EMI affecting IR probe?

Surgikill

Hey all,

I just re-did the wiring harness on my corexy. I had some high grade connectors lying around, and I wanted to make everything modular so I can just pop the whole extruder assembly off. This meant that I wired the IR sensor and filament sensor together. I used CAT 5 cable to wire them together. As you might know, these use twisted pairs. I took a twisted pair and used it for power and ground on the IR probe, and the same for the filament sensor (laser filament sensor). Then I took a third twisted pair, and used it as the data line for both. One wire goes to the IR probe, one wire goes to the filament sensor. However, I am having issues with the IR probe now. It is showing a reading of around 520 when not triggered. I was able to get it to read 0 and homed the machine, and then I started manually pulling filament through the laser filament sensor, and I watched the value on the probe readout slowly creep up.

Is there a way I can fix this? Maybe put a choke on one of the wire bundles, or a pulldown resistor?

Surgikill

So I still have the old cable for the IR probe laying around. I plugged it in and the IR probe works as intended now. So there has got to be something wrong in the cable I made.

I measure the resistance on all the wires for the IR probe, and they are 0.3 ohms or less. I plugged the new cable into the duet and measured a voltage of ~3.16V between the data pin and ground, it showed a reading on the panel due of around P960. I plugged the old cable into the duet and measured a voltage of ~0.660V between the data pin and ground, it showed a reading on the panel due of around P170.

I'm not sure what could be causing this. I pinned the connectors correctly. The older cable is longer and has portions of cat5 on it. Any idea? I'd really like to use the new cable.

What's really confusing is that with the old wire, the IR probe works exactly as intended, showing 0, then only showing 535/536, then showing 1000. With the new wire the value bounces all over the place.

dc42

There is a lot of capacitance between the two wires in a twisted pair. So it is possible that the capacitance between the filament monitor output and the IR sensor output is generating spurious signals on the IR output wire. However, I think it much more likely that there is a bad crimp or other problem in the cable.

Surgikill

@dc42 I still have the issue with it even with the filament sensor unplugged on both ends. I checked the crimps on both ends and found that one pin wasn't seated all the way in the connector. I seated it and I am still having problems. I'm going to check solder connections on the 6 pin connector next, but I measured resistance on each line and all seemed fine.

I'm still puzzled as to why the duet reads a bunch of random garbage with the new wire. I thought I should expect a 0 value, a 535/536 value, and a 1000 value ONLY.

dc42

@Surgikill said in EMI affecting IR probe?:

I'm still puzzled as to why the duet reads a bunch of random garbage with the new wire. I thought I should expect a 0 value, a 535/536 value, and a 1000 value ONLY.

If the output wire of the IR sensor is not connected to the Z probe IN pin, then you will get garbage readings because the input to the ADC is floating.

Surgikill

@dc42 said in EMI affecting IR probe?:

@Surgikill said in EMI affecting IR probe?:

I'm still puzzled as to why the duet reads a bunch of random garbage with the new wire. I thought I should expect a 0 value, a 535/536 value, and a 1000 value ONLY.

If the output wire of the IR sensor is not connected to the Z probe IN pin, then you will get garbage readings because the input to the ADC is floating.

It is wired correctly. The output pin of the IR probe is connected to the Z probe in pin. I just triple checked with a multimeter, the wiring diagram, and my old wiring harness. The multimeter was reading 0.1 ohms for total resistance as well. I can try to measure capacitance if you think that might be causing the issue with the twisted pair.

Surgikill

I fixed it. I think there were some cold solder joints on the 6 pin connectors I made up. I reflowed them and it all works now.