Ah, yes, finally saw it, I haven’t scrolled that far… My bad.
I don’t understand shit on github unfortunately, this website is a total nightmare for the noobs. I think I’ll just wait for the official versions to be released, I really sucks when it comes to firmware and stuff. Basically I just have to get rid of this mintemp error and everything should be fine. Shouldn’t be a big deal though, I recall I already had this issue last year and managed to fix it.
Here is the relavent section, with the somewhat current settings from 1.1.8.
/** * Whenever an M104, M109, or M303 increases the target temperature, the * firmware will wait for the WATCH_TEMP_PERIOD to expire. If the temperature * hasn't increased by WATCH_TEMP_INCREASE degrees, the machine is halted and * requires a hard reset. This test restarts with any M104/M109/M303, but only * if the current temperature is far enough below the target for a reliable * test. * * If you get false positives for "Heating failed", increase WATCH_TEMP_PERIOD * and/or decrease WATCH_TEMP_INCREASE. WATCH_TEMP_INCREASE should not be set * below 2. */ #define WATCH_TEMP_PERIOD 20 // Seconds #define WATCH_TEMP_INCREASE 2 // Degrees Celsius
What puzzles me is that I get this error after the print is finished usually. Never during the print of during the warm up.
Interesting, I didn’t know this new function. Could have been useful before I’ve implemented the autolevel, but it wouldn’t have solved the problem of my bed not being flat anyways… I was actually doing a poor man’s babystepping by manually adjusting the Z height during the print, turning the screw with my hand until the layer height was “close enough”…
Thanks, I’m super happy with the autolevel now, I’m not sure I could get a better first layer. Still need to do more testing but it’s definitely going the right way.
Calibration is always a boring process where you print a lot of useless stuff in order to be able to better print useless stuff later. So why not making useful stuff in the first place?
I quickly draw a beer mug in onshape in order to help calibrating my printer. It allows me to see if the perimeters are well printed, adjust the bond between layer, test retraction and lattices, speed, supports and cooling setup. This part is both simple and complex to print (the handle is a bit tricky, with various kind of overhangs.
Here is the result:
During printing:
[attachment file=57322]
Right out of the printer:
[attachment file=57323]
After removing the supports:
[attachment file=57324]
The supports were fairly easy to remove, I think I got them almost right the first time.
The part looks a bit wobbly and the handle didn’t turn out great. I think the main reason here was that I had no part cooling at all. I tried using my air compressor gun to cool a few layers during printing, which apparently was making a big difference so I guess it should produce good results as soon as I 'll install one.
But now for the main question: is it actually useful? Can it actually fulfill its purpose as a liquid container?
Well the answer is…
[attachment file=57325]
YES!
I’ve filled it with water and let it on my bench overnight: no leak at all!
So, that’s pretty neat, I think that’s a very good start.
Now for the next step: part cooling. I would like to try a very different approach here. As I tested yesterday, using the air compressor looks very effective. Sure, you do not want to use it full blast, otherwise you’ll end up with a useless mess, but using reasonable pressure works very, very well. The good thing with the air compressor is that you can focus the air flow very precisely a few centimeters away from the nozzle, which helps tremendously while doing overhangs or for tiny print areas, like lattices structures. So my plan is to come up with some kind of solution using the air compressor instead of fans to cool the part right out of the nozzle.
The advantages are: probably less weight and vibrations at the nozzle, a more focused and quick air flow, possibility to cool down the plastic almost instantly. The only inconvenience I see for sure is the noise, it might be noisier than just fans.
Anyways, I’ll try to design this system in the next few days, might be interesting. I’m not really sure if this has been done already, so if anyone has recommendations or advices to give me, don’t be shy !
Interesting, thanks.
I wonder how it could perform correctly with such a puny little compressor motor, but eh.
My system as I plan it should be a bit more complex: I’ll use my shop compressor, so I’ll have to lower the pressure and find an easy way to adjust the remaining flow automatically during the print. I’ve already draw and printed a prototype of my cooling system, which was actually quite hard for me to design, since it has a lot of internal piping to direct the air. I’ll need also to create some kind of servo valve, to be able to adjust the flow continuously. Lucky for me, I already made a working prototype of such a valve a few month back for an exoskeleton project, so I can use it again with some minor modification (dual flow command to single flow, easy).
