temperatur reading calibrating, possible?

[3dcase 37]

hI,

Posting the question here, don't now if it is a firmware, ulticontroller, cura or electronics question.

Is it possible, somewhere, to calibrate the temperature readings of the hotends?

Using dual extruders now and there is a 20°C difference between the hotends.

same filament needs a 20° C higher temp on 1 or the other hotend.

No life threatening situation, but would be nice if I could calibrate it.

I know I could check or reinsert the heater and sensor, But don't want to try and get them out if there is another way.

Regards Kees.

 

[gr5 2,265]

I'm a software guy so I would offer a software solution. First I would make a table of programmed temp and actual temp in the region of where you print (for example 160C to 260C).

Then I would update the table in the Marlin firmware.

If the error is a simple offset (like add 30C) then I believe there is a single line in configuration.h to fix this but I don't think that will work for you because that would affect both nozzles. Instead I would... hmm... this is tricky as both use heater number "-1" which is a formula instead of a table (the bed uses a table and is easier to mess with).

Okay so what I would do is hack the actual code that does the calculation. I simplify the errors to just a gain and offset (multiply by A and add B to get the correct temp) then I would do that multiply and add only if it was heater number 2 or whatever.

It's pretty easy to build your own marlin. I prefer the gui but here's a well documented way:

http://www.extrudable.me/2013/05/03/building-marlin-from-scratch/

 

 

[3dcase 37]

I'm a software guy so I would offer a software solution. First I would make a table of programmed temp and actual temp in the region of where you print (for example 160C to 260C).

Then I would update the table in the Marlin firmware.

If the error is a simple offset (like add 30C) then I believe there is a single line in configuration.h to fix this but I don't think that will work for you because that would affect both nozzles. Instead I would... hmm... this is tricky as both use heater number "-1" which is a formula instead of a table (the bed uses a table and is easier to mess with).

Okay so what I would do is hack the actual code that does the calculation. I simplify the errors to just a gain and offset (multiply by A and add B to get the correct temp) then I would do that multiply and add only if it was heater number 2 or whatever.

It's pretty easy to build your own marlin. I prefer the gui but here's a well documented way:

http://www.extrudable.me/2013/05/03/building-marlin-from-scratch/

 

 

Thanks for the quick and detailed answer!!

But euuhm, I'm more a toolbox guy, Glad with myself if I can find out how to work with the software :cool:.

The writing and mod's I leave to you guys.

In my last year in school we spend 1 hour a week looking with 20 guys on the green monitor of an early apple with only a tape deck. Blame the guy who put that antique in a dumpster...

Afraid i'm to old to keep up with software development........ :(.

I might give it a try with a resistor or potentiometer.........

Regards Kees

 

[Dim3nsioneer 558]

I'm not sure if I got you right: do you get a different temperature reading from the Ultimaker compared to a external measurement?

 

[3dcase 37]

I'm not sure if I got you right: do you get a different temperature reading from the Ultimaker compared to a external measurement?

 

I use a dual extruder setup, there's a diffenrence between the 2 hotends.

say, printing white pla in extruder 1 I can go as low as 180°. but same rol in extruder 2 needs 205°.

I don't care too much about the real temperatures, but like the readings to be the same. Need a lot of notes now, to remember what I did :(.

 

[Dim3nsioneer 558]

...But euuhm, I'm more a toolbox guy,...

 

Then I guess you may find it with a multimeter checking the resistance and the connections... :wink:

 

[kjms30 0]

Hi

I have a similar problem were by my thermsensor board works (i.e. the one on the top with the blue led) but when I switch on the ultimaker it reads as 41c and not normal room temp of about 24c.

I have tested all the wires and solder and configuration all is OK from what I can see and tested with digital multimeter and still its high. I believe the AD595 has an internal temp sensor and if that's the one going crazy.

I can still print OK but I have to set the temp higher to compensate for the offset.

gr5 you mentioned about changed the line in config.h do you know what line it is that needs to be changed. I have searched it but can't see anything that stands out.

Hope someone can tell me how to adjust the the offset.

 

[gr5 2,265]

Just edit Configuration_adv.h here:

//These defines help to calibrate the AD595 sensor in case you get wrong temperature measurements.

//The measured temperature is defined as "actualTemp = (measuredTemp * TEMP_SENSOR_AD595_GAIN) + TEMP_SENSOR_AD595_OFFSET"

#define TEMP_SENSOR_AD595_OFFSET 0.0

#define TEMP_SENSOR_AD595_GAIN 1.0

And rebuild Marlin. It's not that hard to build Marlin. I recommend the latest Erik Zalm version. Actually I recommend you go here first:

http://marlinbuilder.robotfuzz.com/

 

And use that to get your configuration.h file. *then* get latest source code of Marlin (which is updated almost every week but which has always worked for me):

https://github.com/ErikZalm/Marlin

Do a diff (winmerge is free and works great) on the Configuration.h file from there and the one from marlinbuilder and pay attention to the changes. They should all make sense. You don't have to know anything about programming to understand these files. Just know that anything after the "//" is a comment. Don't forget to edit the Configuration_adv.h (I've never had to touch that but that is the change you want for your temperature fix).

Then you can build it through the arduino gui (there's instructions inside the source files somewhere) or you can build it according to this detailed instructions:

http://www.extrudable.me/2013/05/03/building-marlin-from-scratch/