Hi all. Just finished the installation of my new Maslow and am working on a “first project”. I did the Bar/Bolt design stand. After calibration, I did my first cuts. Everything in my piano stand project is basically square cuts.
I’m using Inkscape to create the SVG, converting it to Gcode in JSCut and the cut essentially went off without a hitch… Except that the pieces aren’t square, not accurate and I’m essentially butchering a sheet of plywood.
They are really not close to being accurate. But, Close enough I can salvage them as sides. So, I figured I’d do a simple rectangle. 52.5 inches by 12". Started up the cut, and it went to the correct start point. After moving to the left approximately 24", it was already running up the board and off a horizontal cut line by over 1".
What am I doing wrong? How do I make this more accurate? I have read success where people cut smaller parts at the center of the board, but that limits my ability to do bigger projects?
That cut looks like you skipped a chain tooth or two. I had a cut do that on Saturday. It was cutting in the upper right corner and the slack side must have folder over. How is your chain-slack management?
Using Ground Control. Chain slack is the bungees from the kit. I watched the chains and didn’t see any skipping. I’m watching close because in my first cut, I had the pin break from the chain and the router go flying. Found forum posts about making sure the pin goes to chain links that are fully in tact.
If you hadn’t seen it yet. Once you get calibrated and centered on 0,0 mark the top tooth on the motors and the link on that tooth.
It can help identify if the chain skipped and realign/re-calibrate if you have a problem.
Also, a lot of folks went to a pulley/weight system for slack management or a single screen door spring replacing the bungie
Welcome to the forums
Text from post is below. can’t figgure out how to repost video from facebook
Here is what I did, hopefully this will help someone going through the same issues.
Set gears to 12 o’clock.
Made a long compass and drew 3 lines at different lengths from the sprocket, to roughly the center of the spoil board. Repeated on the other sprocket, then drew a line through the intersecting points. This gave me a perfect vertical centre line.
Marked half way up my centre then created a rectangle 800mm high by 1600mm wide in pen on my spoil board.
Entered my measured rotational radius and link parameters in the settings, along with distance between motors.
Ensured my router was central on the sled. Mine was 5mm off.
With the router turned off, I used GC to manually move the tip of router to the center of the spoil board. Then selected “manually measure chains” in settings.
The fun part… this took hours, using GC I entered 400mm and moved the router up. Then i was able to see if it landed on the top line of the rectangle. Sure enough It was way out. Manipulating the motor spacing distance in the settings, I repeated this until It was bang on. I recorded on paper what each setting resulted in. Next I selected 800mm and moved the router to the left, again using the drawn rectangle to visually see accuracy. I then altered the chain length adjustment percentage 0.05 at a time until I was accurate.
I haven’t seen that in a looooong time, but feel with you . Have broken a lot of hardware like this.
Can you give some info about the firmware version and GC version?
Seems GC is not maintained currently. Webcontrol is the main direction for the moment with older hardware. Or investing in a Due the softeware would be Makerverse, a clone of CNCjs.
I really hope you where there to stop it in time.
Upolad the g-code to check to there is something unusual in there.
No need. I’m done. This shouldn’t be this complicated. Done calibration a number of times. The chain breaking and router swinging while running was my first sign that I should quit. This was so stupid to happen, I can’t even begin… My laptop went to sleep while I was marking the rectangle on my frame board. In trying to wake it up, I apparently clicked the Move up 4 inches, which was my last operation to get the sled out of the way. I noticed it moving, and clicked stop, and it wouldn’t. It froze. then gave me errors that I couldn’t clear. I had to stand there and watch it break.
So the ring and all brackets are toast. It also pulled the motors from my top of the frame, and it looks like that board is buckled slightly.
I’m done. Gonna package up and sell what I can. This was a failed experiment, couple weeks in the trash.
I should also add that I just bought this and followed the website’s documentation on setup and software installation. It is disheartening to see stuff like Ground Control not recommended in forum posts. Everything seems so out dated, especially for a person getting into the Maslow now.
I invested in a kickstarter 4 year back, knowing it was in development. Still in development. But getting better and more secure. My doubt is that you either not used the latest software or you have found a bug that no one has found before. The community lives from the feedback and the development thrives by it.
If you really give up, just leave us the information how this happened, please.
things now are 100x better than they were when the project was first launched. things are cheaper and work better. Ground control works fine if you want to use that. There are youtube videos that show one how to change over to webcontrol easily if that is what you wnat to do latter… edited to add. Opps , saw the bent ring. Well I’m sure if you contacted the place that sold it they would help you out. The feet can probabaly be bent back in place, not sure if the ring could be though.
Ground Control 1.26. Same with the Firmware apparently.
Not sure anything truly failed as a result of a bug, but I think this was user error, that I couldn’t recover from. Everything was in motion when I realized the sled was going to keep going up. Stopping didn’t do anything, maybe it was from the laptop having slept.