How would I know if the temp reading is accurate?
 
Notifications
Clear all

How would I know if the temp reading is accurate?  

  RSS
Rutilate
(@rutilate)
Trusted Member
How would I know if the temp reading is accurate?

Folks, I'm having some challenges with under extrusion and poor bed adhesion that appear to be related to low hot end temps. How would I know if the readings are accurate?

My first layer rarely adheres well, even though my Z offset is so low that the first layers are nearly translucent. The extruded lines are very thin.

I'm using an infrared digital thermometer ( https://www.amazon.com/Etekcity-Lasergrip-1080-Non-contact-Thermometer/dp/B00DMI632G ). I point it at the bed and it reads 51-56C in many places--seems legit. When I place the opening right up around the extruder, I get a temp of 150C, when the extruder is set at 215C. I position the opening as close as possible to the heater (where the cables exit the ceramic) and also get 150C.

I used a different infrared digital thermometer and obtain values ~20C lower (but certainly not higher).

I used a FLIR attachment for my Android phone and captured an image of the bed. (ETA: it doesn't appear that the forum software is displaying the images in the same order as they are placed in the editing window.)

The temp reading from the bed shows temps 52.2C to 57C; seems legit.

If I also point the FLIR at the heat block, extruder, and heater, I get a decent thermal image:

and the accompanying image shows high temp of 150.2C:

So, I've tested the extruder/heater/heatblock with three different methods, all of them showing a temp no higher than 150C, which is much different than the 215 setting/reading shown on the LCD.

What reading do I believe? If I believe the FLIR and IR readings, how do I diagnose the discrepancy?

Prusa has been very generous and replaced the heater, heater block, and thermistor. Odds are high that these are not the culprits. What else might explain the problem?

Posted : 16/07/2016 2:32 am
martorias
(@martorias)
Eminent Member
Re: How would I know if the temp reading is accurate?

I'd say the hotend is too small to be correctly measured with your tools. You could use a thermocoupler instead and press it against the hotend. Another way: if you get 150c when you select 215, I'm guessing you get like 100c if you select 150c?
At around 150-160-170 PLA should begin to "run" through the nozzle, which it won't at around 100c so that should give you a hint if it's indeed totally off.

Posted : 16/07/2016 10:13 am
Rutilate
(@rutilate)
Trusted Member
Topic starter answered:
Re: How would I know if the temp reading is accurate?

Martin thanks for the recommendation. Anyone recommend a good thermocouple that'll plug into a multimeter either directly or with an adapter?

Posted : 16/07/2016 9:26 pm
PJR
 PJR
(@pjr)
Antient Member Moderator
Re: How would I know if the temp reading is accurate?

For measuring hotend temps, I use a nitro r/c car thermometer. Very cheap and accurate enough.

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…

Posted : 16/07/2016 10:23 pm
Rutilate
(@rutilate)
Trusted Member
Topic starter answered:
Re: How would I know if the temp reading is accurate?

For measuring hotend temps, I use a nitro r/c car thermometer. Very cheap and accurate enough.

Peter
Peter, all the r/c thermometers I'm seeing are non-contact IR, just like what I'm using but smaller. Is that what you're suggesting?

Posted : 17/07/2016 12:34 am
kevin.b2
(@kevin-b2)
Trusted Member
Re: How would I know if the temp reading is accurate?

Curtis - Keep in mind that by the nature of how thermocouples operate, you will only be as accurate as your cold end compensation - so i'm not sure if you'll get great accuracy using a TC hooked up to a standard multimeter. I would think that the thermistor being used in the hotend would be more accurate than a multimeter type measurement with a TC.

I'm going to try to get a shot of my bed and hotend with an industrial FLIR camera and see how it compares to what you've captured with the Android variety.

Posted : 17/07/2016 4:45 am
Rutilate
(@rutilate)
Trusted Member
Topic starter answered:
Re: How would I know if the temp reading is accurate?

Curtis - Keep in mind that by the nature of how thermocouples operate, you will only be as accurate as your cold end compensation - so i'm not sure if you'll get great accuracy using a TC hooked up to a standard multimeter. I would think that the thermistor being used in the hotend would be more accurate than a multimeter type measurement with a TC.

I'm going to try to get a shot of my bed and hotend with an industrial FLIR camera and see how it compares to what you've captured with the Android variety.

Awesome, Kevin. Looking forward to seeing the results!

Posted : 17/07/2016 5:27 am
kevin.b2
(@kevin-b2)
Trusted Member
Re: How would I know if the temp reading is accurate?

Here you are! Bed was set at 80 degrees C. I found an emissivity value of 0.89 for Ultem (PEI) online and used that, seems to have given reasonable results in terms of absolute accuracy. The temperature distribution is interesting, seems to confirm my "finger tests" that suggested the middle of the bed isn't super hot 😎

I'll work on getting something of the extruder at temp, maybe without the cover as well...so many possibilities...

Posted : 23/07/2016 7:19 am
gz1
 gz1
(@gz1)
Estimable Member
Re: How would I know if the temp reading is accurate?

Here you are! Bed was set at 80 degrees C. I found an emissivity value of 0.89 for Ultrem online and used that, seems to have given reasonable results in terms of absolute accuracy. The temperature distribution is interesting, seems to confirm my "finger tests" that suggested the middle of the bed isn't super hot 😎

I'll work on getting something of the extruder at temp, maybe without the cover as well...so many possibilities...

Nice pic!

I was about to say I believe those IR thermometers make an assumption about the emissivity of the object you're trying to measure.

The cold spot might not be surprising since there are two metal screws right there in close proximity of each other (and if the screws are there, that means the heating element isn't there), and they're leading straight down to the metal bed platform through metal standoffs.

Posted : 23/07/2016 8:29 am
Share: