Problem With Temperatures

OK This may be a slightly long post. I will do my best to get to the point.

Preheated my hot end, didnt realize the thermistor was not installed correctly. The hot end got very hot, spluttering boiling PLA everywhere. No big deal, I replaced heat break, nozzle, heat cartridge and thermistor. All with the same (but new) parts, thinking it would be good as new. Since then I am having under extrusion and skipping e-steps. It seems to me that the hot end it not getting as hot as marlin thinks it is. I can normally manually push PLA through the nozzle without issue anywhere above 140-160°. Now it takes some force at these temperatures. The only solution i have found is to raise the temperature to 210, I have never needed to print as high as this, normally printing rigid ink @ 180-190.

So have have tried 3 different thermistors, and even tried different thermistor profiles in marlin. All 3 thermistors report the same temp both hot and cold (reports ambient room temp correctly across all parts tested)

The only part I have not replaced is the mega/ramps boards. I have been told in another forum that a blown voltage regulator the mega board can result in incorrect thermistor readings.

I know its a long post, but I am truly out of ideas, could it be that the mega board is wrong?

When you replaced the thermistor how sure are you its the same? Did you buy it from a reputable company or was it ebay/Ali? It could be a different thermistor and therefore giving different readings when used with the thermistor table or b value or whatever your firmware uses, which is for the old one.

If you're using marlin try a few different thermistor tables see if any get you closer, use a K type thermocouple probe to check the real temperature or if using an ir probe colour the side of your heater block with black sharpie or heat proof paint to get a good reading.

If your firmware uses b values there are online calculators which can calculate them for you if you give the resistance at two known temperatures.

Edited 1 time(s). Last edit at 12/01/2017 03:48PM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Whether or not your thermistor is accurate really isn't all that important, so long as it is repeatable.

If it was me, I'd print a temperature tower, choose the best temperature, and get on with printing.

That's certainly a very pragmatic and sensible idea and surely the fastest way to get things working again.

Although I do want the temperature shown on the display to be the temperature that the filament actually is heated to, or as close to this as possible.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Is the nozle contacting the lower face of the heater block?
If its in mid air then you can only transfer heat through the threaded part of the nozle so the tip temperature could be a lot cooler than the heater block.
This is also affected by the different nozle designs available where the larger hex on some would enable a better heat path. Small hex nozles may run colder.
IMHO. It takes a bit of care, trial and error to get a good hot end assembly.

The nozzle should look like this for a v6 setup:


If the nozzle were screwed all the way into the heater block then it is impossible to tighten it sufficiently to the heatbreak to prevent leaking material and possibly jams, as it would tighten and bind on the heaterblock, and there are no flats on the heatbreak to hold to apply enough torque to tighten the heatbreak onto the nozzle instead of nozzle onto heatbreak (makes me wonder why not but that's another thread)..

Or do you mean something else?

Edited 1 time(s). Last edit at 12/04/2017 08:30AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
DjDemonD
The nozzle should look like this for a v6 setup:
[attachment 100189 hotend_wired_4web-1000x1000-2_grande.jpg]

Nice photo, but I would definitely not want it it look like that! The crimp ferrules on the thermistor wires are in danger of touching the heater block. All it then needs is an internal short in the heater cartridge between the element and the case (such shorts are not uncommon), and you will feed 12V or 24V into the thermistor electronics. Unless the controller electronics is specifically designed to handle this (most designs are not), this is likely to damage the electronics.

Edited 1 time(s). Last edit at 12/04/2017 11:18AM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I take your point but it was the amount of nozzle insertion that was being demonstrated, best have a word with filastruder then as it came off their website [www.filastruder.com] and to be fair these things are cartridge based now anyway.

