I've been struggling with temperature issues on my hotend for seemingly months now. While the bed temperature has always been rock solid, the hotend can either be very solid, or just wildly all over the place, sometimes triggering a thermal shutdown. I'm running an i3 variant with Ramps 1.4, Arduino Mega 2560 R3, and Marlin bugfix-2.0.0.

I'm running an E3Dv6 hotend that uses the thermistor cartridge. I'm connecting the thermistor to the Ramps board with the E3D supplied cable made for the thermistor cartridge. I've done the PID AutoTune.

What I'm seeing now, is terrible instability in the hotend temp at the beginning of the print, which smooths out later and either maintains through the rest of the print, or goes through other periods of instability. For instance, this is a print I attempted tonight. The first image is showing the last part of the first layer and going into the 2nd layer. This is not even the worst of the hotend fluctuations that occurred previous to what's in the image. You can see the bed switching from the first layer temp to the 2nd layer.


This image is at about layer 8. Temp levels out and remains very stable for some time. The third image shows this stable period.




Then around layer 20, the temp took about a 15-20 degree down spike and while it was climbing back up and I was going to grab a screen shot, it must have jumped the other way and triggered a thermal shutdown.

Hardware wise, I've swapped out just about every possible culprit. I have another E3Dv6 for another printer I'm building, so I swapped out the themistor cartridge with no change. I've gone through 3 sets of Mega/Ramps boards, no joy.

I've seen bugs in Marlin before that were causing some temperature issues, but those were fixed long ago and I see no open issues with Marlin currently. I just updated to the latest pull last night.

Could it be a power supply issue? The only thing I can think, if swapping the thermistor didn't cause any change is that something is causing fluctuation on the AREF which is then causing fluctuations on the readings. But I wish I knew why it always is just the hotend and not the bed. Seems like if it was a power supply thing, the bed would also be showing instability at the same time, but it doesn't.

Does anyone have any other ideas?

Just going out on a limb here and suggest that it might be a faulty MOSFET? Or maybe a break in the solder of the MOSFET?

Check the soldering of the MOSFET and see if it came loose somewhere. I don't know how you can check if a MOSFET has gone bad, but that's where I would check. Maybe just replace the thing since it's a cheap component anyway.

If it was a PSU issue, you would first notice this on the heated bed since that one pulls way more Amps. However, fluctuation on a heated bed is less noticeable since it has a larger thermal mass. But I suspect the MOSFET.

---

http://www.marinusdebeer.nl/

Quote
Ohmarinus
Just going out on a limb here and suggest that it might be a faulty MOSFET? Or maybe a break in the solder of the MOSFET?

Check the soldering of the MOSFET and see if it came loose somewhere. I don't know how you can check if a MOSFET has gone bad, but that's where I would check. Maybe just replace the thing since it's a cheap component anyway.

If it was a PSU issue, you would first notice this on the heated bed since that one pulls way more Amps. However, fluctuation on a heated bed is less noticeable since it has a larger thermal mass. But I suspect the MOSFET.

Hi Ohmarinus,

You mean the MOSFET on the Ramps board? Like I said before, I've actually tried three different Mega and Ramps boards in the effort to figure it out. My bed is a 12v silicon heater which I forgot has it's own control board and MOSFET which could be why I don't see any fluctuation there, but since I've tried different Ramps boards with no change, I'm guessing it's not related to that.

Quote
brucehvn

Quote
Ohmarinus
Just going out on a limb here and suggest that it might be a faulty MOSFET? Or maybe a break in the solder of the MOSFET?

Check the soldering of the MOSFET and see if it came loose somewhere. I don't know how you can check if a MOSFET has gone bad, but that's where I would check. Maybe just replace the thing since it's a cheap component anyway.

If it was a PSU issue, you would first notice this on the heated bed since that one pulls way more Amps. However, fluctuation on a heated bed is less noticeable since it has a larger thermal mass. But I suspect the MOSFET.

Hi Ohmarinus,

You mean the MOSFET on the Ramps board? Like I said before, I've actually tried three different Mega and Ramps boards in the effort to figure it out. My bed is a 12v silicon heater which I forgot has it's own control board and MOSFET which could be why I don't see any fluctuation there, but since I've tried different Ramps boards with no change, I'm guessing it's not related to that.

Ohhh I'm sorry, I missed that part! Then forget what I wrote. Just to be sure on the hardware side, did you replace the complete thermistor wire or only the cartridge?

Otherwise, I wouldn't have a clue about a current issue with Marlin..

---

http://www.marinusdebeer.nl/

Sudden erratic jumps in temperature readings are usually caused by bad connections to the thermistor, or a thermistor wire touching the hot end metalwork coupled with leakage in the heater cartridge. Just possibly by a faulty thermistor.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

One other possibility is overloading the 5A polyfuse - if you have a lot of power to your steppers and the hotend, the polyfuse might start restricting the current and that could cause the hotend to fluctuate. Adding an external MOSFET for the hotend will solve that issue.

