Temperature Sensor out-o-wack

So....

My temp sensor (100k thermistor) started freaking out. Once I got near target temp, it would jump readings + or - 10 degC.

I took the extruder apart, and re-soldered the wires. Yea! Problem should be fixed.

Nope.

I have played around with the temp and found it does fine at 185, as long as its not the target temp. Once the hotend get within 5~10 degrees of the target temp, it starts jumping again.

Any suggestions?

I'm currently in an engineering class so the sooner I can get it fixed, the better.

Thanks in advance,
-Nick

Sounds like a foobared ADC->Temperature table .

Or possibly, noise/crosstalk from turning the resistor of and on again - do you have the heater's legs isolated from the
heat block?

If that is the problem how do i fix it?

What do you mean by heater legs?

-Nick

Check voltage stability across resistor - thermistor voltage divider, also check Mcu GND - Thermistor GND voltage difference.

You may still have a loose wire. Tap the hot end, thermistor, and wires, see if you can get any temperature movement.

I mean the metal legs of the resistor. They shouldn't be touching the metal block.
Or do you have a heat cartridge instead?

What board are you using? The early Gen6 boards suffered from this problem due to an oscillation in the power circuit. Fixed by adding a 1nf cap acress R12.

Geoff

RAMPS 1.4.

If there was a short or the wires are touching the heating block, why would it sputter only when near target temp?

I have an idea of the problem but don't know how to fix it.

It only turns on near target temp, when the heater isn't constantly on... therefore i believe there is some talk between the heater and the thermistor, but I don't know where it could be.

Any ideas on where?

-Nick

Hi,

I am trying to make my home made/desigened gen6/gen7 like board work with my home made excalibur like hot end, and having the exact same problem.

I was glad to have found this thread, but I can't find the actual cure (sorry if it's just me). So, what was it eventually? On another thread I read it was "crosstalk" (http://forums.reprap.org/read.php?1,264712,267231#msg-267231). How did you fix it? Or was it the 1nf capacitor around R12 on the gen6 board?

In my case, there is another thing that appears to "disturb" the temperature reading. Obviously, I am not printing yet, just testing things piece by piece. When the teperature is close to target, but slightly below, and I move in any direction a few mm, there is a spike in the graph, and then back below target. It makes me think the issue might be power supply related. I use a computer AT PSU.

I attached a picture of the temperature graph showing the jumps, and circled the movement caused spikes with yellow marker.

Thanks in advance,
Miki.

is there a feature in marlin configuration files to allow for averaging of temp over several samples. is it called smoothing?, or something like that. I'm not able to access marlin this morning, but will try later on and find it.

Try plugging the hotend sensor into another location at the Ramps board (use the bed temp pins). This would isolate the problem from either the Ramps board, wiring or thermistor. Be sure not to heat the hotend while doing this otherwise it will continue to heat due to a wrong temp feedback. I has a couple of bad solder joints on my Ramps board causing all sorts of weird issues.

My 2 cents,
Greg

I plugged in the hot end to the hot bed connector, but it's doing the same there. I don't think it's the board. It's either the hot end itself, or I have a feeling it may be the power supply.

Is it a problem, if I have the heating and thermostat wires bundled together? (see picture attached)

I had another idea. If you look closely on the picture, you can see I used some heat transfer paste for the resistors and the thermostat. If I'm not mistaken it contains silver that is conductive. But it doesn't show by an ohm meter. The resistor shows 6ohms (two 12ohm resistors parallel), the thermostat around 100k, as expected. Perhaps on higher temperatures its conductive properties change. Although, why would it act up only around the target temperature, and why so prectedly (see graph in earlier post).

I think I got it! :-)

It is indeed the power supply. When the heating is applied, the voltage goes from 5.17V to 5.26V. Yes increases -- I guess the PSU "over-reacts" the higher load.

And this nominal 5V serves as the reference voltage to the thermostat resistor bridge. I think I will have to design a more stable reference voltage into my board.

other thing i was thinking if your using a resistor instead of a heater cartridge and have the wired solderd using leaded solder the melt temp of leaded solder is around 180 so it may be your wires are comming loose when getting up to temp causing wire to touch on and off giving heat up and down.

if this is the case ideally use a heater cartidge there cheap enough these days failing that get some high temp silver solder to connect the leads and cover in PTFE shrink wrap (if you need either drop me a PM and ill send some over)

Edited 1 time(s). Last edit at 11/20/2013 05:52PM by Enlightx.