If taking your computer out of the equation doesn't work then you very well might be loosing steps because of overheating. Maybe add a heatsink and/or fan?

+1 to ARKtest. If it skips steps some time into the print it could be overheating motor drivers. Try swapping the X driver for the Y or checking the current settings of them. Maybe it's a comms/ computer issue as you suspect.
-a

Tech, any update?

I didn't get a lot of time to play over the weekend. I did print this though for my niece's b-day.


It prints in 5 parts. The hands, legs, skirt, body and head.

I did check temperatures on my components though. My steppers run a cool 22C to 25C. Of course they are attached to aluminum mounting brackets and then to the aluminum frame so I'm not surprised they are fairly cool. It helps my basement hovers around 15C at this time of year.

The driver chips on the Smoothieboard though run anywhere from 40C up to 50C. Not sure if that could be the issue. I'll need to research that. At the moment the board is not enclosed or actively cooled. I had it sitting on a cardboard box then moved it up in the air on top of a couple of test print parts. It helped with temps a little bit they still get over 40C. Maybe that's fine, I just don't know right now?

I didn't get any of the offset layers in the Elsa print. I'll give it a work out tonight. I'm going to start printing my Y-axis upgrade parts. Those are 15 to 20 hour prints. I'll run them from the SD card and maybe put a small fan blowing on the Smoothieboard and see if that heps.

Also, I picked up an air pump at the pet store Friday. Very similar to yours Arktest. I'll check out your 4 to 1 adapter tonight as well.

Maybe use that airpump to blow on the hot components of your smoothieboard if you don't have the pump set up yet. Just an idea. The print looks great.

So, I started my to print my new Y-axis slides last night. They are about 24hr long prints so this will be a good test for the layer issue. Started about 6pm last night so should be done about 6 tonight, with the first slide. I checked this morning before heading to work and the layers were spot on so far half way though the print.

After starting the print last night I was carefully monitoring the temperature of the motors and the drivers. The XY motors (I am using an HBot configuration) run at 20 to 25C. No heat problem there. The extruder motor running about 35C. Of course its bolted to plastic so doesn't really have anywhere to dissipate the heat to other than air. The Z motors are no problem running at 20C.

The drivers on the controller are where all the heat is. Not sure how accurate my thermometer is but after 30 mins of printing here's the readings:

X and Y drivers - 35 to 43C
Z drivers - 35 to 40C
Extruder Driver - 48 to 50C

This was with no cooling. The drivers are the Allegro A4982. Rated for -20C to 85C from what I can find online. I grabbed a 20 inch fan that was sitting around the basement and pointed it at the controller. After about 5 minutes the new readings were:

X and Y drivers - 25 to 28C
Z drivers - 20 to 22C
Extruder Driver - 35C

Definitely an improvement. The extruder seems to be working the hardest. Of course its a direct drive bowden setup so I kind of expected that.
I am also printing directly from the SD card. Pronterface choked on the file anyway so it was the only way run it.

I will have an update tonight the first print finishes, but if there is no Z layer offset issue with this then I will consider this solved.

I will design a nice spacious housing for my controller with active cooling around the drivers. I have setup a custom button in pronterface to play my gcode files from the SD card as long as I name them 1.gcode. That should suffice for the time being until I add an LCD display.

I hope the good results continue. It seems like you're really getting the hang of this

Thanks to you ARKtest ! Slowly but surely I'm getting the hang of it. I think tonight I'm going to print out your 4 to 1 connector and hook up the print cooling pump before I print out my second slide. It's probably not a good print to test it on, but I want to get it hooked up and running anyway. I'll just zip tie it place for now.

Here's a picture of my slide carriage:

No problem! the slide carriage looks cool. I'd be interested to see how it works.

Let me know if you need any help with the cooling pump.

I look forward to designing with you.

Thanks ARKtest, I'm looking forward to it as well.

