What is correct chain pitch?

Back to my original goal starting this, I just redid the calibration using 0.99 but with 8114 as the value and Maslow measured at 2902.4 mm vs. 2905. I wonder if my wood shrank at all? It’s been fairly dry lately.

2 Likes

what if someone has a gearbox that actually has 8148 steps/rev? how would they
not get things changed.

There needs to be a guard config somewhere in there

config_version (or similar) lets us make automated changes to the config when
the program changes, but only once.

Another place this would be useful would be to disable custom PID settings when
we make major changes like we did in 0.99, but allow the user to set them back
if they want without them being undone again after the one transition.

3 Likes

Oh that is another good example.

@dlang and @krkeegan Thanks for sticking with the idea. I’m starting to understand the value of a config version number. It would let us force a one time change to the settings which the user could then override. I’m on board!

@madgrizzle I’m about to do a calibration also. Where were we we on your last calibration? Are we getting closer? 2.6mm seems like a pretty solid improvement over just a couple weeks ago when we were seeing 20+mm errors!

3 Likes

Any chance that’s from deflection under load. Wonder if it would change if the moisture content changed because of humidity?

1 Like

Measuring with a tape during the auto-cal would tell the tale :wink:

I ran the test of 500 revoutions again tonight with the updates to GC and the Firmware with the step count at 8113.73 and it is as dead on as I can perceive. Considering our max error is a 10th of this, I am very happy. Thank you everyone.

5 Likes

The 2905mm measured with a tape was done at same time the chain was tensioned during Maslow’s measurement last week. Going from 2888.6 mm to 2902.4mm (13.8 mm) is a big improvement.

When I upgrade to the latest firmware and autocal again (to 8113.73 _encoderSteps), I’ll recheck the tape measurement. Some of it (maybe 1 mm) might be attributable to how I measure from left side of left sprocket to left side of right sprocket. The left sprocket has chain wrapped on it and I eyeball where it would be on the right sprocket if it also had chain on it. I’ll put a link on the right sprocket to see if that makes a difference. It won’t be 2.6mm difference, but it will be closer.

2 Likes

Not to open up some new wounds, but I’ve got a big problem that I’m starting to diagnose.

I’m using 0.99 firmware and GC but I have set the encoderSteps to 8114. After calibrating last night, I went out to do a cut this morning and was moving the sled to where I needed to set home, however, when I use GC to move, say 16 inches left, it also dropped a couple of inches. When I went 16 inches back to the right, it rose back so it returned to the original starting point. After doing a few more movements, including ups and downs, I decided to tell GC to recenter and it repositioned center to about 12 inches to the left of actual center. Weird things are happening.

I’m going to try to get the new GC and firmware working on my laptop I use and try it there.

Indeed, keep us posted. I haven’t noticed anything like this but I will check it out

I wonder if the fact that R didnt get updated after changing encoderSteps was the cause of the problem. Left 16 moves left again, not downwardly left.

So it went away of it’s own accord or did something change?

That tells me something went wrong in the calibration andthe machine dimensions
are not correct (so maslow thinks it needs to feed out too much chain as it
moves left)

I loaded python on my computer and used the latest versions of GC and FW (1.00). It’s working really well. Everything on calibration had gone well using encoderSteps of 2114 on GC/FW 0.99 so I was surprised by the behavior. I do know that kinematics.R does not get updated in version 0.99 when you change encoderSteps. It does in 1.00. So, if R is off, it could easily have been the problem. I’m cutting out a piece at the moment and its tracking very nicely.

2 Likes

Maybe we should think up a fine adjust for one motor mount
That when the wood expands or contracts only the motor position has to be adjusted.

An adjustment is just one more place for inaccuracy, it’s better to just
re-mesure the width if humidity has changed significantly (if your frame is
affected by it)

1 Like

Know I needed an excuse to get a TIG welder (spent some downtime watching welding videos), a steel frame.

Rather than an adjuster a beam across the top would solve it. A T, I, or box out of scrap plywood could work if unistrut or other steel stiffener isn’t practical

1 Like

Even gluing together a couple layers of 3/4 plywood (essentially a home-brew LVL
beam) would do it (and would handle the problem of getting 10+’ of length out of
8’ plywood)

2 Likes

I am hopeful that Unistrut works for this application as it is the solution I am implementing on my build. I haven’t had a ton of time of late, but I am hoping to get cutting by new years.

this is the posting of my build:

1 Like

Ok. Now forgive me for reviving it. But I think it is worth it to close it for good. After all if any newcomer seeks the answer, what a loss to go through 132 replies long and yet get no answer :frowning_face:.

So what is correct chain pitch? The answer would be that chains exact pitch do vary with batches and wear. So that has to be calibrated for a MaslowCNC to be accurate, and is essential to reduce workarea warping. See list of sources of errors.