Have run several circle tests, and they are slightly out of round…
In the center towards the top, 6" wide x 5 7/8 " high
In the center towards the bottom, 6" wide x 5 15/16" high
30" either side the error narrows to 6" wide x 5 31/32" high
What measurement errors cause this problem?
FWIW, I used the machine to measure top beam length, and the result was 3004.57… used this for the temp sled calibration. When I calibrated using the new sled, I used a steel rule to measure the distance between motors (right side of right motor spindle to right side of left motor spindle, and I got 3009.9, which I used.
For a rotation radius, I used the default 140mm, which I also tried to measure, estimating about 5.5" (about 140mm). with the new sled, positioning the ring onto the board was much easier with the 3 holes as guides. The support brackets needed to be pulled out slightly, so there was tension on those brackets. Mounting the router was mostly an eyeball line-up…
So it is probably not dead on. When I get back to the machine in 10 days or so, I will try flipping the sled over and checking the position via the router bit.
Have run several circle tests, and they are slightly out of round…
In the center towards the top, 6" wide x 5 7/8 " high
In the center towards the bottom, 6" wide x 5 15/16" high
30" either side the error narrows to 6" wide x 5 31/32" high
What measurement errors cause this problem?
it’s hard to say. it could be the motor distance, it could be rotation radius,
and it could be that the chain is not quite the length that we think it is.
FWIW, I used the machine to measure top beam length, and the result was
3004.57… used this for the temp sled calibration. When I calibrated using
the new sled, I used a steel rule to measure the distance between motors
(right side of right motor spindle to right side of left motor spindle, and I
got 3009.9, which I used.
that’s pretty close.
For a rotation radius, I used the default 140mm, which I also tried to
measure, estimating about 5.5" (about 140mm).
what did the calibration calculate the rotation radius to be?
with the new sled, positioning
the ring onto the board was much easier with the 3 holes as guides. The
support brackets needed to be pulled out slightly, so there was tension on
those brackets. Mounting the router was mostly an eyeball line-up…
you didn’t use the concentric circle printout to line it up?
So it is probably not dead on. When I get back to the machine in 10 days or
so, I will try flipping the sled over and checking the position via the router
bit.
If this is off, it could throw off the calibration calculations (you did try a
calibration after building the new sled, right?
The first thing I would do is to run the calibration test pattern to establish a baseline for how well calibrated your machine is.
The baseline test will test your machine in the far corners where it is likely to be the least accurate, and it will test across long distances where errors are most pronounced.
I will run a benchmark soon so that we can compare
I manually entered the 140 mm rotation radius value per the calibration script recommendation after I did my best to try to measure it.
Sorry, don’t know anything about the concentric circle printout… where do I find it?
I did calibrate with the new sled.
Bar…
Sorry, I am away from my machine for about 10 days. I was trying to get an understanding of what the causes were for the out of round problem. I cannot run the calibration test pattern until I get back to the machine.
Thanks for all the responses. I will restart this conversation after I tun the test cal.
re: the concentric circle printout, it was not in the package, is not on the packing list, and, as far as I know, is not mentioned in the “cut the sled” instructions. In reviewing the instructions, I probably could have used the plastic base plate that came with the router to help center it over the 3" hole. By the time I got to this step, I forgot it existed.
re: calibration routine calculating the Rotation Radius…
the above screen is the only reference that I can find in the calibration routine regarding the rotation radius. Where is it calculated, and how can it be updated? Here is my .ini file for the latest calibration. Clearly I am missing something.
@Dustcloud In testing I found that using a paper printout to align the ring was not as easy as I wanted it to be, and if anything went wrong there wasn’t really an easy way to print a new one at home because every printer tends to scale things a little bit if you aren’t careful. The new technique is to use the drilled holes to align the ring which I think is easier
I used the holes on the new sled to position the ring. The issue for me ia getting the router dead center on the 3" circle. I eyeballed it, and tried to measure interior plywood off set, and I am sure it is close, but probably not precise. It looks like the concentric circle chart will at least tell me if what I have already made is on target.
Regarding the Rotation Radius, I am using the default Ridgid Router, and the value suggested in the calibration routine is 140 mm, with a note that they are rough values, and should be updated. In trying to round my circles, are we OK with that 140 mm value.
In about 10 days I can run the calibration test pattern and get back to you all with the results.
I think that could be the issue. The real value is more like 138mm, 140 is just meant to be a starting value to get the calibration close and it is dialed in using the test cuts.
I am finally back at it! My machine got “snowed in” at my shop in Wisconsin, so I just got there today. I have cut the Benchmark Test, so what do you need from me now?
BTW, before starting, I used the concentric circle print out to check my router mount, and the spindle seems to be pretty close to dead center. I have not done any changes to my calibration, and attached is my .ini file from a couple of weeks age when I calibrated the machine. I have switched to v 1.11 of both firmware and GC.
How did the measurements come out for the test pattern?
The things to measure are both sides of the squares along with the measurements from the outside edge of one square to the other in both the horizontal and vertical directions like this:
Thanks! I am glad you sent the chart, as I really did not know what to do with the test. Will do the measurements tomorrow.
FWIW, I decided to re-calibrate using v 1.11, thinking there might be some improvements in the results. Went through the whole process, including the triangulation test cut. Then ran my circles test, with pretty much the same distortions… pretty much all perfect widths, with errors in the heights of 1/8” to 1/16”…
Connected with @Jon and @arnoldcp at Fully Calibrated but Oblong Circles
and chain pitch was discussed. Check it out and let me know what you think. I suspect that with the new calibration I will need to re-cut the Calibration test, correct?
@bar, when we take our measurements, I just want to confirm what the “ideal” values are for the squares and the left right and up down measurements.
I think the squares should be 100x100 mm measured across the same side of the cut line. If you measured just the resulting inner square it should be 93.65mm? I think this confused me at first because I was thinking I should be measuring inside of the cut to inside of the cut and I was ~6mm off.
The long left to right measurements should be 1905 mm, again measured to the same side of the cut line.
The vertically, the cuts are 900 mm, measured to the same side of the cut line.