Dialing in Maslow's accuracy

[mrhall4]

All the cuts I’m discussing in this post we’re done in the center of work area.

Hey all,

I picked up the Maslow kit, had it running. But the vertical accuracy was off. Percision was good, all the lines were clean enough. But I was losing almost 3/8" in the vertical. Horizontal was bang on.

Then I fried my v1.2 shield due to a unattended jam. @Bee was a friggin lifesaver and FedEx overnighted me a TLE5206 v1 he had(couldn’t wait for the new one).

So now I’m back up and running. I start the calibration. I extend the chains. I hook the other motor. Retract. Bam. Chain pops off sprocket. Again. Same thing. Next time I left it too loose. Third time I got it to grab and hold.

It set my measurement at 3000.71. So then I tried to understand how the calibration works. If the Maslow received no info from motors. Correct me if I’m wrong @bar but does that mean this number is based on only the fact the chains didn’t pop off?

Not satisfied with that measurement. I 3D printed a tape measure holder with magnet that attaches over the motors axis. With a tape guide and center point on the other axis. My measurement was 3006.50.

With that I cut my sled again. Last sled was an 18" circle. That was more like 17 5/8 x 17 7/8. This new was was like 17 15/16 horizontal and 17 13/16 vertical!

We’re moving in the right direction!

So where do I go from here? Do I keep cutting sleds and measuring? What other settings can I mess with?

Any advice would be appreciated.

Thanks

mrhall4

[Dustcloud]

Yes you are moving in the right direction… a lot has been written on this subject… check out the following:

Let’s dial it in!

Just got my maslow set up. My 18” sled circle is 18 by 17.75 I know this is not a huge deal but I’d like to dial it in. All help is appreciated.

Also, to gain a better understanding about how all this works, check the following:

Accuracy Issues

Definition and the Calibration Model It’s important to make sure you use the correct language when describing your issues and the definitions of accuracy and precision are often confused. Accuracy is how close a measured value is to a desired value. You want to cut a 12-inch square and you get an 11-inch square, you accuracy is poor. Precision is how close multiple measured values are. You want to cut a 12-inch square and do so in 5 passes around the plywood and each cut perfectly overlaps t…

Hang in there! We are all in the same boat!

[mrhall4]

Thanks, ive read through most of the accuracy tagged posts. I certainly understand this is a repeat question. More of a bump. To see if anyone had anything new to drop.

Just trying to get involved with the community

[Dustcloud]

In my case, made a new sled following the “by hand” instructions, which improved things considerably. My chains do not match in length, so I purchased a long enough section that I can get 2 pieces from one lot, which hopefully will result in a matched pair. Since I am replacing the chain, I also raised and widened the top beam to 12’ using heavy duty super strut from HD. This weekend I will find out if all this helps.

Will post the results.

[bar]

@mrhall4

It set my measurement at 3000.71. So then I tried to understand how the calibration works. If the Maslow received no info from motors. Correct me if I’m wrong @bar but does that mean this number is based on only the fact the chains didn’t pop off?

Maslow does receive information from the motors through the encoders so it is using that information to make the measurement.

@mrhall4

I 3D printed a tape measure holder with magnet that attaches over the motors axis. With a tape guide and center point on the other axis. My measurement was 3006.50.

This is the big outstanding question right now. We know that the chain will consistently measure a few mm different than a tap measure, probably due to the accumulated play in the chains adding up, but we have not found a good way to correct for it that consistently improves accuracy.

[mrhall4]

Thanks @bar, this morning I cut a new sled.

This calibration is so frigging close in the middle of work area. 18in circle is perfect horizontal. But off 1/8 in the vertical.

So far I’m pretty happy with my results. Going to keep cutting new sleds. Saving my .ini files and ranking them based on performance. Then use the best one at any given time.

Thanks for the response.

[Willja67]

Here’s another question that I’m sure a solution to could improve the accuracy of my machine: my sprockets on the top motors have a little play in them. I have the set screw on the flat part of the motor shaft and tightened as much as i can by hand without fear of bending the Allen wrench but both sprockets will still rotate a degree or 2 each direction. I can’t help but think when trying to get one tooth perfectly vertical during calibration that my accuracy is suffering.

So the question is this: is the play normal? If it’s not any suggestions on getting rid of it?

[blurfl]

There is a small amount of gear lash in the motor box, but the fact that the weight of the sled exerts a constant pull in the same direction cancels it out. This assumes that the tension on the ‘slack’ end of the chain is less than the tension from the sled, though. Using a strong bungee cord or a large counterweight in the slack management setup could spoil that assumption and bring the gear lash into play.

[Willja67]

Thanks for the reply. My bungees are on the loose side and i had actually been considering tightening them up a little. I’ll leave will enough alone on that then.