Extruder Tempature Where do I measure it????

Hello All,

I am just confused now, where do I want to measure the extruder temperature to get a calibration? 3MM PLA. I have been using 185 but parts have literally started melting halfway through the build, they just sag down. So I am going with this is too hot, but what is a good way to figure out the correct nozzle temperature?

I suspect there should be a number like 180 for the plastic as the tip, so I can put a thermometer in and see what setting gives me that. I have a IR Temp gun, a thermocouple temp reader and some nice meat thermometers what would be an ideal temp for this?


Thank you for your time and help,

Are these small objects? If the object sags after being extruded it probably needs a fan or to be extruded slower or build several at a time.

Extruding PLA too hot causes it to come out like syrup and not form filaments.

I measure it inside the barrel with a thermocouple before I put any plastic through it. It is a bit difficult afterwards.

---

[www.hydraraptor.blogspot.com]

Following on from the original question - how stable would you expect the extruder temperature to be? I have just finished my Prusa and although I cannot try printing yet (one stepper driver was dead on arrival and I'm waiting for a replacement) I have tried heating the tip and extruding some PLA. Using Pronterface and setting the tip temperature to be 175C, the actual temperature seems to vary between about 170C and 185C - there is noticeable overshoot past the set temperature, more so when heating. Is this normal? Acceptable? I do seem to get a very even fine filament of hot PLA appearing.

I am using Sprinter on a Sanguinololu; looking at the code it seems to have some beta code for a PID controller. Is it worth enabling this?

I don't have any way to measure the "real" tip temperature so I am just relying on the thermistor calibration at the moment.

Any feedback or pointers to appropriate info much appreciated!

Thanks,

Hi Nealeb,


I did as nophead suggests and measured inside the tip using a thermocouple on my multimeter. In my case it was after I had previous put plastic through it, so was a pain, but as my probes diameter is similar to the filament diameter I used it to manually extrude the remaining plastic (I did this slowly and carefully, constantly pulling the thermocouple back out and cleaning it off with a acetone soaked rag.

Then I spent about a day heating the tip to various temperatures at 5 degree increments (in printrun) and noting the temperature on the mm. This appeared to show that the calibration of my thermistor wasn't that great.

I also followed up on a suggestion made on the "broken toaster" blog about fine tuning the thermistor tables around the temperatures that are important to you. For example if you are looking for a temp of around 170C then having lookups in your table that are 5 or 10 degrees apart in this region won't help the temp to stabilise. On the other hand if you have values around 2.5 degrees apart then your swing should only be around 2.5 degrees.

I wrote up about this on my blog, but I don't know how good the article is, so if you read it, feedback will be welcome. http://blogger.kritzinger.net/2012/04/recalculating-thermistor-tables-more.html

The distance between table entries shouldn't affect the swing, only the accuracy, because my understanding is the firmware interpolates between entries.

170 - 185 is far too much swing, it needs to stay within about 5C.

---

[www.hydraraptor.blogspot.com]

Quote

The distance between table entries shouldn't affect the swing, only the accuracy, because my understanding is the firmware interpolates between entries.

This understanding is correct.

-- a firmware developer

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

OK - I shall try turning on the PID controller option in Sprinter and see if that stabilises the temp a bit better. I am using a fairly chunky wirewound resistor as the heating element and I suspect that there is a bit too much thermal inertia in the system - PID control might help, I think.

Without any other temp measuring kit, I shall have to suck it and see as far as absolute temp is concerned but if I can get a more stable heater block temp then that will give a better starting point.

Edited 1 time(s). Last edit at 06/12/2012 06:37AM by Nealeb.

Traumflug Wrote:
-------------------------------------------------------
> The distance between table entries shouldn't
> affect the swing, only the accuracy, because my
> understanding is the firmware interpolates between
> entries.
>
> This understanding is correct.
>
> -- a firmware developer


nophead/traumflug,

Thanks for clarifying. I know used the word swing, but I was really meaning accuracy. (This is what comes from trying to blat out a comment while eating breakfast and not thinking it through fully). IMHO With the table values being far from linear, having a wide gap in the zone where you are trying to focus means that the interpolation (which I didn't know it did, so thanks) may well zone in on the incorrect value. (e.g. it focusses on 170 when you want 172). When I was doing the exercise on my own machine I found that some of the tables had very large gaps considering that people are often tweaking temperatures by quite small amounts.

They are a range of sizes. Nothing I would say is particularly small. I had the fan thought first but there has to be a limit. I have a 40mm fan with duct on the backside, and a 50mm fan on the front I started with them 100% on. But now I let slic3r pick the speed with minimum at 50% max 100%.

When I say melting I mean like a pile of molten goo on top. Just a giant blob bring pushed around by the nozzle. It seems worse based on height. The taller they are the more it happens. I can't figure out if it's because taller prints take longer or if it's only related to the height. I have triple checked my Z and it is with in .001in for accuracy. Any suggestions?

Thank you for your time and help.