This is how it should look like, maybe, if it works:
[attachment file=“Nozzle cooler Assembly 1 - Google Chrome.jpg”]
[attachment file=“Nozzle cooler Assembly d1 - Google Chrome.jpg”]
[attachment file=“Nozzle cooler Part Studio 1 - Google Chrome.jpg”]
Those little pumps and coolers work really really well. You really don’t need that much cooling, just enough to get it out of the glass transition phase. Too much cooling and the plastic will shrink differentially and you’ll end up with bad prints.
Well, I tried… but I didn’t try it for long… It was just too damn noisy.
I couldn’t bear the noise more than 10 minutes, plus the compressor had to work a bit too much, so I decided to ditch this idea and go back to the classic fan instead.
I just added one fan now, the prints are way better already.
The printer seems to be quite reliable now. I just have to tweak the settings for infill a tiny bit and it should be good enough to be fully reliable. Now the first layer is just a formality, it never failed once since the autolevel has been enabled, that’s really great.
A few useful stuff I made with my MPCNC last weekend: some backets and boxes for the electronics of my electric motorbike. Those are the most complicated parts I’ve printed with this machine and they turned out usable and pretty much ok.
Most of the defects were actually due to the slicer which is sometimes a bit stupid, adding too much plastic in some spots where it shouldn’t be needed. Anyway, I’ll sand everything and paint it later so It doesn’t matter so much. Plus nothing will be visible once the tank and the plastics will be back on.
A few pics of the parts:
This bracket supports the motor controller. I was surprised it didn’t fail since I completely forgot to put supports but it turned out ok, just needed a bit of manual massaging to fix some areas:
[attachment file=60623]
[attachment file=60624]
That’s quite a big part, it took about 4-5 hours.
[attachment file=60625]
Installed on my bike:
[attachment file=60626]
Then I printed an enclosure for the battery management system:
[attachment file=60627]
[attachment file=60628]
And this is how it looks like now:
[attachment file=60629]
[attachment file=60630]
I’ve changed my battery for a new lithium one, extremely powerful (it came out from a giant pickup truck). Now the motorbike is really fast and accelerates way more than before. It can be compared to a 400cc right now, but I can tune it for 25% more power so it should be equivalent to a 5-600cc in a few weeks if everything goes well.
The MPCNC was really a great tool for this project, super convenient to print these kind of things. You need a part? Just spend 30-40 minutes to design it, put your SD card in the printer, do your stuff and come back a few hours later to pick up your brand new part that fits perfectly. That’s a game changer for me, again, thank you Ryan.
Your welcome Dui, thanks for sharing as always! As the project gets more popular there are more negative things being said, it is nice to hear positive things here from people that actually use the machine.
I have recently been using a larger nozzle for my prints, made me think “Dui would be proud”! HAHA
That battery connector you are using is the same one we used to use when I used to drive forklifts for Costco, something like a 9000lb battery, that connector could handle it. (In the early days we had to swap battery’s once per shift, not easy, by the time I left there the battery’s were able to last more than a full shift!)
They make a battery swapping lift. Next time I’m at the client that has one, I’ll take a picture of it. I’ll be out there in a couple weeks installing more cameras.
All bets are off when the battery comes off the dolly, I watched it happen two times to the same guy. Scary.
It’s real fun when they shove a new battery in, and get the wires caught.
Yeah, don’t bother with negative comments. The MPCNC has nothing to prove now, there are thousands of usable things being made with it everyday, the picture gallery and this very forum is a pretty good demonstration that it works very well when you do it right.
People who trash talk it are just losers who weren’t able to make it work, because they didn’t know what they were doing or were lazy to do their homework and research. They wouldn’t do shit even on a Tormach. XD
Now I got it 100% reliable as a giant 3D printer, I just drop the SD card, press print and move away to open a few beers, it works as flawlessly as my Delta printer and the only quality issues I still have are because of my slicer, not because of the machine itself.
