Hotend temperature jumps?

Hey,

I worry a bit about my hotend temp jumping too much up and down. It should have a certain range correlated with the printspeed but I want to print a 185 but I´m getting reads from 157 to 196 which seems a bit much for my taste.

Whats should and average temp range look like?

That does seem like a lot too much variation, mine (36W 12V hotend, Marlin controlling) normally holds +/- 1.2C during printing. It might be worth having a check that all your wiring is solid, and seeing if the firmware has options for smoothing out readings?

---

Prussa i2 Comissioned 4-Nov-2013
Bitcoin : 1Jcrg5czBtNfKyS4sCjFAamJTcSYdTGk9h

Yeah I just experienced a really big drop, the wiring is the prime suspect.

I replaced the resistor and checked the wiring, its gotten better but I still have 10-20C drops. I also dont think its the PID control because it works fine in a range of +/- 1-2C for most of the time. Any other candidates I should check?

Are the drops a single-reading error? or are they persistant? if they're a single reading it might be interference that can be dealt with by increasing the averaging setting in the firmware. If it's an error that persists over multiple readings then I'm afraid I'm out of ideas and would have to start probing around with an oscilliscope to try to localise the fault.

---

Prussa i2 Comissioned 4-Nov-2013
Bitcoin : 1Jcrg5czBtNfKyS4sCjFAamJTcSYdTGk9h

Ill do a 1 hour test with repetier and take a look at it.

In general are the thermistor physically reliable over long times?

If you've picked a type with a suitable max-temp rating, then yes they're normally fairly good. When they do fail the failure mode is normally either a broken/failing wire (which would look a good candidate here but you say you've checked that?) or a gradual drift in thier value which wouldn't match your symptoms

Edited 1 time(s). Last edit at 11/25/2013 03:10AM by Azrael_UK.

---

Prussa i2 Comissioned 4-Nov-2013
Bitcoin : 1Jcrg5czBtNfKyS4sCjFAamJTcSYdTGk9h

i presume you already tuned your pid settings....

Its weird, my 1 hour test shows nothing out of the ordinary, but in print it starts getting weird. Im printing with Reptier now to shed some light on this.

Is there a guide somewhere for PID control with sprinter?



Update:
As soon as I start printing the temp jumps again, and after some minutes is drops completely. I disabled the fan because it blows on the nozzle, that could explain the normal variation +-10C but not the total failures. Physically its fine, Im getting 12V on the pins, resistor has the right resistance, the wires are clear.

Could that be caused by having not enough power? I recently got a stronger extruder motor and I have only 5amps plus a fan. Could there be an automatic shutdown somewhere in the firmware?

Edited 2 time(s). Last edit at 11/27/2013 05:33AM by sungod3k.

Your last post pretty much confirms suspicion that the problem is due to interference.
Route your sensor cables away from motor/heater cables, that should fix the problem.
/Andreas

I put the motor and the fan cable out of the way and didnt have a complete failure so far and I saw some smooth pattern for some short while, but mainly i get a jump range of +-10C (plus one -40C jump for a second).
The cables for the resistor and thermistor go through the same hose and I havent changed them since I build the machine, so im not sure how they are now interfering. On the other hand I have a big strand of half the cables running down my frame along the Z axis (i have a closed steel tube as frame). Should I start replacing the cables with isolated ones?

it seems like the cable strand was the bad boy. I had taped them together and now that they are freed up I had a rather smooth temp curve. I route the thermistor cable a completely different way next time I open the frame completely.

Thanks for the insight andreas

Update: I bought shielded cable for thermistor today and now its perfect.

Edited 1 time(s). Last edit at 11/30/2013 09:17AM by sungod3k.