I took the liberty to remove the inline connectors and directly solderend the wires in an attempt to proof/solve the thermal runaway issue. To my surprise the thermal run away issue was not solved, meaning the connectors are clearly not the root cause, meaning I jumped the conclusion entirely. Next I ordered a few new Revo heaters form E3D to compare sensor and heater resistance values between my used and malfunctioning Revo and brand new ones, but no significant difference to be noted. Certainly nothing that could explain the thermal run away. Time to dig a little deeper into the problem. Removed the malfunctioning Revo heater from the printer and started to disassemble it and found some interesting stuff that could explain the issue. Once all the potting material was removed it became clear that the NTC sensor which by design is placed in a small cavity at the bottom of the heaters metal body was largely sitting in an air pocket (thermal isolator) while per my understanding it should be fully encapsulated in the potting material. This air pocket theory is also confirmed after closely examining the the sensor surface where only a fraction of the potting material (ceramics) seems to have touched the sensor judged by the sensor surface contamination. Then it became clear why a virgin heater/sensor at first behaves correctly but after some heating cycles start running away. As long as there is a physical connection between the sensor and the ceramic potting of the heater body the regulation loop is working perfect. However due to thermal expansion that physical connection can be lost/broken causing the start of the thermal runaway. So the solution to the problem here is to make sure that when potting the heater during production the NTC sensor is fully encapsulated with potting/gluing material and that this material is resilient to thermal expansion. In fact the main root cause is to be found during the production when poring the potting material inside the heater body. I believe an air bubbel is trapped in the sensor cavity and that causes the poor thermal behaviour.
I can confirm the same problems with at least 2 heater cores. Both worked for about 4 weeks of printing before starting to throw thermal runaway errors on my Prusa i3 MK3s. One was delivered with the Revo Six the other was already a replacement heater core because the wires of the original core broke due to movement. I wrote my findings to support and am curious if those production problems have been resolved, but for now I think I have to retrofit my printers with the original hot end to be able to print reliably.
I started having the same problem today with the heather core in a Revo Hemera XS on my Sidewinder X2 that has been working fine for a couple of months. And indeed it is hit and miss, I have a Revo Mini on my Prusa Mini+ since July 2022 and no problems with it
I have the same problem, its my second now and probably not the last. Have one at work that also has been replaced due to this problem. Could we have some official news from E3d about this problem!?!