Great! I really love the big nozzles, sure you lose a bit of quality but it’s fast and you get to see very well the printing process so it helps a lot for fine tuning any printer later. Show us some of your creations!
That’s actually a “small” connector… I have an other one which is at least twice bigger, but it was just too ridiculous… XD
Hi guys,
It’s been a while, but I’m still using my MPCNC, and it still works pretty well as a 3D printer. I now have consistently good results with it, and my prints come better and better each time.
Here is my last 3D print, made yesterday. It’s a bracket which is part of an assembly for a scientific research project of a friend of mine:
[attachment file=65980]
[attachment file=65981]
[attachment file=65982]
It took about 4 hours to print, I could have gone faster but I wanted the best possible result.
It is almost perfect now, I just have to do a few stuff to improve:
-I’ll add a second extruder motor in parrallel with the main one: I’ve noticed that the print head moves a tiny little bit while printing, because it drags the wire from the spool. This seems to cause very slight imprecision and inconsistency of positioning among layers. The first motor will be next to the spool, and I’ll give enough slack between motor 1 and motor 2 so that the print head can travel anywhere over the surface freely. Since both motors should do exactly the same thing, the slack should be consistent all the time.
-I need to square my print head perfectly in all directions. I’ve noticed that this is actually the main factor to get right if you want perfect prints. When the head is perfectly square and the extrusion is set up properly, you could print the top layer with almost no visible mark. I managed to get it right once and it looked amazing.
-I need to modify my probing system so that it locks the print head in place during printing. Currently my print head needs to move because I use it to probe the table during the autolevel before each print. But once it starts printing I need to manually lock it firmly in placed so that it doesn’t move at all while printing. Problem is I need to be around to unscrew it at the end of the print, so I cannot leave the printer completely unattended, which is quite annoying.
On the positive side:
-The bead heating system works well enough to avoid warping, it’s pretty nice.
-The autolevel works flawlessly. I never had any issue with the first layer for a very long time.
-Printing quality is good enough for what I’m making, and speed is just fantastic.
Nice, I have been wondering what you have been up to. …Chasing perfection as always!
hey dui,
Going to use your water cooling instructable to mod my anet a8.
Ive got it in an enclosure to prevent warping at night but the other day the whole enclosure got too hot and jammed my e3d clone as the fan was just blowing hot air onto the heat sink. before that the prints were getting really good. Got a new hotend coming too and replacement parts for the old one. Going to try bigger nozzles too soon.
I found a massive improvement when I put my spools on 4 bearings in the dry box I had. I can believe that two extruder motors will really make a big difference.
Great to see you posting again. Exciting stuff.
Interesting. I had a similar problem and I fed the filamemt through some PTFE tube to the print head, and attached it somewhere on the frame. The distance of the tube is constant so the motion of the head won’t change the tension. I thought about a second motor. I wonder how close they will operate since the interface isn’t as precise as the belts. I look forward to seeing how it goes. It’s like a direct drive bowden, almost.
What Slicer do you use? I add a little code in as a post process to move the X, Y, and Z to specific coordinates after the print has finished so all i have to do is walk up and pull the print off the bed. Seems like this would be easy enough to do. I use Slic3r personally.
I’m using Kisslicer, and I can do that too if I include a post Gcode.
Problem is, I’m not entirely sure this will work fine all the time, since those coordinates might interfere if I’m printing really big objects and forgot to disable the function before or to edit the coordinates.
Plus I’d still have to be there to tighten the screw before any print, and not forget to untighten before a new print.
But it could be a temporary fix until I find a better solution, thanks.
I’ll need to redesign my print head assembly to be more rigid anyways, it’s working fine but I think it could be better if the extruder was closer to the tubes (shorter lever).
Next project is to print this thing:
[attachment file=66180]
[attachment file=66181]
It’s a new version of my laptop arm. Nothing was really wrong about the previous version, but now this one should be more rigid and much better printed, the other one looked kinda crappy because my printer wasn’t setup properly at that time. I’ve already printed a few parts and they came up beautiful!
