If you are saying the right sprocket turns CW when you press CCW, I don’t know why that happens. As for “moves further than normal”, I suspect this is caused because the routines to set the tooth at 12 o’clock tell the motors to move a certain distance of chain. This is calculated based upon the number of teeth in the sprocket and the chain pitch. It appears that when the chain tolerance was added, this wasn’t taken into account. Since setting the sprocket is independent of the length of chain, it appears that we need to back-out the chain tolerance from the set-teeth-vertical routine so that when the controller adds it back in, then it gets to where it needs to be.
By “further than normal”, I mean several times more movement than usual. The 5 degree will cause a move of several inches. It might be possible to get it to 12 o’clock, but it also turns ccw when “tightening”, making the routine non-operative.
Thanks!
1 Like
I’m trying to replicate this right now and I will figure out what is going on and fix it as soon as I do
3 Likes
I have spent more time with the left chain routine, and have discovered the following…
- The right cog does not move during the tightening phase (I think). Messing with the chain tolerance changes how much chain is moved (I think) during the 12 O’clock routine, and maybe that is why the chain was constantly falling off. A clam sloved this problem
- depending on the value input into the chain tolerance, the motor will reverse directions, move more when trying to adjust to 12 O’clock. The left side worked fine with an input of .058248, but the right went into reverse mode with a value of .213065.
- I was able to operate the Left side, so I proceeded with the chain measuring routine, with the chain tolerance setting of .058248. The result was 964.39! Couldn’t believe it, so repeated, and got 964,29.
- tried changing the chain tolerance setting to .00058248, and the Left motor went into reverse mode for the 12 o’clock settings. Clearly either I am way off on how to compute and input the corrections, or … ?
- I really want to do something productive on the machine, so I went back to the chain pitch strategy (which, in my case, will only address the error on one side, but maybe I can get perfect circles out of my head for the time being. After several Maslow Left chain measurements, with the pitch set to 6.356, I get a Maslow measurement of 3003.48, really close to my Man measurement of 3003.5495.
Unless there is a process to get me or Mr. Maslow to understand how to use the Chain Pitch adjustment, I will run with this and see what the Benchmark does.
Thanks!
1 Like
It sounds like there is absolutely a bug in the chain pitch code. I haven’t had a chance to look at that section of code before and I’ve just run into this but:
but I WILL make the chain pitch code behave. It’s important for the kind of discrepancy you are seeing between your chains. I will let you know when I’ve been able to go through the process you outlined above without issues
4 Likes
Perhaps it makes more sense to have a separate pitch adjustment for each side, rather than the % Chain Tolerance. It was neat to see the Maslow motor distance change as I adjusted the Chain Pitch.
Thanks for your attention on this!
4 Likes
Out of curiosity, after you set the chain tolerance values, did you perform a manual chain recalibration to feed out the full chain lengths from zero?
1 Like
Since there is a problem using the chain tolerance values, I have not been able to fully calibrate the machine using them.
1 Like
It’s not uncommon to get to around 500 links in use, it doesn’t take much error
to be significant when multiplied so much
I have used the Maslow motor spacing / chain measuring tool multiple rimes in the last couple of days since you suggested making pitch adjustments and then checking the result. That is also what got me to check the right chain, which is actually worse. It is pretty neat to be able to correct the pitch number, and then go back and check the result. Sadly, you cannot change it for each side, and presently the right and left chain percentage adjustments are buggy. @bar is working on a fix.
Sure would be cool to have a way for the machine to remember the 12 O’clock settings, as that gets pretty tedious when trying to dial this stuff in. I suspect many others have unmatched chains, and for precision’s sake, an easy process to check the lengths and make corrections would be a good addition to the cal routine.
2 Likes
This will be added in version 1.13 it’s a top priority!
3 Likes
chain pitch cannot be changed, but chain tolerance is configurable for each
chain
We really need to adjust chain tolerance, not chain pitch, specifically to be
able to deal with the ‘pin at 12 oclock’ issue.
Very excited for this. I think it’s a key feature!
4 Likes
I just wanted to let everyone know that I believe 
that I got the bug in the left and right chain tolerances fixed in time for the 1.12 release.
I haven’t tested it much yet because it was an 11th hour fix and I would love feedback on how it’s working.
I’m optimistic that if it seems to be working we can get an automated process for computing those values into 1.13 which would be amazing! I’m thinking something like manually measure the edge to edge distance on the motors, then pull chain tight from the left and right to compute chain tolerances and you are done.
6 Likes
That sounds like the right process to use.
1 Like
Great, @bar ! I will take a look at it tomorrow PM. Will it be explained how to calculate the correction values to enter? Thanks!
Also, with the chain tolerance feature working, we would then return the chain pitch setting to the default, correct?
trying again with a longer message
Yes, you are correct
The edge to edge measurement should be taken with the chain pulled tight so I think the order is off a little. Am I correct in thinking the idea is to measure each chain’s tolerance?
- Extend left chain from left sprocket to right sprocket, pull tight for Maslow to measure
- Manually measure edge to edge distance on motors while still tight.
- Relax left chain and disconnect.
- Extend right chain from right sprocket to left sprocket, pull tight for Maslow to measure
- Manually measure edge to edge distance on motors while still tight. Compare to measurement of Step 2.
- Relax right chain and disconnect.
4 Likes
@madgrizzle Excellent suggestion. I was imagining that with the top bar we are getting no flex under tension, but there is no need to make that assumption. I like your method of measuring under tension.
@Dustcloud I’m actually not 100% certain what the right way to compute the that value is. My understanding is that it is a percentage error so if you are measuring the chains as 1% longer then the value to enter would be 1. It seems to me like you could just enter the values you computed previously:
Yes, resenting the chain pitch to the default should be correct because the tolerance values will be applied to that length after the chain pitch