Milling a hole (conversational) and hole creeps

All things related to the Centroid Acorn CNC Controller

Moderator: cnckeith

bbatarelo
Posts: 36
Joined: Fri Jan 10, 2020 6:25 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No

Milling a hole (conversational) and hole creeps

Post by bbatarelo »

Hi all,

I wanted to ask if any of DIYers here have experienced a following artefact?
creeping.jpg
This is a basic hole made with very conservative numbers, new brand name end mill, there's no chatter, operation is very quiet, material is aluminum and whatever I do it always creeps. On this photo it's conventional milling and hole creeps to the right. If I change to climb milling it reverses direction - it creeps to the left. Machine is new, steppers are strong (for machine size) Nema 34 (6 amp) with 1:2.5 ratio and they don't skip(for what I know), GeckoDrives 201X are inside, machine has ball screws and linear rails. This happens only on x axis only. Hole is round and with ok diameter but at each plunge it just moves a bit and move is very precise and consistent - not erratic and occasional. Plunge type also doesn't affect this. I played with feeds/acceleration - nothing. Pretty consistent as if it were programmed - which it wasn't. Hitting F8 in conversational produces nice hole and proper path - not something skewed.

Tried few other basic things such as facing operation or plunging and slotting using g code - no problems at all. Currently I am out of ideas :/

Bruno
Attachments
report_38D2694DFDF9-0110180504_2020-01-18_00-51-25.zip
(188.81 KiB) Downloaded 123 times
gerald martin
Posts: 90
Joined: Mon Oct 15, 2018 7:37 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC11: No
CPU10 or CPU7: No

Re: Milling a hole (conversational) and hole creeps

Post by gerald martin »

That's too consistent a problem to be missing steps, I would say...check the value for revolutions per inch on the X axis maybe? If you mark a location and move the x axis exactly 12 inches on the control, and measure the actual distance moved, do they agree?

Edit: Probably ignore that, I did not catch that the problem is not present when using the conversational CAM system. Sorry!

Hmm...on second thought, doesn't F8 graph it? I would expect the preview to be correct even if something is off in the settings for the x axis. Better let some pros weigh in here!

Gerald
cncsnw
Posts: 3765
Joined: Wed Mar 24, 2010 5:48 pm

Re: Milling a hole (conversational) and hole creeps

Post by cncsnw »

How does the tool move if you dry run it without material?

How about if you re-run the job after the skewed cut (so it should theoretically be retracing its path without cutting any more material)?

How exactly did you check to confirm that you are not skipping steps?
cnckeith
Posts: 7166
Joined: Wed Mar 03, 2010 4:23 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: Yes
CNC Control System Serial Number: none
DC3IOB: Yes
CNC11: Yes
CPU10 or CPU7: Yes
Contact:

Re: Milling a hole (conversational) and hole creeps

Post by cnckeith »

sounds mechanical.. loose pulley to me... run it in air like cncsnw suggested .....got some photos of the machine? and how it is built?

and this doesn't have anything to do with the pulley slipping..but please update to CNC12 v4.20 you are running v4.12 which is quite old.;-)

from another post on missing steps..

missing steps are caused by:
1.) bad wiring practices
2.) mechanical force (drag) exceeds stepper motor power at a given rpm, therefore it skips position (lost steps)
3.) running a stepper motor at high RPM speeds where it has the least torque, therefore it skips position (lost steps)
4.) running Acorn at a step frequency higher than the stepper drive can handle, therefore it skips position (lost steps)
5.) motor size and drive size ( or amp settings) not matched, therefore it skips position (lost steps)
6.) bumping the axis into a hard stop
7.) hitting Estop when axis is rapiding and then the axis coasts after the estop condition.
8.) if you drive any stepper outside it’s speed/torque curve it will lose steps. Bigger is not always better. Triple check your wiring, voltages and drive settings. Check your motor wiring, how are you wiring the 8 wire motor? Series or parallel? This changes the inductance and drive/motor performance. Are any of the motors getting hot in any configuration?
Need support? READ THIS POST first. http://centroidcncforum.com/viewtopic.php?f=60&t=1043
All Acorn Documentation is located here: viewtopic.php?f=60&t=3397
Answers to common questions: viewforum.php?f=63
and here viewforum.php?f=61
Gear we use but don't sell. https://www.centroidcnc.com/centroid_di ... _gear.html
cnckeith
Posts: 7166
Joined: Wed Mar 03, 2010 4:23 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: Yes
CNC Control System Serial Number: none
DC3IOB: Yes
CNC11: Yes
CPU10 or CPU7: Yes
Contact:

Re: Milling a hole (conversational) and hole creeps

Post by cnckeith »

looking at the error log in the report.zip.. you have alot of communication errors.
what are the results when you run a ping test? viewtopic.php?f=61&t=1451&p=8880&hilit=ping#p8880

be sure to have windows setup properly.. viewtopic.php?f=61&t=1456&p=8872&hilit=windows+10#p8872
no USB Ethernet adapters, use a shielded Ethernet cable, no anti virus software installed.

