Boxford Z Axis Backlash?

52 views
Skip to first unread message

Andrew Carson

unread,
Sep 4, 2021, 7:34:21 AMSep 4
to reading-...@googlegroups.com
Hello,
I was using the Boxford this morning and I noticed some oddness: It was seemingly not responding to every keystroke. It looked like the gcode was being generated and sent in the console window, but the machine was not moving.

This seemed to happen pretty much only for when switching direction for fine Z axis adjustments.

Once I was aware, it wasn't too difficult to work around, so I was able to finish my piece - for which accuracy was not a big deal.

Because why not, I gave it a reboot and then opened up CNCjs. Now there was no console window and the 'Port' drop down was empty, so I was suspecting a communication issue. A second reboot, this time with an actual sustained powered down period and the console now connects, but I can still reproduce the fine Z axis movement issue. 

Having played with it for a bit, to novice me, the issue now feels like backlash. After a series of fine z axis movements, the next one or two fine movements in the opposite direction appear to do little.

Andy

Steve Rodway

unread,
Sep 4, 2021, 11:12:34 AMSep 4
to reading-...@googlegroups.com
I tweaked the gib strips on the boxford a few days ago to try to reduce
the slop in the axis, might have over-done it; it's always a bit of a
trade off between getting out the slop and introducing backlash.

I might have time to check it out again today but if not then I'm stuck
at the lab on Monday dealing with the problems with the Axminster air
filter so I'll take a look then.

Communication issues seem to crop up from time to time, mostly dependant
on the current phase of the moon and if there's a weather balloon
overhead, hopefully that'll go away with the new control board when I
get around to doing that.

Steve
> --
> You received this message because you are subscribed to the Google
> Groups "rLab / Reading's Hackspace" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to reading-hacksp...@googlegroups.com
> <mailto:reading-hacksp...@googlegroups.com>.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/reading-hackspace/CAGTmkM0_ppQFY0GHZ4duG%3D9_k7qvNr_fzk12VsgkzfW0mV1L2w%40mail.gmail.com
> <https://groups.google.com/d/msgid/reading-hackspace/CAGTmkM0_ppQFY0GHZ4duG%3D9_k7qvNr_fzk12VsgkzfW0mV1L2w%40mail.gmail.com?utm_medium=email&utm_source=footer>.

Andrew Carson

unread,
Sep 5, 2021, 2:07:52 PMSep 5
to reading-...@googlegroups.com
Awesome, thanks Steve.

On a slightly different note, I want to try writing my own g-code for simple operations. Obviously, I'll go (very) carefully, but is there any reason that this is a terrible idea?

If it's not, does anyone have recommendations for a good g-code editor?

Andy



To unsubscribe from this group and stop receiving emails from it, send an email to reading-hacksp...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/reading-hackspace/83c7ef9a-7484-becc-6f09-07e8002b9a47%40legionelectronics.co.uk.

Steve Rodway

unread,
Sep 6, 2021, 10:24:52 AMSep 6
to reading-...@googlegroups.com
I've had another go at the Z-axis gib adjustments, it's now running with
a backlash of 0.015um and a slop of about 0.035um. That's about as good
as it's likely to get on the current setup. If you need better than that
then there's work that could be done on the boxford to make it better
but we've already got the low-hanging fruit and further improvements
will need increasing amounts of time and effort spent. If you need
better than we currently have then we could have a talk about how you
could try making those improvements.

As for writing G-code manually, I normally discourage that because it's
a very error prone process but if there's a specific reason that you
need it then it's an option. As you say, go very carefully indeed, use
lots of break-points in your program and pay attention to air-passes and
being ready on the E-stop but there's no fundamental reason why you
can't do it. You'll probably want to take a copy of the post-processor
from the boxford and study it so you can understand what it's doing and
how it's customizing the g-code for the boxford.

I understand there's g-code language modules available for notepad++ and
for Visual Studio but I've not used either of them for that so can't
comment on their merits.

