My best guess would be a earth problem in the A channel. So when you connect both probes to the source then A benefits from a good ground connection from B.
One things you could try to see if this is connect is to connect A to your source but also connect just the ground clip of B to the source ground as well to see if that cleans it up.
If it does then it means there is a continuity problem in the earth somewhere along A. If you swapped the probes and the problem stayed with channel A then it is not in the probe itself but in the Quad either with the connector itself or the grounding of the connector inside the quad. You could try tracing that connection through with a mutli-meter or you could just RMA it.
Thanks, thats exactly what the problem was. Faulty ground connection on channel A in the Quad itself. Will try to trace it but if I can’t find it RMA it will have to be.
Probably accidentally connecting channel A ground to a different voltage than channel B. This gives a short-circuit, and apparently the Quad has quite thin traces at that important point.
However, it shouldn’t break anything else so should be an easy fix. If you want to make it extra robust, solder a wire directly between the MCX connectors.