1) 01-17-2020 23:02:58 Tool change: T1 10mm 3f end mill
(4) 01-17-2020 23:06:28 9033 Reset Initiated, Press Reset to Clear
(4) 01-17-2020 23:08:58 4032 Reset Cleared
(4) 01-17-2020 23:09:04 9033 Reset Initiated, Press Reset to Clear
(4) 01-17-2020 23:09:05 4032 Reset Cleared
(4) 01-17-2020 23:09:12 9033 Reset Initiated, Press Reset to Clear
(4) 01-17-2020 23:09:13 4032 Reset Cleared
(4) 01-17-2020 23:15:23 9033 Reset Initiated, Press Reset to Clear
(1) 01-17-2020 23:22:32 204 Exiting CNC...Normal Exit (66)
(1) 01-17-2020 23:31:46 199 CNC Started
(1) 01-17-2020 23:31:46 Acorn-F (Free) Mill v4.12
(4) 01-17-2020 23:31:51 944 MPU requested resend # 12
(4) 01-17-2020 23:31:51 946 PC resending # 655
(4) 01-17-2020 23:31:56 944 MPU requested resend # 13
(4) 01-17-2020 23:31:56 946 PC resending # 660
(1) 01-17-2020 23:31:56 407 X+ limit (#50001) tripped
(1) 01-17-2020 23:31:56 407 Y+ limit (#50002) tripped
(1) 01-17-2020 23:31:56 407 Z+ limit (#50003) tripped
(4) 01-17-2020 23:31:57 4032 Reset Cleared
(4) 01-17-2020 23:32:40 944 MPU requested resend # 14
(4) 01-17-2020 23:32:40 946 PC resending # 664
(4) 01-17-2020 23:35:43 944 MPU requested resend # 15
(4) 01-17-2020 23:35:43 946 PC resending # 666
(4) 01-17-2020 23:36:32 944 MPU requested resend # 16
(4) 01-17-2020 23:36:32 946 PC resending # 667
(4) 01-17-2020 23:36:32 946 PC resending # 668
(4) 01-17-2020 23:37:53 944 MPU requested resend # 17
(4) 01-17-2020 23:37:53 946 PC resending # 670
(4) 01-17-2020 23:37:53 944 MPU requested resend # 18
Need support? READ THIS POST first. http://centroidcncforum.com/viewtopic.php?f=60&t=1043
All Acorn Documentation is located here: viewtopic.php?f=60&t=3397
Answers to common questions: viewforum.php?f=63
and here viewforum.php?f=61
Gear we use but don't sell. https://www.centroidcnc.com/centroid_di ... _gear.html
bbatarelo
Posts: 36
Joined: Fri Jan 10, 2020 6:25 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No

Re: Milling a hole (conversational) and hole creeps

Post by bbatarelo »

Thanks Keith,

I'll investigate more. This led me to Marty's video about minimal step width which took me towards speeds and feeds field which brought me towards fix - although it's possibly a temporary fix - feed dropped to 50% of what I was doing. My initial feed was 1500mm/min.
Regarding feed and torque, ballscrew thread is 5mm and reduction from stepper to the ballscrew is 1:2.5.

I checked if going to 100khz pulsing will have some effect, but to no avail. I believe that Gecko 201X should be able to handle 200khz.

As far as the other things are concerned

- I discovered that steppers are not that strong after all - 3Nm holding torque. I'll need to open the motor enclosure of the mill to check the wiring, but according to this document sent by mill manufacturer (
stepper-wabeco.pdf
(554.21 KiB) Downloaded 124 times
) it should be a 4 wire bipolar connection. They are working off the 48V 600W power supply. They are probably undersized compared to 7k5 rpm 2KW spindle.

- Windows 10 is set according to Centroid's recommendations and I think you already checked that half a year ago or more - we had one TeamViewer session.

- I will update Centroid software :) - I just didn't want to introduce more variables or possible delays until I figure this out.

- Computer is dedicated NUC 8th gen, i5 8259U at 2.3 GHz, 4GB ram and SSD.

- Wiring is done completely with ferrules (around 200 connections for the whole cabinet, no exceptions), ferric cores on both ends of step/direction cables, heat shrinking everywhere, shielding where possible. I unsoldered DB9 connectors and replaced them with XLRs, grounded everything.

- There is something I am guilty of and that is shielded ethernet cable - I temporarily switched to standard one because the one supplied with Acorn was too long/bulky. I am about to switch back to shorter SFTP that will be more convenient. I'll check pings.