Steve

On 05/09/2021 19:07, Andrew Carson wrote:
> Awesome, thanks Steve.
>
> On a slightly different note, I want to try writing my own g-code for
> simple operations. Obviously, I'll go (very) carefully, but is there any
> reason that this is a terrible idea?
>
> If it's not, does anyone have recommendations for a good g-code editor?
>
> Andy
>
>
>
> On Sat, 4 Sept 2021 at 16:12, 'Steve Rodway' via rLab / Reading's
> Hackspace <reading-...@googlegroups.com
> <mailto:reading-hackspace%2Bunsu...@googlegroups.com>
> > <mailto:reading-hacksp...@googlegroups.com
> <mailto:reading-hackspace%2Bunsu...@googlegroups.com>>.
> <https://groups.google.com/d/msgid/reading-hackspace/CAGTmkM0_ppQFY0GHZ4duG%3D9_k7qvNr_fzk12VsgkzfW0mV1L2w%40mail.gmail.com?utm_medium=email&utm_source=footer
> <https://groups.google.com/d/msgid/reading-hackspace/CAGTmkM0_ppQFY0GHZ4duG%3D9_k7qvNr_fzk12VsgkzfW0mV1L2w%40mail.gmail.com?utm_medium=email&utm_source=footer>>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "rLab / Reading's Hackspace" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to reading-hacksp...@googlegroups.com
> <mailto:reading-hackspace%2Bunsu...@googlegroups.com>.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/reading-hackspace/83c7ef9a-7484-becc-6f09-07e8002b9a47%40legionelectronics.co.uk
> <https://groups.google.com/d/msgid/reading-hackspace/83c7ef9a-7484-becc-6f09-07e8002b9a47%40legionelectronics.co.uk>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "rLab / Reading's Hackspace" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to reading-hacksp...@googlegroups.com
> <mailto:reading-hacksp...@googlegroups.com>.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/reading-hackspace/CAGTmkM3Kg%2Bha8D-OasoK9Siv%3D6BUroOnSbsXS%3D2-DL2UZq0ZTw%40mail.gmail.com
> <https://groups.google.com/d/msgid/reading-hackspace/CAGTmkM3Kg%2Bha8D-OasoK9Siv%3D6BUroOnSbsXS%3D2-DL2UZq0ZTw%40mail.gmail.com?utm_medium=email&utm_source=footer>.

Alex Gibson

unread,
Sep 6, 2021, 1:43:46 PMSep 6
to reading-...@googlegroups.com
Notepad++ with GCODE colour code add-on is great.
Follow instructions here to download:
https://www.reddit.com/r/3Dprinting/comments/ffe4tn/gcode_in_notepad_syntax_highlighting_for/


Alex Gibson

+44 7813 810 765 @alexgibson3d 37 Royal Avenue, Reading RG31 4UR

admg consulting

edumaker limited

• Project management
• Operations & Process improvement
• 3D Printing
To unsubscribe from this group and stop receiving emails from it, send an email to reading-hacksp...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/reading-hackspace/b2ea8379-a3ad-e4b0-7088-489addadc2af%40legionelectronics.co.uk.

--
This email has been checked for viruses by AVG.
https://www.avg.com

Richard Ibbotson

unread,
Sep 6, 2021, 2:56:42 PMSep 6
to reading-...@googlegroups.com
Steve, Those backlash and slop figures look unrealistic. Are they actually in mm ?
Richard

Sent via Richard's Phone

> On 6 Sep 2021, at 19:36, Alex Gibson <al...@alexgibson.net> wrote:
>
> Notepad++ with GCODE colour code add-on is great.
> To view this discussion on the web, visit https://groups.google.com/d/msgid/reading-hackspace/00f701d7a346%2462950310%2427bf0930%24%40alexgibson.net.

Stephen Rodway

unread,
Sep 6, 2021, 5:07:50 PMSep 6
to reading-...@googlegroups.com
Ooops, yes that's 0.015mm and 0.035mm. If it had been in um I'd be
trying for a job at Kern :)

Steve
Reply all
Reply to author
Forward
0 new messages