Thermal Runaway because of fan deflection?  

Page 1 / 2
  RSS
thomas.p19
(@thomas-p19)
Active Member

Hello, I need some help trying to diagnose an issue.

I've been getting a Thermal Runaway error from your typical Hotend dropping temp for too long kind of situation.  First layer goes down fine, fan spins up, hotend cools down.  You might think, "Ah!  fan is overpowering the heater cartridge, easy diagnoses."  or "Ah! I the fan shroud is angled up and cooling off too much of the hot end or thermister".  The Problem is, I'm not sure its those and I want a second opinion.

Some notes: 

Hotend gets up to temp at a normal speed, can still get to higher than  necessary temperatures, seems intact.

Fan spins up on layer 2 at full blast(as is default), in a controlled elevated position, hotend is not cooled by the fan enough to reduce its displayed temperature.  Fan shroud hasn't been displaced.  R4 extruder cover is a little warped but its not sagging yet.

When the hotend is against the heatbed and the fan is on, now I get a drop in temperature.  It seems to me that the air is deflecting into the hotend suddenly and I don't really understand why.  At fan speed 180(out of 255), the fan pulses Rrrrrrr Rrrrrr Rrrrrrr but is able to cool without thermal runaway.  

The first question I'm trying to answer is whether this is an indication that my heater cartridge is dying

And secondly, has air deflecting off the heat bed ever been a problem someone has seen before?  I'm not able to find a case of it so I question whether that's my root cause or just a symptom.  Has the fan speed been adjusted in any of the recent software/firmware?  I'm a bit perplexed as to why this is happening now.

Opinions welcome!

Posted : 25/09/2019 12:49 am
thomas.p19
(@thomas-p19)
Active Member

You know what would help?  The Model.  This is a Prusa i3 MK3s(upgraded from a 3).

Posted : 25/09/2019 1:38 am
Tim
(@tim-m30)
Illustrious Member

Common issue; run a PID at the temp you plan to print most often.  That will help the system know the thermal properties of your specific printer.  A sock also helps keep the thermal jolt of the fan from being a problem.  If you install a sock, do another PID.

This post was modified 10 months ago by Tim
It is always wise to get more than one opinion......
Posted : 25/09/2019 8:32 am
thomas.p19
(@thomas-p19)
Active Member

Update time, thanks for the suggestion Tim!  A PID calibration was one of the first things I did when this problem was starting but I gave it another shot.  I still saw the temperature dropping too fast when the nozzle is close to the bed.  What really sucks about this is that the zone that this occurs in is like only the millimeter or two above the surface.  

Next I cranked the fan up and did a PID Calibration.  It managed to complete but once I got to printing, that second layer fan defeated the hotend yet again.

I don't have a sock around and that seems like a reasonable suggestion.  While I've been thinking of installing a sock for some prints where I want to control heat bleeding into a small object, I'm a bit worried this is a symptom of a something not right.

I'm trying to think of any way I could test for hardware faults I can understand.  I'm not much of an electronics guy.  So again, I wouldn't mind suggestions 🙁

Posted : 26/09/2019 12:27 am
michael.s233
(@michael-s233)
New Member

Interesting you're having this issue.  I've had my MK3S since it came out.  Kit, not fully assembled.  Been running it non-stop since.  I do tons of preventative maintenance, keep it oiled and clean.  I've made NO changes to it whatsoever, and this started happening to me as of yesterday.  The only way I can reliably print now is with the fan off, which was NEVER an issue before.

Also, in the past week or so, it's started stringing really badly, which it never did before either.  Been tweaking my slicer and nothing seems to help. This is the same rolls of filament I've been printing with for a long time, and multiple rolls now all exhibit the same behavior.

Posted : 26/09/2019 3:13 am
Tim
(@tim-m30)
Illustrious Member

The print fan shroud position has been suggested as a possible variable.  Make sure it points downward, away from the heater block.  Seems they can be bumped to point a bit upward directing air flow directly on to the heater block.   Short of that - make sure the print support of the shroud is removed - the bar between the horns.

It is always wise to get more than one opinion......
Posted : 26/09/2019 9:45 am
guy.k2
(@guy-k2)
Noble Member

Is the system actually indicating a HEAT BED thermal runaway rather than hot end thermal runaway. The heat bed runaway looks very similar and manifests when cooling fan comes on. It tends to happen more on colder days or when using higher bed temperatures like 100C. With days getting colder, this might be masquerading as a hot end thermal runaway.

I solved mine with a piece of furnace insulation tape over the heat bed thermistor (bottom side of heat bed). Only needed to cover about  5 cm radius around thermistor to stop the problem.

Posted : 26/09/2019 11:28 am
thomas.p19
(@thomas-p19)
Active Member

I’m heading to work so I won’t be able to experiment some more for quite a while.  

Thanks again for suggestions!  I’ve taken the extruder partially apart and it’s possible I put the shroud back the same but I will double check that it’s tilted down as far as it will go.

its definitely a hot end run away as I’ve left the bed off for a couple of my tests.  It also doesn’t call out heatbed thermal runaway on the screen.

I also did a temp cal and pid cal back to back in case there’s a disagreement on what temperature it’s actually producing but it was time for bed.  I’ll see if that helped later.

My shroud might still have supports on it.  I feel like there was more material present than in the render Tim posted.  I did check it for deformation and obstruction.

Alright this gives me something to try later, thanks.