- I did measurements for dry run - still skids more or less the same amount as with aluminium stock, but when tested pushing it a bit harder, it started skidding a bit towards Y as well.

- Checked mechanical things - all seems solid and well tightened.

- Machine is really German made - assembled by order (small and relatively expensive given its size) - compared to things like Optimum it is completely another league - fit and finish is really nice, paint is excellent, electronics are well done, everything is silent and chatter free. I can get some photos later if needed.

- I am about to drop in bigger power supply and bigger steppers if this really proves to be a deceleration issue due to weak steppers.

Thanks for this checklist!
Bruno
Nigelo
Posts: 366
Joined: Tue Dec 11, 2018 4:03 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: F045DA80C8B8-0905181172
E415F6F70BC3-0318203049
98F07B91FC6B-0123236802
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
Location: UK

Re: Milling a hole (conversational) and hole creeps

Post by Nigelo »

Bruno

Sounds like you have Wabeco's F1410-C mill with ball screws similar to my F1210-C but with lead screws. If so, your steppers are Sanyo Denki SanMotion 103H8221-6241 which should not be powered above 40vdc - I am using 36vdc through Gecko 203V drives which achieves a very satisfactory performance @ 200Khz including rapids to 110 ipm (about 2,800 mm/pm) all with no lost steps

Your report shows you have not set your lash compensation as 3 axes are 0. This cannot be correct especially as all transmission is with timing toothed belts.

Therefore suggest you check belt tension on all axes, especially X axis, and then check/calibrate lash compensation. This in addition to sorting the comms errors in your log, of course. I think this is what Keith is alluding to above

Hope this helps
Nigel
"You can lead a horse to water but you cannot force it to drink"
Hope this helps
Nigel
bbatarelo
Posts: 36
Joined: Fri Jan 10, 2020 6:25 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No

Re: Milling a hole (conversational) and hole creeps

Post by bbatarelo »

Hi Nigel,

thank you for chiming in. From google images it seems to me that those might be correct steppers though the photo I took when I had my cover taken off doesn't show the sticker. Suggestion from Wabeco was to have up to 40V so indeed I powered them with 36V such is your case. However, I left backlash compensation zero on purpose. When I was measuring backlash, it was almost non existent (I'd say about 2-3 microns) according to my .01mm Mitutoyo gauge. It was also measured with rather heavy 125mm vice on table. At one instance I did try adding some backlash (0.1mm), but results were worse. Also, with slow feed I am satisfied with finish look and feel forboth climb and conventional milling, though 30mm circle is some .08 undersize which I am yet to resolve.

Again, I don't hear any weird noises from steppers or see them stalling, it's clear that they can move 1500mm/min, but for some reason, deceleration is causing me an issue or at least that's how I perceive it. I'll check more stuff tonight including belt tightness jus in case, but last time I was satisfied with what I felt. Mill is practically new - I bought it about a year ago and in the meanwhile I had it working for half an hour max in a manual mode.

Bruno
cnckeith
Posts: 7166
Joined: Wed Mar 03, 2010 4:23 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: Yes
CNC Control System Serial Number: none
DC3IOB: Yes
CNC11: Yes
CPU10 or CPU7: Yes
Contact:

Re: Milling a hole (conversational) and hole creeps

Post by cnckeith »

check for pulley being loose on the shaft
Need support? READ THIS POST first. http://centroidcncforum.com/viewtopic.php?f=60&t=1043
All Acorn Documentation is located here: viewtopic.php?f=60&t=3397
Answers to common questions: viewforum.php?f=63
and here viewforum.php?f=61
Gear we use but don't sell. https://www.centroidcnc.com/centroid_di ... _gear.html
bbatarelo
Posts: 36
Joined: Fri Jan 10, 2020 6:25 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: No
Oak CNC controller: No
CNC Control System Serial Number: none
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No

Re: Milling a hole (conversational) and hole creeps

Post by bbatarelo »

Just came back from garage - checked everything and crimped a sftp cable. Pulleys are tightened strongly and both timing belts are very stiff.

Progress is in the sense that there are no more resends - communication seems stable and fast. Today's log doesn't show anything weird.
Attached is a new report.

However, problem still persists. I attached two photos - one (upper) with feed rate started at 1500mm/min and turned it in two iterations down to 65% where eventually hole stood still.
Photo below is at 65% - all looking ok regardless of the spindle speed which I varied between 1500 and 3500 rpm just to see if it will have any effect.
IMG_1182.jpeg
IMG_1183.jpeg

Depth is in both cases correct.
Attachments
report_38D2694DFDF9-0110180504_2020-01-20_23-16-08.zip
(194.37 KiB) Downloaded 121 times
Post Reply