Quote
dc42
Sudden erratic jumps in temperature readings are usually caused by bad connections to the thermistor, or a thermistor wire touching the hot end metalwork coupled with leakage in the heater cartridge. Just possibly by a faulty thermistor.

Hi dc42,

That was my first thought, but I've got two thermistor cartridges and they show the same behavior. Unless I got two bad ones from E3D, it appears to be something else. The heater block has a silicon sock and I've made sure the wires are not trapped underneath or rubbing against anything hot. The wires pretty much come straight out away from the heater block. I'm swapping out the thermistor cable and trying one last mega board today and we'll see how that goes.

Quote
brucehvn

Quote
dc42
Sudden erratic jumps in temperature readings are usually caused by bad connections to the thermistor, or a thermistor wire touching the hot end metalwork coupled with leakage in the heater cartridge. Just possibly by a faulty thermistor.

Hi dc42,

That was my first thought, but I've got two thermistor cartridges and they show the same behavior. Unless I got two bad ones from E3D, it appears to be something else. The heater block has a silicon sock and I've made sure the wires are not trapped underneath or rubbing against anything hot. The wires pretty much come straight out away from the heater block. I'm swapping out the thermistor cable and trying one last mega board today and we'll see how that goes.

I expect swapping the cable will fix it. Let us know!

---

http://www.marinusdebeer.nl/

I have the same issue, almost exactly the same.

Bed temperature is rock solid, but hotend varies between stable and a bit wild. Can sometimes throw THERMAL RUNAWAY errors. I've upped the time period and temp variation allowances in configuration_adv.h in the hope that will help prevent failed prints.

I've also changed RAMPS boards, I'm using a new 1.4 board, with a 1.5 and a 1.6 to try also. I've also ordered a new thermistor cartridge from E3D (I'm running a genuine E3Dv6) so and will take extra care to make sure the thermistor cables are not unduly bent or touching the heater block, essentially I'll make sure the wires are in perfect order as they reach the RAMPS board.

Following this thread with interest!

Joe

---

JK3D Blog - [jk3d.wordpress.com]

Well, I was going to hold off on replying until I did another print later today, but last night I did 2 prints for a total of about 3 hours and didn't see a bit of instability in the hotend temperature. I changed two things, the mega board and the thermistor cable. Now I have to see which one was actually the problem. It should be pretty easy because the old thermistor cable is still there, I just ran the new one temporarily, so I can swap it back easily. The cables I'm using are the ones that came with the E3Dv6's. If it turns out to be the cable, then I guess I just got a bad one. I hope I'm not speaking too soon, but I will update after the next print, which will take about 2 1/2 hours once I start it.

Yep, mine was the thermistor (or most likely the cable from the thermistor that goes past the hotend). Changed it yesterday and temps are solid now. Glad it's fixed. I made sure to give the hotend a wide berth in terms of cable tidying.

---

JK3D Blog - [jk3d.wordpress.com]

I made my other print which was 2 1/2 hours with the old thermistor cable and didn't see any issues. I'm not ready to declare total victory yet, as I've seen cases where it could work for a print occasionally, so I'm going to print a few more to be sure.

I have the very same issue.

Bought myself a genuine E3D V6 hotend for my printer and installed it ok. It prints well, but almost immediately my printer started to intermittently shut down with Thermal runaway warnings. So I went into Marlin and changed the Thermal Runaway Hysteresis from 4 to 10 deg C and that seemed to cure that false warning. But then during printing I noticed that the hot end temp would fluctuate quite a lot from my 205 deg setting to 190 and even sometimes down to 170 before clawing its way back, this would trigger a cold extrusion warning and stop the extruder for a few moments). The MOSFET light on my Bigtreetech SKR 1.3 board was flashing intermittently and not smooth at all which indicates its reacting correctly to the fluctuating inputs from the thermistor. I checked in Marlin again and set the hot end thermistor to number 5 to match the 104GT thermistor E3D say is supplied, but the fluctuation is still there during printing. I tried changing it in Marlin to item 55 which is the same 104GT thermistor with a 1k pullup resistor (rather than 4.7K) but that gave the cold idle temp reading of 56 deg c in a 21 deg c room. So I changed it back to item 5 again and it now reads 21 deg C correctly. I've ordered a replacement thermistor but I doubt it will fix it. Going to try separating the thermistor cable run from the rest of the cables going to the hot end to see if that Isolates the issue.

I ended up replacing mine with the thermocouple and thermocouple board from E3d. But I'm not actually that happy with it. The temperature still fluctuates by about +- 2C even when not heated and the printer is sitting idle. Seems thermocouple cables are very sensitive to magnetic fields and you can't run them tied to other things like your heater or fan wires which is difficult to maintain. But I've just resigned myself to what appears to be the fact that I'm just not going to get anything better. Even though I once had a very stable hotend temperature, doesn't seem to be achievable anymore.

I had similar issues. Turned out it was the fan. A PWM calibration with the fan running helped, but socks work better.
Initial instability was caused by the fact that the air is bouncing off the bed onto the hotend. Later it does not hit the hotend because of the height.