Hot End Heater wiring crimping techniques and Thermal Runaway  

Hot End Heater wiring crimping techniques and Thermal Runaway

So I've been happily printing away on my MK2.5S with a Mosquito and have been having a little trouble with a 3 hour print....

First, the print quality deteriorated and hotend temperature became slightly erratic.  ... then "Thermal Runaway" 2 hours into a print. 

Oh, no not again! A look with a FLIR and yes, the heater wiring is extremely hot about 25mm from the heater cartridge. 

So, pull back the sleeving and the source of the heat is Slice's crimp, where the stiff single strand resistance wire coming out of the heater cartridge joins the thicker multi-strand wire that leads back to the MiniRambo.  

Seems to me that I "just" had this issue several  months ago with an E3D heater cartridge.  Not again! And this time with a different manufacturer.

By comparison, E3D uses a much longer crimp in an apparent attempt to prevent this. My failed E3D 12v / 40w heater had a poorly formed crimp, which led to overheating of the connector, causing excess voltage drop, then underheating of the heater and finally the dreaded Thermal Runaway.  As the Slice heater is 12v and 50w, the shorter crimp area means less contact and a higher chance of failure with the higher current draw.  The Slice crimp looked correctly made, but yet it was getting extremely hot on one wire.

Last time around, I just replaced the E3D heater cartridge which worked great until I replaced the E3D with the Mosquito.  

As I have a spare heater element, I thought to experiment a little.  Why not see if I could repair the crimp? I do have some experience with high-load high-temperature wiring repairs and a whole bunch of various high-quality crimp connectors.

This time, I set about repairing the crimp.  I merely cut out the old crimp, stripped back the insulation, cleaned up the bare wire and re-did the crimp with an "open-barrel" tin-plated brass eyelet connector with the eyelet cut off.  This gave a crimp similar to Slice's but about 1mm longer, at about 5-6mm contact area.  Start it up and preheat, and the crimp now remains cool, though the resistance wire does get pretty warm.  Re-assemble and  start the print and the temperature is once again extremely stable. 

From the number of threads on Thermal Runaway errors, I suspect that this is a pretty big issue.  Yes, replacing the "bad" heater cartridge is the best immediate solution, but even heaters from high-quality companies like E3D and Slice have failed on me.  I can't imagine the failure rate of El Cheapo heaters.

So, I was wondering if there is a better way to join the stiff resistance wire to the flexible multi-strand wire.   I think E3D has taken a good step forward by merely using a longer crimping ferrule which increases the surface contact area between the dissimilar wires.     In applications with space to spare, a high-temperature terminal block  could be used with screws to maintain tension on the resistance wire for good contact even after many heating and cooling cycles.  I have even seen wires spot-welded to each other to avoid the chance of a failed crimp, but could this work on the resistance wire?   Of course normal solder would just increase resistance and probably quickly melt if it even stuck to the resistance wire at all, but maybe some kind of brazing?