Printing Polycarbonate - Max extruder temp (software regulated) question

[atilla_the_fun]

the cooling fans should stop the carriage from warping.

Can't think of any reason you'd want to go above 280.
Pc prints at 260 and above.

From what I could gather about my Alchemist PC, the closer to 300c you get, the better your results - both for clarity and layer adhesion.


I took some photos of the motherboard just to understand the temperature sensing circuit. It looks like it uses the TI ADS1118 temperature IC. The circuit itself is right out of the TI spec sheet figure 57.

[Graham]

Yes, the ADS1118 circuit is right out of the data sheet and that would be fine except they made one BIG mistake.
The thermocouples are connected to terminals on the wiring board at the extruders and then copper wire is run from there to the ADS1118. This means that the cold junction to be used in correcting the temperature starts at this interface. The ADS1118 is using its internal temperature sensor for the cold junction temperature.

We do not know what FF is doing in the software (they may have put in some kind of fudge factor) but I do know (have measured) that as the extruder area heats up, the reported temperature goes down. This error is directly related to the difference in temperature between the ADS1118 and the extruder wiring board.

In my testing, I found that when extruder temperatures were at or below 230°C, the error was not significant (<2°) and constant once things had warmed up. But when the extruder temperatures were set higher, the error was significant (8-10°) after about 30-45 minutes of printing and KEPT CLIMBING. These tests were done with the door closed but the top off as I have seen at least one totally melted extruder carriage.

What happens with this error is as the cold junction temperature increases, the temperature reported to the processor goes down so it increases the heater on time. Theoretically it is possible that this could lead to a runaway situation where the processor kept the heater full on trying to get to temperature, not knowing that it was already too high.

The best work around is to replace the thermocouples with those from the Creator Pro and run the thermocouple leads directly to the connector on the main board.
Graham

[atilla_the_fun]

I agree, I realized that the disconnect at the wiring board could cause an issue, even with minimal research into the subject (which is curious for FF engineer(s?) ).

The data sheet for the ADS1118 does show a minimal error due to cold-junction temperature wander, but that was at 25C.

What effects do you think I would see after replacing the thermocouple with the set from the creator pro? More accurate temperatures, sure - but whats the difference between 240 and 250 when printing ABS. For PC, a higher temperature would actually be good.

[Graham]

Yes, I am surprised the FF made that mistake. I suspect that it was a case of form over function and easier maintenance. It is nice that the heaters, fans and thermocouples can all be changed easily without chasing long wires through the printer.

I actually made a different fix to mine. My Thermocouples still go to the wiring board but I connected thermocouple wire at the same screw as the thermocouple leads and ran those wires through the loom to the mother board. Because these connections are at the same temperature, there is almost no error introduced by having a connection in the thermocouple wiring.

That error in the data sheet refers to the temperature of the ADS1118. In our case when we print above 230 or so, the cold junction temperature approaches 40 (at the extruder wiring board) but the ADS1118 is closer to 30 depending on the ambient temperature. That means that the actual temperature that the extruder is maintaining is actually 10° higher than indicated.

The big benefit of fixing the thermocouple wiring is that you will KNOW what the REAL extruding temperature is and there is no chance of it running away and destroying your extruder or worse. As I said before, this risk is minimal if we stick to the suggested 230°. I think one reason FF may be slow to "permit" other filaments (nylon, PC, etc. is because of this risk.

Graham

With ABS the only real difference between 240 and 250 is that the 250 is usually too hot and you would not know it is 250.
Certainly when you get up near 250 the ABS carriage is also getting close to or above 50 even with the top off. With the top on, I think there is a real risk of warping it. I warped mine in the first week I had it before I knew about the thermocouples.

For higher temperatures, I think an aluminum carriage is the only way to go but even at 230/240 there is risk for long jobs in cooler ambient temperatures. The reason the risk increases in cooler temperatures is because the extruders still heat up but the ADS1118 will be much cooler this increasing the error.

