Honest Appraisal of UCCNC

[dubble]

[dubble]

[clark25]

Hi,

the UCR200 is compatible with UCCNC + UC100 ?

Thanks,
Enzo

[insidelegman]

There is no speed issue on the 4th axis, but the UCCNC does not support the radius based surface speed control for the 4th axis.
In other words this is not something to be corrected in the software, but to support this will be a further development and a new option to select...

Be that as it may - call it what you will. If you have any serious 4th axis rotary work to be done , be prepared to wait a loooonnng time for completion - that is the bottom line which renders UCCNC useless for my purposes. Best try rotary work BEFORE you buy!

[dubble]

I'm calling it exactly what it is, it is not an error, but an unimplemented feature which seems you wanted to use.
And sure, I also advice to not to buy it if you need to do a work which is with 4th axis and if you need a radius based surface speed and if your CAM software can't generate the feedrates for your code, because the 4th axis radius based feedrate calculation is not implemented in the software.

The software demo mode can be tested prior to any purchase (the download is available on our website), the demo mode emulates all motions the same way as it happens with the licenced version,
so everybody can test it and can make a decision before purchasing.

[dubble]

Enzo,

It is not working yet, but as the plugin-ability was done now, my collegue started to write a plugin for the UCR200 wireless pendant .. will see how fast he can finish.

[Bad Wolf]

Solved one of my software issues, it appears it was a configuration issue on my end not the software. While he's in pendant mode any chance of ShuttlePRO v2 Controller support?

[daniellyall]

bad wolf only got one thing to say about M4 dam 0.02 + or - mm

[Bad Wolf]

lol I could live with +- 0.02, I m currently struggling to get my CNC to cut within 2mm of size
I have made another calibration change and will see if it is any closer next time.

[daniellyall]

I had to drop acc to 6% to get that got the wrong pitch ballscrew on my Z axis

[Bad Wolf]

I'm currently in the middle of changing my version 1 rack and pinion Z axis to version 2 ball screw, but I can live with the Z not being overly accurate for now. It's the X and Y I am having problems getting right. I'm not sure if some of that is attributed to flex in some temporary plywood parts, or the job moving.

I did a comparison between UCCNC and Mach3 and Mach 3 gave me rounder circles, I tried making sure I had the same settings for both.

I used 5mm pitch on my Z and 10 on both X and Y. I figured I wont need to travel massive distances on the Z, but need perfect depth.

[daniellyall]

you have to have all axis bang on it can make thing good in one direction totally walking of the table in another direction had X axis cut straight in one direction in another direction it was doing a nice circle when it was meant to be going straight fixed the calibration and it came right.

the best thing I found to do is do one axis at a time with the others turned of run a long code with lots of speed changes in that axis if its out by it`s self it will be worse with all axis going

I have two code I use one that is a 1000 in any direction the others 10000 if the 1000 is running fine I useley leave it but when I services the machine I do both 4 times a year then I now I can leave it for 18 weeks before i services it again unless we have had a earth quakes whats quite often at the moment

[dubble]

You have control over how round a circle must be in the CV mode. In exact stop mode the circle should be perfect (by means of the path and signals generated by the software.),
but in CV mode the linear position error settings for the CV mode defines how much position error can the software make in order to blend the path and to make the velocity close to the programmed.
The software will always obey your settings, so if you set a linear error rolarance of e.g. 0.01mm then the software will not generate more error than that when blending the path and this will define how preise your circle will be. (Again talking about the produced signals only and not the machine precision.)

The most problematic are the small circles, because the movement direction vector changes rapidly, so the optimised path should be logically a smaller circle if a fast enough feedrate is set and if the circle is small enough.

The UCCNC may make less round circle if the linear error parameter is set high, but again the non roundness value can be the maximum you setup.

As far as I known in Mach3 you do not have control over the error it makes in the CV mode, so you can only guess about how it will optimise the path and what size of position error it will make in order to blend the path, but the same is true for Mach3, it will optimise the small circles to a smaller circle or oval if the feed is fast enough and if the circle is small enough which makes the software hard to keep up with the velocity...

So, it cannot be said that the UCCNC makes less round circle than Mach3, because it depends on the settings.

[Bad Wolf]

Mach3 made rounder holes, that's not to say it was entirely down to the software. I still have considerable slop to weed out of the CNC, so I wont be concerned about it until everything is finalised. But interesting in a like for like situation Mach3 produced a rounder hole. The hole cut with Mach3 is on the bottom UCCNC on the top.

Neither provided an acceptably round hole, but with only one X axis ball screw currently in place I am surprised it was that round.

[dubble]

Hmm, I see quite a lot backlash there, in my opinion the difference can be from 2 things that the 2 software can accelerate up and deccelerate down differently and/or can they achive different feeds causing the backlash to apply on the workpiece differently.
The second possible thing is if the linear error is set too high in the UCCNC, comparably high to the error seen there.

We very often cut supports for bearings with the UCCNC and that needs to be in the tolerance range of 0.005mm or so, it cut without problems, the bearings fit always very well.

[Bad Wolf]

I think the error is more likely to be coming from the unsupported side of the CNC, and using the wrong cutters (hence why I am not currently worried).
That photo was cut with a 6mm 4 flute cutter (since buying the cutters I have learned 4 flute need to be run at twice the speed to cut right), but this latest piece (top of new ball screw Z axis) was cut with a 3mm 4 flute cutter. It cut much cleaner and smoother than the 6mm cutter at twice the depth per cut (1mm instead of 0.5mm).

[ger21]

Small circles are difeinitely the hardest thing to cut accurately on a CNC.
At work, we have a $100,000 router that cuts cabinet parts all day at 1200ipm, with perfect accuracy.
It has all ballscrews, and zero backlash.
Try to route a small hole, and it'll look like yours.

[daniellyall]

dont use 4 flute with wood your wasting your money 1 or 2 flute is all you need for wood, plastic and aluminum.

cutting small holes is best done slow unless you use a helix cut

[Bad Wolf]

I have a number of 2 flute cutters on the way. Buying 4 flute was part of the learning curve. Been raining here for the last few days which stops me wheeling the CNC outside the shed to use, but I am keen to see how a second cut with the 3mm goes.

[Bad Wolf]

Edit due to relevance.