Multichannel analog board with saturating opamps

Thread Starter

sim10

Joined Oct 6, 2022
24
Hi everyone, i have an analogic board of about 20 channels each of which is made by a circuit with an opamp. Each circuit is positioned in the PCB board forming a matrix, imagine that each circuit represents a node of the matrix, I have observed that, given the same Vin to all channels, the gain of the opamps more distant from the input connector gets higher hence the output saturates earlier than the ones nearer the connector. Does someone have a general idea? Can it be caused by some common-mode tension?
 

ericgibbs

Joined Jan 29, 2010
18,869
hi sim,
Welcome to AAC.
Do you have a photo of the layout you post, it sounds like the ground lines are the problem.
E
 

Ian0

Joined Aug 7, 2020
9,844
1. The circuit appears to be a transimpedance amplifier, with a current input from the socket being turned into a voltage. What controls the inout current?
2. What is your supply voltage, and what voltage is on the non-inverting inputs of the op-amps?
 

Thread Starter

sim10

Joined Oct 6, 2022
24
1. The circuit appears to be a transimpedance amplifier, with a current input from the socket being turned into a voltage. What controls the inout current?
2. What is your supply voltage, and what voltage is on the non-inverting inputs of the op-amps?
the supply voltage is +- 15V. the signal is given with an external signal generator.
 

Ian0

Joined Aug 7, 2020
9,844
the supply voltage is +- 15V. the signal is given with an external signal generator.
What Voltage is on the non-inverting inputs of the op-amps?

Does your signal generator have a current output or a voltage output?
If you drive a transimpedance amplifier with a voltage, then the output WILL be saturated.
 

Thread Starter

sim10

Joined Oct 6, 2022
24
What Voltage is on the non-inverting inputs of the op-amps?

Does your signal generator have a current output or a voltage output?
If you drive a transimpedance amplifier with a voltage, then the output WILL be saturated.
to the non inverting input I give a sinusoidal signal of amplitude 1v and frequency 30kHz. This input is given to all the channels but the ones further saturate before, this implies that I have to reduce the amplitude of the signal in order to have all the channels not saturating. However, I don't use some of them at the best
 

Ian0

Joined Aug 7, 2020
9,844
to the non inverting input I give a sinusoidal signal of amplitude 1v and frequency 30kHz. This input is given to all the channels but the ones further saturate before, this implies that I have to reduce the amplitude of the signal in order to have all the channels not saturating. However, I don't use some of them at the best
Whatever is happening on the inverting inputs is causing the problem.
 

Thread Starter

sim10

Joined Oct 6, 2022
24
Whatever is happening on the inverting inputs is causing the problem.
on the inverting input I have a capacitance that's what I want to measure with each channel. i don't get why more distant channel behave differently with respect to others, is there some way to avoid this?
 

Ian0

Joined Aug 7, 2020
9,844
on the inverting input I have a capacitance that's what I want to measure with each channel. i don't get why more distant channel behave differently with respect to others, is there some way to avoid this?
How big is the capacitance you wish to measure? And how much capacitance is there between the pcb tracks and ground?
 

Thread Starter

sim10

Joined Oct 6, 2022
24
How big is the capacitance you wish to measure? And how much capacitance is there between the pcb tracks and ground?
the capacitance I wish to measure is 10 nF, for what regards the capacitance between the pcb's tracks and ground I cannot tell because my measurement instrument doesn't measure anything. The minimum value that the instrument can measure is 1nF
 

Thread Starter

sim10

Joined Oct 6, 2022
24
the capacitance I wish to measure is 10 nF, for what regards the capacitance between the pcb's tracks and ground I cannot tell because my measurement instrument doesn't measure anything. The minimum value that the instrument can measure is 1nF
Could it be that something affects the impedance of the circuit? Like parasitics capacitances?
 

Ian0

Joined Aug 7, 2020
9,844
What value of resistance and capacitance do you have between outputs and inverting inputs of the op-amps?
[edit] and what is part number of the op-amp. Have you checked what its open loop gain and phase shift is at 30kHz.
 

ericgibbs

Joined Jan 29, 2010
18,869
hi,
Perhaps a circuit diagram, with component values, for just one element of the matrix would help us to see the problem.
E
 

Thread Starter

sim10

Joined Oct 6, 2022
24
What value of resistance and capacitance do you have between outputs and inverting inputs of the op-amps?
[edit] and what is part number of the op-amp. Have you checked what its open loop gain and phase shift is at 30kHz.
the gain allows me to go from 650mVpeak-peak to 26 V, so it's about 40
1665063736575.png
 
Top