Posted : 26/09/2019 1:03 pm
floyd.lenny
(@floyd-lenny)
Active Member

Hi

Yeah, I've been getting the same issue..

MK3 upgraded to MK3S - Latest firmware.

Managed a Benchy and Whistle and then whilst doing a larger flat print (coffee cup coaster) it started struggling.

Completes first layer without issue but as soon as the fan powers up (full tilt) the temp gradually starts dropping. Most prints get to around 70% then drop too low and thermal runaway.

Done the PID cali and visual checks of hardware.

I have a silicone shroud on order and will do some more testing tonight with varying fan speeds.

Either the MK3 upgraded to MK3S can't deal with the amount of cooling the upgraded blower directs or its a hardware issue in which case I'll start with the Thermistor, heater etc.

I ran a previously failed print again without the fan on and it completed fine and the temp stayed constant through out the 90 minutes .

 

 

Posted : 26/09/2019 4:37 pm
thomas.p19
(@thomas-p19)
Active Member

I check the shroud, even reinstalled it.  Its angled as low as it can go.  No Change.

Then I went back a firmware version(3.8 -> 3.7.2).  I'm actually not sure if I was on 3.7.2 or if I skipped over it when I updated to 3.8.  Anyhow, I completed a print with 3.7.2.  The temp still dropped hard when the fan came on but it was able to recover and overpower the fan before thermal runaway cut it off.

So 3.8 might be what tipped me into thermal runaway, but losing 13C when the fan comes on seems like more than I recall previously so I'm not ruling out that something on my printer is underperforming.  If you're one of the ones reporting the issue here, maybe try the previous firmware and report back?  I haven't the slightest idea what would be different here.

Posted : 27/09/2019 1:22 am
guy.k2
(@guy-k2)
Noble Member

Another reason to use a silicone sock. Lots of upsides to using one. For what they deliver, silicone socks are low cost, solve a host of thermal issues, speed clean up, and reduce blob of death risk/severity. 

Posted : 27/09/2019 2:10 am
floyd.lenny
(@floyd-lenny)
Active Member

So I dropped fan speed on my last print (3 hours) and the temp on the hotend stayed constant.

Fan Speed was 160.

I also did one where I made the fan min/mac 40% and that two completed without issue.

Did a few tests with just heating up the hotend and bed to temp and putting fan on full and the hotend temp divided from 210 to 200 then came back to 210... not sure it would do that if close to the bed with deflection though.

 

I have some new parts on order though as all though the prints finish and look good, this can't be normal.

 

Posted : 27/09/2019 7:27 am
thomas.p19
(@thomas-p19)
Active Member

@floyd-lenny

Your issue sounds more severe than my own.  Outside the second layer or so, my printer operates as it has for 1.5km of filament.  I'd be interested to hear if going to an older firmware version helps you print at 100% fan speed though.  Really I'd be interested in hearing from anyone with a similar issue rolling back a version.  

Does anyone know how to produce the temperature profile provided with the printer?  I'd be interested in comparing my initial test graph to whatever performance I'm currently seeing.  I suspect they'll be pretty similar though.

Posted : 28/09/2019 12:13 am
thomas.p19
(@thomas-p19)
Active Member

Well I can rule out firmware.  I’m still seeing the issue on older firmware.  If anyone knows of a more in depth way to analyze hot end performance, I’d be happy to hear it.  

In the mean time, I’ve purchased a new heater cartridge as a final resort and I’m going to try it again with a sock once that arrives.

Posted : 01/10/2019 9:31 pm
jerry.s3
(@jerry-s3)
Trusted Member

Consider replacing the thermister. Mine declined slowly sort on like this. Replacing worked for me. After almost 2 years.

Posted : 02/10/2019 11:29 pm
thomas.p19
(@thomas-p19)
Active Member

@jerry-s3

I'll keep this in mind.  An extra thermister is something I haven't ordered yet but I'll keep an eye on it.  When you had this issue, was the hotend getting hotter than reported or was the thermister more sensitive to temperature change but otherwise accurate?

Posted : 08/10/2019 10:18 pm
floyd.lenny
(@floyd-lenny)
Active Member

I've been printing at 60% fan for the last two weeks without issue but did order a silicone sock which has now arrived and been installed.

 

Back up to 100% with no thermal errors at all.

 

Posted : 08/10/2019 10:33 pm
rosterloh liked
Tim
(@tim-m30)
Illustrious Member

Thank you for getting back and reporting success.

It is always wise to get more than one opinion......
Posted : 10/10/2019 8:38 pm
nicolas.b19
(@nicolas-b19)
New Member

Hello, I am having the same problem. I've been getting a Thermal Runaway error, it is appears only when I use PLA after the first layer. I dropped fan speed to 60% and changed the filaments temperature, that was the solution. The first layer temperature it must be 5 degrees higher than the other layers (When the fan is on).

I have another 3d printer, a Reprap Prusa with Marlin, to resolved this problem I had to change Delta temperature and the recovery time to avoid the Thermal Runaway error. I think the Prusa MK3S new firmware has wrong paramters

 

Posted : 01/11/2019 8:46 pm
Tim
(@tim-m30)
Illustrious Member

The issue has nothing to do with print temperature. lol.  

It is always wise to get more than one opinion......
Posted : 01/11/2019 10:06 pm
Page 1 / 2
Share:

Please Login or Register