Edited 1 time(s). Last edit at 12/04/2017 11:22AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
DjDemonD

If the nozzle were screwed all the way into the heater block then it is impossible to tighten it sufficiently to the heatbreak to prevent leaking material and possibly jams, as it would tighten and bind on the heaterblock, and there are no flats on the heatbreak to hold to apply enough torque to tighten the heatbreak onto the nozzle instead of nozzle onto heatbreak (makes me wonder why not but that's another thread)..

The correct way to assemble a hot-end is:
1. heat to operating temperature
2. Screw in the nozzle all the way.
3. Back out the nozzle a quarter to half turn
4. Screw in the heatbreak all the way, finger tight
5. Screw on the heat sink, finger tight
6. Tighten the nozzle against the heater block with a wrench

The idea is that the nozzle *should* bind against the bottom the heatbreak to prevent leaking material and possibly jams. The heatbreak is thin (to minimise conduction) and not strong, so applying torque to that is liable to break it.

FWIW, none of my nozzles has ever tightened all the way into the heater block. The thread on the nozzle just doesn't go all the way to the hex flats.

I tape (Kapton tape) the thermistor wires to the heater cartridge wires to prevent movement of the thermistor wires relative to the heater block, and the fatigue that would otherwise happen. I also wrap a few layers of tape around the heater wires and back around the heater block to prevent movement of those wires, and cable tie them to the heatsink or effector. This doesn't completely eliminate wire breaks and shorts around the heater block, but it does reduce their frequency.

That's broadly how I do it too, but the suggestion was the nozzle should be all the way into the heater block for thermal reasons which isn't true, and which would only work if the heatbreak could be screwed tightly to the nozzle, which isn't possible as there are no flats and it's too delicate.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

I also use the same process. And this gets the nozle as close to the block as possible.
I was making the point that if the nozle is something like 2mm from the block - like in the photo you will probably get different behaviour compared with if the nozle is close to the heater block. Even though the thermistor reports the same temperature.
So for the OP this could explain why he has to modify the temperature settings.
I have no issue with the v6 being setup as shown if that is the designated "correct way". However it is possible that those with a V6 like this will have higher set temperatures than those with the nozle closer but iIts not as simple as this because it depends on the internal design as well = how much nozle is inside the heater block.
Sorry for any confusion but the OP wanted suggestions for why his new hot end has different performance to the original.

I think you make a very good point, I apologise if I seemed overly critical, I just didn't want lots of newbies to screw their nozzle all the way in, as this does seem intuitively the right thing to do. And I agree if you can get that gap between the flange on the nozzle and heater block smaller then do so. But unless you are actually going to tighten it down I can't see it affecting thermal transfer to the nozzle by much. Maybe a smear of high temperature thermal grease in the gap might help?

I'd like to suggest the best way to measure the temperature in your hotend is the way chosen by the Deltaprintr mini hotend, with a thermistor designed to be retained in a recess in the nozzle itself:

This is not a photo of that hotend I couldn't find one that shows the thermistor clearly, and I'm not at home to photograph mine, but this scheme is the same, the thermistor is retained by a fibreglass (now silicone) sleeve and measures the nozzle temp. Also as shown is the novel heater design which wraps around the heater block and nozzle, you can see the recess in the nozzle for the thermistor. Surely this location for sampling the temperature is preferable to measuring the heater block temperature? The only issue with these hotends is the thermal capacity of the heater is lower, so rapid changes from slow details to fast infill can cause the temperature to fluctuate more than with a larger block.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
DjDemonD
...The only issue with these hotends is the thermal capacity of the heater is lower, so rapid changes from slow details to fast infill can cause the temperature to fluctuate more than with a larger block.

OTOH the lower thermal capacity and tighter coupling will reduce the dead time, which will allow for faster PID control. So it may actually be better, using the optimum PID parameters.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

You might be right, I have only used my mini hotend with Marlin so haven't had chance to see if RRF does better, the autotune in marlin struggles with it perhaps assumptions are made that are not such a good fit for this unit which heats extremely rapidly, and the thermistor supplied has no preconfigured table for marlin.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions