# Distortion in a negative channel of the differential amplifier THS4505

I designed a single-ended to differential signal amplifier using a tool provided by Texas Instruments:

The source signal passes through a voltage follower (OPA810IDBVR), so the output resistance should basically be negligible. My problem is that there's an asymmetry, that is clearly visible in the oscillogram below (CH1 & CH2 = outputs, CH3 = input). The output signal is sent through the 1 meter long cable and is terminated with a LPF ($$\33\Omega/22 pF\$$ RC filter) before it's supplied to an ADC differential input.

Can you suggest me any possible cause of this problem? I also attach the application schematic (the voltage follower operational amplifier is outdated here):

Update 05/30/2024

I found the source of the problem - it was caused by exceeding voltage reference's max output current. After amplifying the current, this part of a problem was solved.

For anyone interested, there's the waveform of overloaded voltage reference's output in the photo (it should be stable $$\1.50 V\$$). :

Unfortunately, my problem doesn't end here. Although the differential amplifier output is not distorted anymore, the gain factor of both channels is asymmetric. Each output channel $$\V_{PP}\$$ should be equal to $$\1.00 V\$$. I measured the equivalent resistances of both feedbacks and it's perfectly symmetric. However, I noticed that when I change the input signal from $$\1.5 V_{PP} + 0.75 V_{DC}\$$ to $$\1 V_{PP} + 1 V_{DC}\$$, it starts to look fine and the gain becomes exactly $$\1.33 \dfrac{V}{V}\$$ per channel. If anyone solved similar problem before, I'll be glad to read about your solutions, while I'm considering possible parameters' modification.