
The PSRR is ~ 70 db. Your gaining up by 20, 26 db.

The bypasses should be raised significantly in value for starters. Look for any
common mode stuff input referred and take these parameters into account.
Like ground loops.....

The spikes are 10 Hz, a clue ? You are driving a 4 ohm load, I cannot see what
waveform vertical scaling is, driving too much current ? L load ?

Power is + 12, what is - Vee ? The input CM range is 5V min off either rail at +/-
25 V supply rails. You input in CM range ?


Regards, Dana.
 
Last edited:

Thread Starter

yuesu jin

Joined Aug 22, 2019
20
Your .1 uF bypasses at 20 Hz or ~ 80 K ohms, basically not even present.
The PSRR is ~ 70 db. Your gaining up by 20, 26 db.

The bypasses should be raised significantly in value for starters. Look for any
common mode stuff input referred and take these parameters into account.
Like ground loops.....

The spikes are 10 Hz, a clue ? You are driving a 4 ohm load, I cannot see what
waveform vertical scaling is, driving too much current ? L load ?

Power is + 12, what is - Vee ? The input CM range is 5V min off either rail at +/-
25 V supply rails. You input in CM range ?


Regards, Dana.
Good evening, Dana!
Thanks for your reply!! I am an amateur in electronic parts, what I am trying is to make a driving circuit for our electromagnets. Vcc=10v, -Vee=-2.8V. The load here is an oscilloscope. The common mode voltage is -50mv-50mv, non inverting input is -100mv-100mv, inverting input is grounded. Could you give me more information on bypass? What's the usage of bypass capacitor? Where can I find more information about these stuff?
Best regards,
Yuesu
upload_2019-9-16_21-6-55.png
 

Audioguru

Joined Dec 20, 2007
11,248
The datasheet says the minimum supply totals 16V and is +8V and -8V. Your negative supply voltage of -2.8V is much less than -8V.
 

Thread Starter

yuesu jin

Joined Aug 22, 2019
20
lue ? You are driving a 4 ohm load, I cannot see what
waveform vertical scaling is, driving too much current ? L load ?
The datasheet says the minimum supply totals 16V and is +8V and -8V. Your negative supply voltage of -2.8V is much less than -8V.
Hello, Audioguru,
Thanks for the reply. I input a 30Hz signal, the result is pretty good. It seems like weak relationship with input voltage.
upload_2019-9-16_22-41-54.png
 

danadak

Joined Mar 10, 2018
4,057
The datasheet says the minimum supply totals 16V and is +8V and -8V. Your negative supply voltage of -2.8V is much less than -8V.
But datasheet shows several curves down to ~ 5V, which cannot possibly meet
their CM specs. Datasheet needs some love.


Regards, Dana.
 

Thread Starter

yuesu jin

Joined Aug 22, 2019
20
But datasheet shows several curves down to ~ 5V, which cannot possibly meet
their CM specs. Datasheet needs some love.


Regards, Dana.
Hello, Dana,
You are right, the reason why I choose 12V voltage is that on the datasheet the supply voltage could work around ~10V. I cannot understand why the amp could not work in low frequency regime, because op-amp could work in DC range, it's strange that the amplifier will distort a low frequency signal very seriously.
 

Audioguru

Joined Dec 20, 2007
11,248
The datasheet does not show a 10V supply. It says that +/- 8V (16V total) is the minimum then your -2.8V is much too low.
You have distortion at low frequencies because your power supply voltages (especially the negative supply) are dropping with the signal current because then have a high impedance at low frequencies (not enough filtering capacitance).
 

Thread Starter

yuesu jin

Joined Aug 22, 2019
20
The datasheet does not show a 10V supply. It says that +/- 8V (16V total) is the minimum then your -2.8V is much too low.
You have distortion at low frequencies because your power supply voltages (especially the negative supply) are dropping with the signal current because then have a high impedance at low frequencies (not enough filtering capacitance).
Hello, Guru, I know one possible problem, the power supply might have bad quality that sends an AC component into the amplifier, tomorrow I will use another source for the test, I will update my result tomorrow.
 

Audioguru

Joined Dec 20, 2007
11,248
In your other thread, the LM675 opamp is wired to use a dual polarity supply but it has NO negative supply so it rectifies the signal producing severe distortion. I showed how to bias its input at half the supply voltage and use coupling capacitors allowing it to work perfectly without a negative supply.
 

Thread Starter

yuesu jin

Joined Aug 22, 2019
20
In your other thread, the LM675 opamp is wired to use a dual polarity supply but it has NO negative supply so it rectifies the signal producing severe distortion. I showed how to bias its input at half the supply voltage and use coupling capacitors allowing it to work perfectly without a negative supply.
Thanks a lot!! My voltage source arrived today, I will have a try on that! I appreciate your help!!
 

KeithWalker

Joined Jul 10, 2017
3,063
Hello, Guru, I know one possible problem, the power supply might have bad quality that sends an AC component into the amplifier, tomorrow I will use another source for the test, I will update my result tomorrow.
That should improve the waveform. The display showing the 10 HZ is being badly modulated by 60 Hz ripple. Check that the circuit common and signal common lines are all connected to to the same circuit reference point to eliminate noise pickup.
 

Thread Starter

yuesu jin

Joined Aug 22, 2019
20
In your other thread, the LM675 opamp is wired to use a dual polarity supply but it has NO negative supply so it rectifies the signal producing severe distortion. I showed how to bias its input at half the supply voltage and use coupling capacitors allowing it to work perfectly without a negative supply.
Hi, Guru,
The problem has been solved, thank you a lot!! You really contribute to this project, thanks !
upload_2019-9-23_22-37-24.png
 
Top