Help With Thermal Runaway

Troubleshoot issues with the Tarantula
Post Reply
rebeltaz
Posts: 28
Joined: Sun Mar 26, 2017 6:28 am
Been thanked: 6 times
Contact:

Help With Thermal Runaway

Post by rebeltaz » Sat Aug 19, 2017 4:34 am

I need help! A few days ago, my hotend started dropping temp so I canceled it, reset the printer and OctoPrint and everything was fine. The other day installed Marlin 1.1.4 when installing a capacitive sensor. I printed twice with no problem. Tonight, twice it locked with a "Thermal Runaway" error. I checked the setting under [Thermal Runaway] and they were different between the old version and the new version, so I fixed those and reflashed. It shut down with that same error again. My target temp was 235, but it had dropped to 223.

I found last night that the heater was loose in the heatblock, so I tightened that up, but it is still shutting down with the same error and it is still dropping temperature. I zip stripped the heater wires to the fan duct so they don't flex and I am printing the same file now empty - no filament - just to see if it does it again. So far, at 66% into the file, it's still printing.... If it keeps going, maybe bad wires?

The errors occur at different points in the print, by the way, and I have resliced the part, so it's not a corrupt file. This takes anywhere from 30 minutes to an hour and a half to show up.

Any ideas why this is happening all of a sudden?

-=[EDIT]=-

CRAP. After completing successfully with no filament loaded, I went for a full run. Half hour or so in, temperature drops to 225° and shuts down. I have a fang duct that drastically drops the temperature, but a long time ago I figured out that raising the temperature slowly over a millimeter keeps that in check. I have never had this problem before and nothing has changed other than the firmware and the capacitive sensor. I don't understand why it works fine on a dry run, but not with filament loaded.

User avatar
royg
Posts: 81
Joined: Tue Mar 28, 2017 10:56 pm
Location: Australia
Has thanked: 33 times
Been thanked: 21 times

Re: Help With Thermal Runaway

Post by royg » Sat Aug 19, 2017 9:18 am

Have you seen this paragraph in the Marlin FW documentation for configuration.h ? It contains a few hints regarding what the cause possibly is ? The following is extracted from the configuration.h:

"If the measured temperature drifts too far from the target temperature for too long, the machine will shut down with a “Thermal runaway” error. This error may indicate poor contact between thermistor and hot end, poor PID tuning, or a cold environment."

Hope that info helps,

RoyG
Tevo Tarantula - Large Bed - E3dV6 - BLTouch
OctoPi (on a Raspberry Pi 3B)
Cura

rebeltaz
Posts: 28
Joined: Sun Mar 26, 2017 6:28 am
Been thanked: 6 times
Contact:

Re: Help With Thermal Runaway

Post by rebeltaz » Sat Aug 19, 2017 11:39 pm

Well, I didn't specifically see that, but I have checked that everything is tight on the hotend; I redid the PID tuning twice; and I can ASSURE you that is is FAR from COLD in here! lol... I don't run air conditioner and it is mid-summer here in the Heart of Dixie. Need I say more? :)

What I don't understand is why it has been working all this time, but now all of a sudden it's throwing up these errors. Even more odd is that it didn't have one problem "printing" without filament throughout the complete file, yet when I load filament - BAM! Error.

I am reprinting now with the fan speed dropped down to 75%, but if that fixes it - I'll know in a couple of hours - why now after all this time?

I cannot ever have normal problems....

User avatar
royg
Posts: 81
Joined: Tue Mar 28, 2017 10:56 pm
Location: Australia
Has thanked: 33 times
Been thanked: 21 times

Re: Help With Thermal Runaway

Post by royg » Sun Aug 20, 2017 12:54 am

When it comes to 3D printers - there's no such thing as a normal problem........
Tevo Tarantula - Large Bed - E3dV6 - BLTouch
OctoPi (on a Raspberry Pi 3B)
Cura

rebeltaz
Posts: 28
Joined: Sun Mar 26, 2017 6:28 am
Been thanked: 6 times
Contact:

Re: Help With Thermal Runaway

Post by rebeltaz » Sun Aug 20, 2017 1:48 am

You seem to know Marlin. Is there anything in there that I could check that could cause the hotend to no longer be able to keep up with the fan running full speed whereas it was before the firmware update? After dropping the fan speed, It printed relatively trouble free. The temperature is fluctuating between 229 and 237, mostly when the carriage is moving faster - I guess the airflow from the fan is moving across the nozzle as it moves - but it recovers relatively quickly. There has to be something in the firmware that I am missing.

