isn't one of the perks of using RTDs having 4 leads of which one is a ground? I may have misunderstood at some point but I have been assuming the roaster case/ground potential should not affect the RTD in any way.

Depending on where the bean mass is in your roaster, you should be able to make slight adjustments to depth of the probe without affecting your roasts. In fact I'd make sure the tip of the problem is 1cm or so away from any metal vanes, drum surfaces, etc anyway to avoid skewed readings. I'd do that before I bent the probe which can surely weaken it leading to premature failure.

It is possible the probes themselves could be grounding but not likely and you can test this with a continuity tester/multimeter where one lead is on the probe sheath and then you test continuity against your roaster, computer, etc. to see if there is a short.

Another culprit if it's a grounding issue are ground loops through your computer and USB devices. This can lead to intermittent noise if the ground floats back and forth on your USB port. Some folks try operating their laptop not plugged in during a roasting session to mitigate this, but then others say it's needed to avoid them. Do you have other devices, monitors, going into your computer? Are some of these devices plugged into different outlets? I'd recommend removing any devices you can and make sure everything is plugged into a high quality isolated surge protector going into the same wall outlet. And to make sure the wall outlet is grounded to earth (which sometimes is regretfully not the case).

P.S. What diameter probe are you using? 3mm?
P.P.S. I may be leading your down a rabbit hole, but these are some ideas to try.

Quick summary, it is a 3.2 mm diameter probe that is 11 cm long. It has worked for about a year in its current configuration with no problems. It will accurately sense temperature up to charge temp of about 410F but soon after beans are added to drum, it wigs out. I'm getting message in phidgets control panel that saturation is detected on that probe. I'm thinking it's internally damaged causing intermittent failures. I'm running a USB cable from the Phidgets hub to a desktop. My monitor is a 30" flat screen tv all are plugged into same grounded outlet with surge protection. I've ordered a new probe and will install it not fully interested in drum and make sure there is adequate clearance fro rotating parts and avoid hand forming of the new one. It should arrive next week.

Sounds like a good plan to replace the probe. Report back if you're still having issues. Definitely try to reposition the probe so as not to have to bend it. I backed mine off about 2cm from the stock position and didn't note any differences in my roaster performance. Of course this is dependent on your charge size and roaster, but worth a test.

Do you have any further Phidgets connected to the HUB0000, like an ambient HUM1000 or PRE1000, or any unused module?

MaKoMo - for me I had connected to my Huky

Into HUB
2 X TMP1200 (for BT and ET)
- ET no longer connected
HUM1000 - for temp and humidity
REL1100 (for LED's) - different colors depending on been stage
1136 pressure sensor - monitor build up of charff
TMP1101 ( cyclone temp - so I don't melt the plastic)

I use a laptop powered and have touch screen monitor

I would start having crazy spikes as ROR reached its peak and then so messy it was useless
I played around with
- removing all the different phidgets, removing the touch screen monitor. Really felt like I was chasing my tail as it was just so random. Sometimes I would think I had solved it and then next run, same issue. Once I disconnected the ET RTD the problem went away for good. That was over a year ago

Using the latest Phidget driver and updated firmware version (using the Phidget Control Panel of the latest Phidget driver) for all modules?

on the thermopar

Using the latest Phidget driver and updated firmware version (using the Phidget Control Panel of the latest Phidget driver) for all modules?

MaKoMo
At the time I believe so. I went back in and updated the firmware for phidget. The rehooked up the ET RTD

My Et read but BT did not. Disconnected ET, BT still not reading after computer reboot. Checked Phidget control panel it shows a saturation error

rebooted again, BT is working. I stopped. This was part of the issue before as well that ET and BT would randomly stop

I see a thread on this but looks like the solution was for a PT100. I followed the wiring shown in the image from Phidget for the PT1000 (TMP4109_0 - PT1000 4-Wire RTD 11cm)

What to do about RTD Phidget (TMP 1200) giving saturation detected error

Chuck; refresh my old noggin. Did this start when you added that touch screen and did taking it out cause the problem to go away? Or am I just mis-remembering?