Unfortunately it hasn't been a good couple of days with the printer. I'm trying to print out new carriages for my Y axis to use larger bearings and guide rods but I'm running into more issues. I have had 2 prints stop on me now. The first one stopped sometime after 12 hours and the second print stopped sometime after 15 hours. Here's a couple of pictures.


I have had a couple of other prints stop on me before but I know for a fact those were because of my computer.

The first fail for this print (the short one) I tried printing from the sdcard. I used the verbose setting so I could see the code on the computer screen. I left it connected then when I found the failed print there was a usb error. I just assumed it was because it was connected to the computer still.

For the second print I used the same gcode file, printing directly from the sdcard with no computer attached. All was fine, the print had run for 15 hours and looked good when I went to bed. I expected to find a completed print this morning and found another fail (the taller print).

Needless to say I am quite disappointed with this. I have a 20 x 20 inch print bed that if I utilize the majority of it I can see the printer easily running for 20+ hours at a time.

At least the offset layer issue seems to be gone since I started cooling the controller with a large fan.

I restarted the print this morning, changed some settings to print a thinner perimeter, rougher support and less infill. If this doesn't work I may have to try a different slicer. This was done with KISSlicer. It's a very good program and I've gotten my best results with it of any I have tried so far but it does create quite a large file of code, although that shouldn't be the problem.

If anyone has any other ideas for me to look into I would appreciate the suggestions.

Also, if anyone would find it useful to look at my gcode file I could host it.

That is awful the only thing I can think is if you where willing to sacrifice a computer to stay hooked up. I know that as long as my updater doesn't pop up I haven't had a problem so far.

Your prints do look good though.

If for some reason your printer just refuses to work for a while, you could always engineer the carriage in two pieces.

I am kind of at a loss for the moment. I've put a little something out on the Smoothie forum so maybe I'll get some information there. There's not a lot of activity though so I'm not really optimistic. I did have an older laptop that was going to be strictly for the printer, but my wife can't remember her log in so it's just a paper weight at the moment.

I am debating changing the mounts to 2 or 3 piece design and maybe gluing or screwing them together. But in the end I should be able to run the machine for that long so I'll work toward figuring that out first. Of course for a couple of pizza's I can have the tool room here at work make them for me, but it still doesn't solve the root issue .

I can't complain much about the print quality. Those are printed at .3mm layer height and I think they look damn good. My daughter wants a golden snitch printed up soon, maybe I'll try that at .1mm layer and see what I get. I'm also going to venture into ABS here soon.

At least I am not dead in the water. I can print shorter duration prints and still utilize about 5/8 of the printer bed. I am more of a half glass full kind of guy.

Well stay optimistic. I hope the more specifically purposed forum can help you more.

Good luck on the snitch, and don't let my wife know, she loves HP.

a little late to this thread but I recently had an issue with prints jumping off line. They would offset 2mm every few layers. Turns out my issue was due to CA on one of the rails for my Z axis. Cleaned the rail, and woot no more stepping.

Not sure if this is the fix for your issue but worth a shot. Hope you get it squared away soon.

Finally, a successful print (I'll try to post a pic later). Not exactly sure if it was chance or the different slicer settings, but it worked. I'll find out when I print the 2nd one. First things first though, the wife has requested 24 mini Yoda's to give to her class on May the 4th day. That'll keep the printer busy for the next couple of evenings then back to the upgrade parts.

@chris - Thanks for the reply. Beyond the cooling I did also clean all the rails and switched from lubing them with 3 in 1 oil to grease. Not only does everything move much smoother, I haven't had the offset issue since .

You should be sure to show us the army of Yoda's when that happens. Congrats on the print.

Thanks. I found a funny thing with my print when I got home tonight. I actually ran out of filament as it was finishing the last layer. That's perfect timing. At least something finally happened in my favor.

Nice. So the last layer not printing doesn't affect it does it?

I don't think so. Here's what it looks like

You can see the small area that didn't quite finish in the upper left hand corner. So I should be good.

I think you should be good.