User avatar
royg
Posts: 81
Joined: Tue Mar 28, 2017 10:56 pm
Location: Australia
Has thanked: 33 times
Been thanked: 21 times

Re: Help With Thermal Runaway

Post by royg » Sun Aug 20, 2017 3:00 am

There may have been code changes in the latest version of Marlin that improve the hot end temperature management in certain conditions, but potentially those code changes may be having unintended consequences in your particular situation.

Short of modifying code in files other than configuration.h and configuration-adv.h, I can't see any way of configuring marlin to overcome your problem.

When I was still using the original Tevo Hotend, I created my Over-kill Layer Fan design https://www.thingiverse.com/thing:2410225. When I started using that layer fan, I started getting all sorts of Marlin Errors involving hotend heating. The problems were only solved by turning the layer fan down to it's very lowest speed - thus defeating the purpose of having a Layer Fan.

For other reasons (mainly frequent hotend blockages, and frequent leaks), I ended up replacing the Tevo Hotend with a genuine E3Dv6 Hotend. To my surprise, the E3Dv6 included a sample of a then new product from E3D. The new product was a Silicon Sock that fits over the E3Dv6 hotend to insulate the hotend.

Image

With the Silicon Sock fitted over the hotend the PID Tune ends up with hugely different values - and the hotend temp stays very close to target temperature, even with the Over-Kill Layer Fan blowing a gale onto the print job. Printing unsupported bridges is much easier with the hotend insulated and using a really powerful layer fan. With the hotend heated to 240 degrees C for PETG, I can still place my finger on the silicone sock, which is only slightly warm.

So, I got to thinking. What sort of insulation could you assemble together to insulate a Tevo standard Hotend, or any other hotend?

NOTE: I have not tried the following - it's simply an idea that I'm throwing out there.....

The silicon material used in the E3D Silicon Socks looks and feels a lot like this material that I bought from a local kitchen supplies store. The material is a thin flexible Silicone Baking Sheet. Some of these Silicone Baking Sheets are thick (3mm and more) and do not roll up. The type you want for this job would be the thin (my stuff is around 1 mm thick), and is usually sold rolled up in a carton, as per the following photograph:

Image

I use this material, attached with high temp double sided tape, to insulate the underside of my Heated Print Bed. And, if the material can stand up to 250 degrees C in a kitchen oven, then it will be able to handle those same temperatures when wrapped around a 3D Printer's hotend.

What I'm thinking, is that you could use some of this silicone baking sheet - cut to a suitable shape, to fold around your hotend, and held in place with Kapton Tape.

I suspect that the problem you're encountering is a common problem, because E3D in their marketing blurb for the silicone sock, talk about it being of great value when using Layer Fans. I think that playing with the Marlin Firmware to solve this problem is not actually fixing the root cause of the problem. The root cause of the problem is that the Layer Fan is over-cooling the hotend. Insulating the Hotend is one way of directly addressing the root cause.

The other way of directly addressing the problem may be to fine tune the design of your fan duct to maximise the air flowing onto the print job, whilst minimising the airflow flowing directly and indirectly onto the hotend - not an easy design task.

I hope this information is useful to you.

RoyG
Tevo Tarantula - Large Bed - E3dV6 - BLTouch
OctoPi (on a Raspberry Pi 3B)
Cura

rebeltaz
Posts: 28
Joined: Sun Mar 26, 2017 6:28 am
Been thanked: 6 times
Contact:

Re: Help With Thermal Runaway

Post by rebeltaz » Sun Aug 20, 2017 4:39 am

You have most definitely. I'll be honest with you. The only thing keeping me from reflashing with the old firmware is that it doesn't have the bilinear leveling mode and my bed is not exactly a perfectly flat plane. I'll have to sleep on this. Thank you so much for your time.

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest