Extruder temperature reading inacuracy

My printer with the SOD-27 style axial leaded thermistor was reading about 20 degrees low at working temperature, so a set temperature of 190C was giving more like 210C. I had a miniature PT100 sensor that I could mounted on the heater block for actual readings. I saw someone else on the forum mention a similar error, but can’t find that posting now.

The first thing I did was set the cold extrusion limit lower in Marlin, but even then I was hitting the limit when performing filament tests.

I had a stand alone temperature controller to hand, so today I hooked it up to the PT100 and to the heater via a solid state relay and a transformer. Using this I was able to get thermistor readings at 25C, 160C and 220C to feed them into the thermistor table generator available from; thingiverse.com/thing:103668

I added a “table 8” in thermistortables.h and changed the extruder sensor to type 8 in configuration.h

Now I have readings between the Repetier and the pt100 that agree within a degree or so.

Maybe I had a rogue thermistor, or perhaps it really doesn’t match table 5 well.

Just in case, I’ll post my table here;

{ 488 , 300 },
{ 519 , 295 },
{ 552 , 290 },
{ 589 , 285 },
{ 628 , 280 },
{ 670 , 275 },
{ 717 , 270 },
{ 767 , 265 },
{ 822 , 260 },
{ 881 , 255 },
{ 947 , 250 },
{ 1018 , 245 },
{ 1096 , 240 },
{ 1181 , 235 },
{ 1274 , 230 },
{ 1376 , 225 },
{ 1488 , 220 },
{ 1611 , 215 },
{ 1745 , 210 },
{ 1893 , 205 },
{ 2055 , 200 },
{ 2232 , 195 },
{ 2427 , 190 },
{ 2640 , 185 },
{ 2874 , 180 },
{ 3129 , 175 },
{ 3407 , 170 },
{ 3710 , 165 },
{ 4040 , 160 },
{ 4396 , 155 },
{ 4780 , 150 },
{ 5193 , 145 },
{ 5633 , 140 },
{ 6102 , 135 },
{ 6597 , 130 },
{ 7115 , 125 },
{ 7655 , 120 },
{ 8212 , 115 },
{ 8781 , 110 },
{ 9357 , 105 },
{ 9934 , 100 },
{ 10506 , 95 },
{ 11066 , 90 },
{ 11608 , 85 },
{ 12128 , 80 },
{ 12619 , 75 },
{ 13078 , 70 },
{ 13503 , 65 },
{ 13891 , 60 },
{ 14242 , 55 },
{ 14556 , 50 },
{ 14834 , 45 },
{ 15078 , 40 },
{ 15290 , 35 },
{ 15473 , 30 },
{ 15629 , 25 },
{ 15762 , 20 },
{ 15874 , 15 },
{ 15967 , 10 },
{ 16044 , 5 },
{ 16108 , 0 },

In the newest original Marlin software there is support for PT100 resistors inside the file thermistortables.h.
I just copied those lines of code into the thermistortables.h coming from Velleman.

In Configuration.h I set #define TEMP_SENSOR_0 1047
This is actually a PT1000 for the extruder using the standard 4k7 pull up resistor on the controller board.
I recommend using a PT1000 instead of a PT100 to have a higher voltage change.

The PT1000 can be easily attached to the extruder using the standard screw and its washer.
PT1000 resistors have normed linear temperature curves so they can be optained from the prefered manufacturer/distributor.

I had bad printing results with my K8200. I exchanged the NTC’s a couple of times. First I had sucess but it did not last very long.
Comparing the K8200 LCD readout with a thermocouple confirmed me that somthing does not work well.

Since I use a PT1000 instead of the standard NTC, my K8200 prints reliable. I therefore recommend to use PT1000 and not the standard NTC.
I guess the problem is that the standard NTC not having a linear temperature curve is senitive at ambient temperature but not sensitive at filament melting temperatures.

[quote=“linurs”
Since I use a PT1000 instead of the standard NTC, my K8200 prints reliable. I therefore recommend to use PT1000 and not the standard NTC.
I guess the problem is that the standard NTC not having a linear temperature curve is senitive at ambient temperature but not sensitive at filament melting temperatures.[/quote]

That is why i mounted a thermocouple type K for both hotend and heatbed. :wink:

A 20 degree error was what I was getting too and I generated a new table.

Would any of you guys consider making an instructable for doing this? This seems like it might be a problem that could occur on several printers, and you probably would not even know the problem is there.

Cool to see my post here in the forum :wink:
I have done 1 year 3d printing with PT1000 resistors and I had no more temperature issues.
PLA works without any hassle.

I added my experiences with the K8200 to my page

http://www.linurs.org/print3d/Printing3d.html#K8200OrReprap3drag

I hope I will find time to explain more in details what I did with PT1000.
Currently I use my 3d printing time to print objects in ABS and have there some smaller issues.