Nozzle temp not correct

[DIY3D]

Hi All,

A few days ago I noticed strange nozzle temp behavior (see attachment)




I did a wigle wire check (using my multimeter with sound ) between thermistor connector and Rambo connector. I didn't find a broken wire.
Then I did a PID autotune using: Send: M303 E0 S210 C8. but that failed (error temp to high)
Next thing was measuring the resistance of the thermister and this is new for me... so, for sure I post my setup below:

- I disconnected the thermistor and used a simple voltage divider (5V power, 4.7k and thermistor in serial)
- The voltage over the thermistor was 4.775v and that means 100k for the thermistor (I saw changing the values when I heated it for short), so I think the thermistor is working well?


2 Questions for now:

- Is the above method correct?
- Anyone other suggestions to control whats wrong (maybe some measurements at the Ramboboard?)

Kind regards,
Marco

[Tim]

Is the above method correct?

It's correct for telling whether or not you have the right kind of thermistor or whether the thermistor has a definite open circuit or short. But the temperature readback is more or less constant over the long term; it looks like it is getting mis-reads in short spikes, which you would likely not see on a multimeter.

The quick diagnostic test is to swap the thermistor connectors on the RAMBo for the extruder and heated bed. Note that this is inherently dangerous: when the heater does not get feedback from the thermistor measuring its output, it will just keep on heating until something breaks. But it's good for diagnostics if you keep your hand on the kill switch. With the thermistors reversed, first turn on the bed heat and watch the graph of the extruder temperature (which is actually the bed heat now). If it appears to be reading the correct bed temperature, then everything's okay on the RAMBo side, if it's flaky, then something's wrong with the controller board. Turn power off before the heated bed kills itself (although that would take several minutes at least). Now try the opposite test, which is to turn on the extruder heater while watching the graph of the bed temperature (which is now reading the extruder temperature). If that appears to be correct, then the thermistor is okay. If that one is flaky, then you need a new thermistor (if there is a wire open or short, it's still probably easier to replace the thermistor than to try to find and fix the wire). Again, don't let the hot-end get over 100C or so before killing power. And make sure you swap the thermistor connections back again afterward.

[ednisley]

I didn't find a broken wire.

The symptoms show a problem between the thermistor and the RAMBo. The challenge is to find it...

You can measure the thermistor resistance directly, without conjuring a voltage divider, by unplugging its connector at the extruder and poking the meter leads directly into the connector. It should read about 100 kΩ and probably won't be intermittent, because fiddling with the wires & connectors will change the layout.

The temperature graph shows downward spikes, which suggests a broken thermistor lead: broken wire = very high resistance = very low temperature. The wires emerging from the thermistor are very delicate and easy to damage, so that's the first place to look; they shouldn't be loose or twisted around each other.

In all the recent problems of this sort, replacing the thermistor has resolved the problem...

[DIY3D]

Thank you guys for the quick reply and advices, I'll take a look again!

Kind regards,
Marco

[innkeeper]

Hi,

if all the above dosn't result in anything...i have a few other things you can start checking.

i've seen where noise in the power will cause this kind of issue. - in fact had to change out a dc-dc converter on another type of board for this problem just recently.
What happens is, first the noise spikes start sending false readings, then the board tries to compensate and causes larger swings of temp.
could be somthing as simple as a bad power supply or a loose power connection.

if you have an oscilloscope, take a look at the power and the power begin supplied to the thermister.

Also... and i hope this is not your issue....it can be the rambo board itself (personally had this issue)
as an experiment, you can try using one of the other temperature inputs, (in example changing the second extruder mosfet and its associated thermister inout and drivign it as the second hotend) this will let you see if the issue is only on one of the temperature ports.

[DIY3D]

Hi,

Yes I'm up and runnning again



Finally I found the problem,



... a bad connection on both wires at the connector, I could pull them out and unfortunately the NTC was also broken...



Friend of my gave me a few 100K NTC's and I decided first to make a kind of bypass.



...took a small alu tube...



... used a little Silicoset 158 (thanks Jeroen , heat shrink tube and soldered the wires ...


... this is a bypass (test setup),.................Bobby was monitoring the whole process.



... Cut the wires to the original length, I mounted a new molex connector, did a PID tune and now, happy printing again

Thanks for the advices and ideas here, nice too for those who have the same problems.

Kind regards,
Marco

[jsc]

That's a monster print you've got going there.

[DIY3D]

That's a monster print you've got going there.

Oh no, it's a little misleading, the object is 100 x 80 mm

Kind regards,

Marco