I'm having some troubles getting rid of an annoying offset, therefore I need your help!
The power rails are +-12V and I am feeding 0 to 500uA to an LM13700 OTA (U5.2) in this configuration:
***edit: I wrote 0 to 500mA instead of 500uA


Even when there is no input signal at the non inverting input, i get an offset at the output (pin 12 of U5.2) that swings between 0 and around -400mv depending on the amount of the Iabc current.
And when the signal passes through the non inverting input, it get summed to it and everything is shifted down.
This is quite annoying for my application and I’m trying to get rid of it but I’m having some issues understanding why this is happening.
Do you know why this offset is there?

The following doesn't have offset at the output but is not a solution since it doesn't remove it at the source, but it just cut it out at the price of an extra OPAMP and it doesn't look too nice to me.

Any help appreciated!

 

if i set the trimmer (R100) so that 11.61V pass through R89, the offset disappear but I can find it again when I amplify the signal passing through the non inverting input.
I discovered that i can completely get rid of it if I remove R107 and so there is no diode bias current but in this way i will need to reduce the signal at the non inverting input in order to avoid distorsion. This will also make the circuit noisier.
I am indeed omitting something.

 

The datasheet shows that the absolute maximum allowed amp bias current at pin 16 is only 2mA and you are feeding up to 500mA! That might have damaged the LM13700.
Your LM13700 has no negative feedback which would cancel the base-emitter voltages of the buffer transistors.

 

hi 1919,
This is what LTSpice shows, check my circuit for accuracy.
No input signals applied.
E

 

excuse me all. I'm talking about 500uA, not 500mA.
@ericgibbs what is the Iabc current you are feeding? is it a constant 0uA?
@Audioguru again do you have any circuit i can check to understand what precisely you refer at?

 

hi 1919,
I am not inputting any voltage or current signals, tell me what you are using, and I will add them to the simulation.
E

 

with 0 uA, i get the same result as the simulation.
When the current is about 500 uA, then the output (tp12 in the simulation) shows around -400mV.

 

Hi,
This what 500uA gives at tp12, note the 625mVppk is set to 0v.
Make sure my circuit is correct.
E

 

it is correct and it gives an offset as well. Different in voltage from what i encounter but looks like the same issue.
I don't understand why it happens.
I discovered that, even without signal applied, the non inverting input sits on top of an offset (in my case is around 100mV). this offset appears when I use the 12k to feed 1mA at diode bias input (pin 15).
Beside the fact that i noticed when it appears, i don't know why it appears.
I tried to change R89 to 100k and set the trimmer resistor R100 so that I also shift the inverting input to the same 100mV.
In this case the offset at the output cancels out.

 

@Audioguru again do you have any circuit i can check to understand what precisely you refer at?

Negative feedback is shown on circuits in the datasheet: