When printing ABS, I tend to heat up the chamber to 50° using the Aux fan and bed temp at 110°.
A single print I do will usually take about 1h50m. I clean the bed and immediately try to run a 2nd print.
This time around, the slicer (Bambu or Orca) will have a hard time connecting to the printer. The results will usually be connection drop outs or no connection at all. I cannot send the print job to the printer.
Before printing ABS, I isolated the electronics and added a 2nd Noctua fan to exhaust any hot air from the MC board, but this doesn’t seem to help.
I have vast experience with ruggedizing technology for extreme environments over my career. 50 °C is a magical number in that many commercial semiconductors will only guarantee their product to that limit. Truth be told, you can almost always counts on on their being 15 °C of headroom because the specs are usually set conservatively.
Now having said that, I’m going to pull a page out of the PC modder’s playbook where we are always pushing the temperature envelope of performance as we overclock everything.
If you really want to get to the bottom of this and you strongly suspect temperature to be an issue, then the best way to prove this is to put two thermocouples onto your critical parts. This is what I did with my P1P from the start when it was at one time recommended to add a fan to the MC board. As it turns out, I’ve had that connected to an external thermometer ever since it it was an unwarranted scare, the MC never went above 55 °C which is well within the tolerance of the micro controller.
Now this suggestion isn’t has hard as it may seem. Here’s how to do it.
By a cheap thermometer with a thermistor probe. The following uses the same 10k NTC thermistor that’s used in the hot end of the Bambu nozzle.
I paid $8 on Amazon but I also used this too and made my own probes with wire and cheap thermistors, this accommodates 2 probes. :
Then simply tape them on to the part you want to monitor and set the thermometer outside of the printer so as not to influence the measurements. 1/2" Kapton tape is a good material to have around anyway and has other purposes such as taping down something onto a heat bed.
Then you want to apply a probe to these two locations. Note that they the back cover needs to come off to get to the MC and you’ll have to peel back the black cloth tape on the MC interface board.
Now that you have two temp probes placed, you can truly measure “real” temps. If you find that these areas develop hotspots, then the remedy is to provide cooling. That part is going to be tricky but a 40mm fan will work. I prefer these little USB products for exactly this purpose and use them on my PC overclock builds if I suspect that a component might be overheating.
I found these Gooseneck extenders to be handy if you want to have a more firm control of where to place and focus the fans. Agreed, there is some engineering of a mount involved as you will have to fabricate a mounting scheme, but hey, that’s what 3D printing is for.
Mate, thank you very much for your detailed and thorough explanation!
Have you tried adding a fan over the wifi antenna?
or do you have an idea what is a good DIY way of isolating it from the chamber heat?
The Wi-Fi antenna doesn’t have any active components to be affected by heat. It is literally just a copper strip of flat wire in a specific shape. You’d have to go to very extreme temperatures to warp it or cause temperature-induced anomalies. I have no experience in that realm, but I suspect it would have to be in the 400°C range, where solder would melt, which is obviously not what we would see inside a chamber.
Having said that, I still suspect that if it’s thermally related problem, it will be either in the MC board or the interface board.
Here’s one trick you can try. Take a can of compressed air and turn it upside down to blow on the component. Now I realize that is tricky because you will be changing the test conditions simply by opening the door when you try to place the air can inside, it’s also damned near impossible. So here’s a trick.
Purchase the air can along with longer straw which can be found on Amazon and Alley express. Then tape one end of the straw near the component needing cooling and thread the straw outside of the case and when your Wi-Fi disconnects, connect the can, turn it upside down to guarantee fluid comes out and give it a squirt. Not to much because you could crack the components if you freeze them. This cold fluid will give a few seconds of lower temps. If you combine that with the thermocouple trick above, you can closely monitor the temperature.
Here’s one source of such a straw for this purpose. They can be cut to size.
Yes and… no.
A bad solder joint can fail with temperature as the physical expansion can let to a non-contact situation of the joint. It’s a very common issue in electronic component PCBs.
