If you’re going the same differential IR sensor I’ve got, it can run in either an analog or digital mode. Analog gives a variable result as the probe gets nearer to the surface. Digital triggers at a consistent height (which will depend on the ir reflectivity of your surface). I needed digital and for some reason on my ramps board it would come up in analog mode - I think it has to do with how soon 5v was provided to the board during startup. By adding a small resistor I was able to ensure it always came up in digital from then on. Results have been rock solid since then.
I have the DC42 IR sensor. Probably the same one.
If the input is left floating, it comes up in analog mode. If there is a pullup resistor, then it comes up in digital mode. 5V or 3V3 doesn’t matter. (EDIT: Fixed that. I had it backwards!)
All irrelevant to this printer now, because I switched to a BLTouch clone. For the E3D hotend, I have a good mount solution for the IR sensor, but not for the Mk8 one. It’s too short to get a nice close mount.
Well, I ended up with a 36mm Y offset for the BLTouch anyway, so I probably could have managed the same with the IR sensor. Doesn’t matter though.
The other problem was that the BuildTak sheet that I put on the hotbed is very black, and reflects very little IR, so it was having a hard time sensing the surface, and had more variability than I liked. I have the aluminum heat spreader painted flat black on my other printer, and have a PEI sheet over that, which works well with the IR sensor on my main printer. It’s using Duet electronics, so it’s in analog mode there. I like it because I can set a fast dive speed, and it will slow down when it gets close.
Not sure that I like that BuildTak much though. Well, the prints certainly don’t lift while printing, it’s just that it’s really hard to get them off when it’s done.
Something has gone sideways here, and I’m not sure what.
I’ve started getting heating failures. I set the hotend temp to 250°C for PETg. The hotend heats fairly reliably to about 240° then starts getting weird, and eventually the LCD reports heating failed, and I have to reset.
I thought it might be the heater cartridge wiring, so I re-did that, added good crimped ferrules to the wiring. It doesn’t seem to help. It’s also seems very reliable and consistent up to the temperature where it goes nuts, though this is sometimes as low as 210°
The thermistor wiring looks a little suspect to me. The PTFE insulation on the thermistor right next to the block looks like it was a little over-crushed by the retention screw, so I’m wondering if it might be something there with heat expansion of the heat block and/or retention screw on the Mk8 extruder causing problems. Seems a little dodgy to me, but… maybe? I’d think that it wouldn’t bounce around near the same temperature if it was shorting, but go to extreme hot or cold.
The only other thing that I can think of is that maybe there’s something flaky with the MOSFET tot he hotend. I know that board has a bad pullup resistor on Z-MIN, maybe something else has gone, and I didn’t notice before on the Primo because I wasn’t using the heater circuits.
So… I have more thermistors, I’ll try swapping that out first. It’s a pain with the wiring sleeve to the hotend, but I’ll try taping the new thermistor to the wiring for the old one and pull it through. If changing the thermistor doesn’t work, I have another identical board here already, I’ll just have to flash the firmware to it. I was really hoping NOT to have to resort to that though. Maybe I should just do it anyway, given that I know this board has that pullup resistor issue.
I really don’t want to change the thermistor cartridge again. I do have a couple more 24V ones, if it comes to that, but since I have to take the throat off the block to get to the set screw, and it was such a pain to get just right, I don’t want to have to deal with that again. If I do that, I’m going to swap out for an all-metal hotend, and be done with the PTFE tube in the throat. Even if I never actually do try to print nylon or polycarbonate, it’s an option that I was trying to keep open with my main printer.
What bugs me most with this is that it started happening after getting successful prints off. It’s like the printer is pouting or something.
Okay, well, it’s neither the motherboard, nor the thermistor.
It’s oddly consistent, the temperature climbs to ~245° then drops back to about 240, then I get the “heating failed” message.
If I try to immediately go back to temperature again after reset, I will get the heating failed message, almost immediately. I can only try again once it drops below about 190°
After a few attempts, sometimes, it will come up to temperature. If that happens, I can apparently print OK.
I wonder if there’s anything in the firmware that I’ve done wrong, or if I have a hardware issue elsewhere. About the only thing left is the heater cartridge there. Well, I guess I’ll try that next…
Did you buy all of the thermistors together in bulk? Its possible that the high temperatures is causing it to react badly as they could be defective.
What board and thermistors are you using? I’m curious to see what kind of hardware you are using as I don’t have any experience with building a 3D printer, just using one.
Did you run the hot end PID tuning procedure?
