Missing Steps in circular interpolation

[ShawnM]

- Are the headers not the same as the DB25? I am curious why they would introduce this problem.

Not even close, please read the install manual, this is all covered in section 5.1 on page 24. It tells you where to hook up stepper drives and where to hook up servo drives so there's no confusion. What step rate are you using in the wizard? As Gary mentioned, set it at 100,000, no higher.

[kman1967]

I have received my breakout board and will rewire the drives.

Another question, Should I bring back a separate COM from each drive back the break out or just jumper the COM from drive to drive and bring back one wire.

[tblough]

If you want to minimize the chances for noise induced problems in the future, you'll run a shielded twisted pair cable for step (step and com return) and direction (DIR and com return).

[ShawnM]

I have received my breakout board and will rewire the drives.

Another question, Should I bring back a separate COM from each drive back the break out or just jumper the COM from drive to drive and bring back one wire.

It is bad practice to "daisy chain" any connection for step and direction and as Tom noted they should be shielded, twisted pair wires. This is probably the most delicate pair of wires in the whole system and can cause the most trouble. Please take the time to do it correctly and use the proper wire. Honestly, it'll save you a bunch of headaches down the road.

[kman1967]

So here is an update...

I am using solid foam board as a test...

- Starting from 0,0,0. Z-0.1, cut 3" circle, Z0.2, X5, Z-0.1, Z0.2, X0, Z-0.2 and repeat cycle. I take a 0.1" deeper cut each time. I move to the X5 and plunge the cutter to help see if the tools is still in the same spot.
- This seemed to work good. I tried the same cycle now with a 1" circle and there were some changes
- I switched to a 1/8" end mill and then took a 0.1" deep cut and a 1" circle. I repeated the circle 5 times without lowering the tool. By the time I was done the circle was oblong. I tried this several times at different feeds, slower feeds were worse.

SO, still missing steps

[Sword]

It's been mentioned that your PC's processor doesn't meet the required single thread spec. Have you tried one that does? I've had one that was low and while linear motion was fine, as soon as I tried running a v-carve file, it stuttered and lost steps greatly.

[ShawnM]

Until you correct all the deficiencies in the system this will continue to happen. Just fixing one wont do it. The PC is probably the most important for now. Get a proper PC and report back, and I don't mean a laptop.

[KWORK]

May bee I am complet wrong.
I have may be an art of same issue.
Have only seeing it in probing Bore, abaut 50% of time it stop every time motors change direction you can go further with start.
I only used Probe Bore for test.
Not sure I fix it the right way but it works. Faund a old power supply from an old PC 0.6 volt less, and switched CNCPC power supply. NUC.

Look in Utility log errors. I have a lot of resendings.

I run Barrel nut final.cnc from mm dir, it run in 22 min at my cnc and stop exactly on zero on dial indicator, so NO missing step in linear moves.
Think noise filters on input on CNCPC and monitor need ground. but they get there ground against the Shieded net cable.

Input filter in the new supply is 2 x 1.2 nF open loop 100 - 110 volt input to net cable
Input filter in the old one is 2x 0.62 nF open loop 80 - 85 volt
Messuret on 1 pin on net supply and minus on output
Greetings Knud
Sorry for my bad english an spelling

[suntravel]

Probe stopping comes from the resending, had this to and worked long time on noise cancelling, but the solution was in the ethernet settings in the PC...

Uwe

[KWORK]

UWE thanks for reply
I spent some time searsing the forum, dit not find the right answer.
Can you remember what to change in pc setup or is it in forum.
Will like to set the orginal power back and check it out.
Thanks for your time.
Knud