Extruder temperature registers 340C when its really ~220C

Hi All,

I have built a Mendel90 that is basically all set up and ready to print except that the reported extruder temperature is wildly inaccurate. Initially, I was puzzled why the extruder would not push filament during testing, but I discovered that the extruder temperature was far to low to melt the filament.

In order to get the extruder up to around 220C, verified via external thermometer, I must set the temp through repetier to 340C (had to bump up the HEATER_0_MAXTEMP for Marlin to even allow me to set the temp that high).

The software/hardware is Marlin1.0/RAMPS with a qu-bd extruder.

I initially thought the problem was caused by the thermistor included with the extruder (I added the recommended table to my thermistortables.h and pointed to it in my Configuration.h , but still had this same undertemp problem.)

So I figured it must be a problem with the thermistor and replaced it with a Honeywell 135-104LAG-J01 and updated (#define TEMP_SENSOR_0 7), but I still had a problem with the temperature of about the same magnitude (i.e. actual temperature is around 220C, but set/reported temperature is 340C).

Thanks for any suggestions!

Dan


Here is Marlin telling me the temperature - the extruder pushes fillament reliably at this reported temperature, but it still feels like I could push the temp a little higher.
< 10:12:00 PM: ok T:341.2 /522.0 B:27.0 /5.0 @:99 B@:0
< 10:12:02 PM: ok T:340.6 /522.0 B:26.9 /5.0 @:120 B@:0
< 10:12:03 PM: ok T:341.2 /522.0 B:26.9 /5.0 @:98 B@:0
< 10:12:04 PM: ok T:340.9 /522.0 B:26.9 /5.0 @:108 B@:0
< 10:12:05 PM: ok T:341.2 /522.0 B:26.9 /5.0 @:97 B@:0
< 10:12:06 PM: ok T:340.9 /522.0 B:26.9 /5.0 @:108 B@:0
< 10:12:07 PM: ok T:340.6 /522.0 B:26.9 /5.0 @:117 B@:0
< 10:12:08 PM: ok T:341.2 /522.0 B:26.9 /5.0 @:96 B@:0
< 10:12:09 PM: ok T:341.2 /522.0 B:26.9 /5.0 @:97 B@:0
< 10:12:10 PM: ok T:341.2 /522.0 B:26.8 /5.0 @:98 B@:0
< 10:12:11 PM: ok T:340.6 /522.0 B:26.9 /5.0 @:119 B@:0
< 10:12:12 PM: ok T:340.3 /522.0 B:26.9 /5.0 @:127 B@:0

You don't say what electronics you are using. If it is Melzi then some people make them with 10K resistors instead of 4K7. That would cause it to read high with a table intended for 4K7.

---

[www.hydraraptor.blogspot.com]

The electronics are a pre-assembled RAMPS 1.4 board that I purchased from Lulzbot (not sure of the ultimate origin) and a Tosduino MEGA 2560 (haven't had problems with any other aspect of this setup). Is it possible to accurately measure the (pull-down?) resistor while it is still in the board? Where would it be located?

OK, so I looked it up - for Thermistor 0 on the RAMPS board, its R7 - I measured the R7 resistor on the board and found resistance of 4.55K which seems pretty close.

The resistance of the honeywell thermistor at garage temp (19C) is 128K - not sure if that is too low or correct...

One other thing I was thinking about is that in the first case, I installed the stock thermistor with (teflon?) shrink tubing insulating it per the standard install instructions. The second time around, I used kapton tape to insulate the leads on the Honeywell thermistor and added a little thermal grease (antec) for better heat conductivity. I just tried to measure the resistance of a blob of the thermal grease on a 200K scale and got no conductance, so it seems that is not causing any problems.