The only other things I can think of is, make sure it is the right therm in the firmware. Also you have enough time set so it can get to that temp and stabilize, I think that is in watchdog settings. Oh, last thing make sure you set the max in the firmware higher to account for overshoot.
PTFE will not last very long at 250, at 210 I would get just under 6 months before it would either swell, burn, or crack. Pretty sure any time I printed anything at higher temps the PTFE died within days.
Nope. The first came with the Mk8 extruder, the replacement was one from a batch, where another is known good.
The board is a MKS Gen L v1.0, the thermistors are the base generic ones.
I did with the first board, but not with the replacement. Doesn’t seem to make a difference. I’ll run it again. I did use the “U” parameter, maybe I’ll just enter the specs myself.
I’d have suspected that, but it never gets up to temperature, let alone overshoot. The max temp is set in firmware for 260 to allow for overshoot.
I double check my thermistors now with boiling water. I had some that were entirely different from the listed spec once that cost me a proprietary heat break. At this altitude, water boils about 97°C so I consider anything from 95-100 acceptable precision. The firmware was listing 98°.
Yeah… I’ve been spoiled by all-metal hotends for a while. My PETg filament lists 240 as the recommended extrusion temperature, I have been getting the best results with 250 in my main printer, so I just kind of stuck with that. PLA prints cooler, which is a nice property.
Chances are pretty good that I’ll switch to all metal at some point, probably sooner than later. The PTFE lined throats are a hassle, even if printing cooler, they have limited lifespans, whereas the all-metal E3DV6 in my main printer hasn’t been touched over many spools of filament. (Maybe not the kind of duty cycle that happens in a print farm, but all the more reason to want low maintenance.)
Well, at this point, it’s pretty standard troubleshooting. I’ll try looking for an all metal hotend, and check those thermistors again. I’ve been using room temp and boiling water as data points, though this past couple of weeks, room temp has been somewhat higher than usual. Maybe I should use ice water, like when testing a thermocouple.
I usually print PETg at 230 and get good results. As for the heat failures, I too have wrestled with this with my MP3DP w/ MK8 hotend. I think what I ended up doing was running the PID tune at 250 I think, and having the part fan on at 100% the whole time. This will force the hot end to compensate for the cooling effect, and not give you thermal failures since it will have a wider margin when the fan isn’t on. Do you have a silicone sock, or cotton wrap on the heat block?
1 Like
I do have a silicone sock. I ran the PID tune at 240, since that seems to be what I can get it to reliably. I entered the parameters manually and saved them. Trying a print now to see if it works out.
1 Like
Well, seems to be working now. Maybe manually entering the PID parameters was what it needed, or maybe I incidentally fixed something else while poking around. Either way, seems good.
Prints look a little overextruded, with some stringing, so I’ll have to re-tube the extruder steps. It’s been a while since I’ve done anything at high resolution.
It’s been a while since I’ve done much at 0.1666667mm layers. There is some loss of detail where it looks a bit over extruded, I’m sure I can make it better. Maybe I forgot to set steps/mm on the new board.
4 Likes
So close. It looks pretty dang good if it weren’t for the blobs.
1 Like
My understanding of these thermistors is that they are relatively consistent, but absolutely pretty bad. So 240C on one printer may be 230C on another. Just because of variances. But if you go 10C higher, it will be within 0.5C of 10C difference.
If you get the wrong thermistor in the table, or if there is a funny resistor for the thermistor or something, you would have a similar problem, where you’re off by 20C, but 5V difference is still going to be close. They won’t show much difference at room temperature and I would have to look at the tables to see how much they change at 100C.
One thing I’ve always wanted to try for calibration is sugar. It melts between 175C and 190C. I have used it to check ovens before (setting a little cup of it at 350F for 30 mins, no melting, boost to 375F, melts). But I can’t figure out how to not make a mess with a printer. You would want to stick the nozzle in the sugar, I think, and see if it will stay solid at 175C. But you would also want to see it melt at 190C, and that would make a mess.
The inside looks pretty bad for stringing. Looks like a few PETg spiders have been busy in there… I think that it is overextrusion, not terrible, but it does build up. I probably need to re-tube retraction for the Mk8 nozzle as well, but this is standard printer troubleshooting at this point. The heat failures were worrying, I still have some things to run there while I tune the rest of the settings.
Yeah, that pretty much agrees with my understanding as well. They’re pretty good across a relative spread, but need some good base calibration if you wanted to say, use one as a room thermometer.
But if it were just in a little cup (say, a printed cup?) you could just replace the nozzle and toss the melted sugar cup, right?
1 Like