Hello Guest it is March 19, 2024, 06:24:40 AM

Author Topic: Bridgeport Knee Mill Conversion?  (Read 269145 times)

0 Members and 1 Guest are viewing this topic.

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Bridgeport Knee Mill Conversion?
« Reply #550 on: October 06, 2016, 05:06:49 PM »
What do you think they are meaning about Gauge Block DRO? Are they saying it should be set to zero and they think there may be some value in it?
If you load the standard screenset you will see what value is in the DRO, you can then set to zero if that is what is required and then go back to your normal screenset.
Hood

Offline Davek0974

*
  •  2,606 2,606
    • View Profile
Re: Bridgeport Knee Mill Conversion?
« Reply #551 on: October 07, 2016, 02:10:03 AM »
It's zero, checked on the 1024 set.

I have no idea what they are leaning towards, I also asked Gerry who did the 2010 set and he has no idea either, the set does not use the 1001dro though.

Baffling

No macro's were in play at the time - it was before the M6 so M6Start was not called and M6End is empty anyway, no probing was being done so the gage block height is irrelevant.

I have passed the log data back to cs-labs
« Last Edit: October 07, 2016, 02:12:14 AM by Davek0974 »

Offline Davek0974

*
  •  2,606 2,606
    • View Profile
Re: Bridgeport Knee Mill Conversion?
« Reply #552 on: October 07, 2016, 06:11:08 AM »
Ok, reply from CS-Labs, my comments quoted in the reply start with >>>....

Dear Dave,
 
>>> As you can see, the system started up normal, then had a software limit switch activated - this was caused buy this line of code - G28 G91 Z0.
 
Soft limit limits are set out and passed to CSMIO/IP-A by  Mach3.
If Soft Limit was set correctly the obviously Mach3 made a mistake during Soft limit calculations or during passing to the controller.
 
>>> Then once i reset Mach it followed with an ePid fault on this line I think - G0 X109.987 Y168.641 or this one - G43 Z25. H2 which followed.
 
The e-PID mens too high difference between  currently set by Mach3 trajectory and currently set axis position read out from encoders.
The fault may be caused by Mach3 and servo drives as well.
 
- E-PID caused by Mach3.
The fault is mostly caused when Mach3 makes a mistake during calculations and it wants to perform very fast uncontrolled axes move.
A corrupted  XML file is usually responsible for that. The e-PID fault may be also caused by incorrectly written VB macros.
 
e-PID caused by servo drives.
That's nothing else but max. following position error exceeding of one of axes.
In this case you should verify your PID loop settings (tuning).
 
I would only like to add that for a milling machines we always use 03.043.022 Mach3 version as it's the most reliable in our opinion.
Version higher than 03.043.045 of Mach3 can sometimes cause problems with synchronization of internal coordinates so it's easier to get the e-PID fault.
 

Now, in my mind this looks like an internal Mach/CSMIO thing because in theory you should NEVER be able to exceed or trip a soft-limit - thats the WHOLE reason they exist, to control the boundaries of motion right?

So, somehow, my code for home is trying to send Mach Z axis beyond Mach Z++ soft-limit home setting - this is clearly wrong no?

Any thoughts?


So, it seems i need to downgrade further, anyone have a link for V022 please?
« Last Edit: October 07, 2016, 06:16:13 AM by Davek0974 »

Offline Davek0974

*
  •  2,606 2,606
    • View Profile
Re: Bridgeport Knee Mill Conversion?
« Reply #553 on: October 07, 2016, 09:17:53 AM »
Just been advised by a respected guy that v022 will not work as it has issues with tool-change routines that are not out-of-the-box seems v028 is better.

I'm holding off on downgrading until i have retested the same code that failed.

Slowly losing faith.........

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Bridgeport Knee Mill Conversion?
« Reply #554 on: October 07, 2016, 01:09:43 PM »
I use 062 and have never had an issue but I suppose each system can be different. You will find all available versions on the ftp site.
ftp://machsupport.com/

022 is going back quite a bit and I wouldn't be surprised if some things don't work, as an example I was messing around trying to find the problem with the CSMIO and Lathe threading and went back to an older version and threading was suddenly much improved (slow pullout being the problem) The Spindle Speed Override however didn't work in the older version.

Anyway, it seems like the G28 line is the problem, I never use G28 I use G53 instead, wonder if that could be the difference? I also do not go to Z0, I think mostly I do a G53G0Z-2.

Hood

Offline Davek0974

*
  •  2,606 2,606
    • View Profile
Re: Bridgeport Knee Mill Conversion?
« Reply #555 on: October 07, 2016, 01:19:17 PM »
Yeah, i could get the post modified - this is Fusion360 and the posts are gibberish to me ;) but the forum is great.

In my Aspire posts I use G53 G0 Z0.

Z0 or Z-2 will work, 2mm is not worth worrying about, but its odd that it only fails AFTER a tool change - something must get messed up internally i guess.

I cammed the job with one tool per file and it worked fine, at least it got the part made :)

Will be playing this weekend, just have the encoder to fit now, Mach is controlling the new VFD/motor nicely now.

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Bridgeport Knee Mill Conversion?
« Reply #556 on: October 07, 2016, 01:54:47 PM »
When I had the CSMIO controlling a servo spindle on the wee lathe, in the days before the IP-A allowed Step/Dir spindle, it was very good. Very linear and very accurate in revs. I thought this may just have been because it was a servo but a friend has a lathe with a DC spindle motor and a Mentor Spindle drive and he is using 0-10v and he says RPM are spot on as well.

Hood

Offline Davek0974

*
  •  2,606 2,606
    • View Profile
Re: Bridgeport Knee Mill Conversion?
« Reply #557 on: October 07, 2016, 02:02:16 PM »
Yep, my mini-mill took hours of farting about with spindle pulley values to get reasonably good linear control - the CSMIO just nails it straight out of the box, its about +10-15rpm out at any speed but thats no bad thing really.

I might increase braking time a tad - 1s stop from 3000rpm is pretty hard and tends to slam my worn dogs out on the back-gear clutch.

Apart from that its all gone well so far, just need to rig up a bracket for the spindle encoder, wire up and see what happens then :)

Offline Davek0974

*
  •  2,606 2,606
    • View Profile
Re: Bridgeport Knee Mill Conversion?
« Reply #558 on: October 07, 2016, 02:25:41 PM »
I've grasped enough to edit the post processor and change the G28 lines to G53 G0 Z0 :)

I'll still air-cut the original file - it would be nice if it fails again, then try the new version and see if that passes or fails :)

It would at least point the finger at the G28 stuff.

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Bridgeport Knee Mill Conversion?
« Reply #559 on: October 08, 2016, 03:46:37 AM »
Will be good to see what you discover.

Hood