[Graham]

Just went back and read your first post again. I see you have the all metal hot end. Is it the MicroSwiss or Carls? I made my own but still want to make an aluminum carriage before I go to temperatures as high as you.
Graham

[atilla_the_fun]

I'm curious to know at which temperature ABS warps. The glass temp (like 110C) ? Although I haven't checked the air temperature after a long night's print (I've done ABS prints at 245C for as long as 19hrs - enterprise model =) ) the other extruder tends to hover around 45C or so. Since its physically connected to the working extruder I would assume it to be close to (and greater than) than the carriage temperature. Those rear fans do work to exhaust heat.

I've installed the MicroSwiss kit, which was nice. I emailed him by chance and found out he was offering them. Its a guy just north of me (twin cities). I did add thermal paste to the fan heatsinks when mounting to the aluminum block.

[Graham]

I hesitated replying to this as I am by no means a thermoplastic expert but here is what I think.
I understand that the glass point for common ABS filament is around 105°C but it does vary significantly among manufacturers depending on the blend of A, B, and S

ABS does not warp at a specific temperature. If you uniformly heated an ABS part and then allowed it to cool slowly and uniformly, it would not warp. ABS does, however, shrink when cooled... as much as 1-3%. The problem we face with our printing results from adding a hot layer onto already cooled layers. As the hot layer cools and shrinks it pulls on the layers below it, tending to pull them up and in. This results in the warping and pulling up from the build plate. Because we used a HBP, the first few layers are kept warmer so the shrinking/warping tends to occur a little higher up in the model.

There is no secret in how to prevent this, but more an inability for hobby type printers to manage this effectively. Stratasys along with all their closed patents for managing the chamber/model temperatures while extruding has no problem with warping. The key is to keep the entire model at a high temperature somewhere below the glass point. I believe the optimum chamber temperature is around 80°C.
In our machines, this is not easy. The mechanics of the extruder, the yet to be extruded filament cannot be kept at this high a temperature if we expect them to continue to function for long.

I believe that we can obtain satisfactory results in minimize negative impact on our extruders if we keep the chamber enclosed and get the temperature up to around 50°C. We also must eliminate all drafts e.g. the fans exhausting hot air and drawing in cold air and the extra nozzle cooling fans that some have added to extruders for PLA, etc.

Now back to the Dreamer... If the chamber temperature is 50°C, but the main board is 30°C, there will be a 20°C error in the measured and displayed temperatures assuming that the extruder wiring board is also at 50°C. BUT a 50°C chamber temperature will also reduce the cooling of the cold end, the stepper motors, etc., so it is likely that we will still see increasing temperatures (and error) in that critical area and still some tendency to runaway. My other concern is that I don't believe the plastic case and ABS parts will last long if they are constantly exposed to 50°C plus temperatures.

I see the thermocouple fix as critical for any attempt at higher temperatures. Once that is done, I think that we cautiously increase chamber and extruding temperatures and test where the upper limits are. Without the thermocouple fix, we have no real idea what the temperatures really are at the extruder other than to be sure they are higher than the displayed value.

I print a lot of ABS and as long as my ambient temperature is around 30°C and I keep the doors closed an eliminate drafts, I have had reasonable results. I've only had to scrap one model due to warping and that was when I tried to print a 4" x 6" thin walled box with an ambient temperature close to 16°C
I have already meted my first X carriage so I know it is easy enough to get extruder temperatures too high with the current software.
Graham

[curious aardvark]

okay just looked at the polymaker label on the roll of polycarbonate they gave me.
states print temp 255-280.
I know this isn't always a great guide for all manufacturers - but I have found with polymaker that the termps they state are the temps that work best.


@ graham

and as long as my ambient temperature is around 30°C

You're saying the temperature in your workshop/room is 30C !
Holy crap dude that's hot.
I'm currently looking at 16.8 and considering it nice and toasty lol