Improve x-axis (short) carriage on Sophia

[amcewen]

Sean S reported problems with Sophia's accuracy, on the Google Group:

I used the Sophia laser cutter on Thursday last week and just got around to using the stuff I cut. There seems to be something wrong with the accuracy. Has this already been noticed and dealt with? I think the Y axis (long) is okay but the X axis (short) isn't/wasn't. A hole separation was expected to be 138 mm but was actually 144 mm. You can also see that the holes are kind of elongated in the X direction. I also have a piece where larger holes were cut and the ends of the circles don't meet the starts.

This ties in with Emma having problems with it ghosting during an engrave at Maker Night too. @magman2112 reckons there's a problem with the belt and the tensioner wheel.

[Sean-anotherone]

Having watched the kind of issues Sophia has I think the best repair would be to the sled to remove skipping and misalignment issues - this isn't at all difficult, but will take a bit of time and would probably be best undertaken once Monty(*) comes online. A bit of something "compliant" to make new wheels out of would be a good first step - 1" thick polythene, or delrin being first choices Bar or slab would not matter as both could be turned down. Rebuilding the sled as a 3-wheel zero backlash design would be a significant upgrade and is far easier than it might sound.

(*) a better name for Morris?, (who should only formally christened once up & running.)

[amcewen]

Improving the carriage would be good, but as you say, progress on Morris/Monty/Hugo/... is taking priority.

In the meantime, can we get a minimum fix (given it's possibly the idler wheel rather than the carriage) done so that we can start using Sophia more. In particular, @jackie1050, @dazz-ling and I were just discussing how it would be better to run inductions on Sophia to stop everyone from just booking Gerald because they're not familiar with Sophia, but we can't do that with Sophia in her current state.

Maybe something for the rest of the @DoESLiverpool/laser-maintenance crew to look into?

[Sean-anotherone]

With much help from @magman2112 the sled on Sophia has been rebuilt, it is now a 3-wheel design with spring-loaded tensioning. Following a quick align and test it was cutting 50mm squares within 0.2mm on both axes and with a bit more tweaking will get even better. There has been a price to pay however - due to an unforeseen lack of clearance between the sled and the Y rails the cut-able area is limited - about 23mm less front -to-back than previously. Some countersunk screws and a bit of 1mm steel plate to remake the third-wheel mount will resolve this, but for now, it's working again. The idler wheel on the X axis still requires work but for now it appears to be working, so I propose leaving it alone until after the big laser is commissioned and doing it properly then.

[amcewen]

Thanks for the update. @dazz-ling is going to mark the bed and make a sign to put up next to Sophia to warn people that the bed size is a bit restricted.

[amcewen]

Although the new sled has made the carriage lots better, it hasn't addressed this issue :-(

I've just tried cutting a Museum in a Box design, and over 165mm it lost 3mm on one part of the box and 6mm across the same distance on another piece of the box. It's also not a consistent, the two tabs on this photo should line up against the ruler...

Marking it as broken as I don't think it's usable for people at the moment

[magman2112]

I tested this on Thursday afternoon and identified that a 10cm square was OK on the long axis, but 9.7cm on the short axis.

I adjusted some software parameters that @Sean-anotherone and I had set late on the last Maker night, putting this back to defaults and I then got a good square. I then tested this at various positions on the bed to verify these measurements again.

Sophia was then in use for the rest of the afternoon and through into Maker night without any issues being raised.

I will leave this issue open for now though, as ideally @amcewen should test a cut of a Museum in a Box to confirm that Sophia is behaving herself.

[Sean-anotherone]

Despite best efforts, Sophia is not doing well and is currently completely out of action. The temporary sled fix is holding up but the tensioning mechanism on the short axis has fallen apart.

I will need to take the bed completely out of the machine to fix that, so while I'm at it I intend to implement a more permanent sled upgrade, too. Got a few things to track down first (40mm long countersink M4 screws, nuts, washers, nylocks etc.) but if all goes well I might get it all done today.

The current position however is that Sophia is broken.

[amcewen]

Thanks for the update (and the Broken label addition) @Sean-anotherone. I've blocked out next week on Sophia's calendar so no-one can book in to use it.

[Sean-anotherone]

I have fixed the axis so Sophia is back in use... but. The internal socket where the usb cable plugs into the controller is in really poor condition - the upper board from the controller needs to come out and have the socket resoldered (as a minimum).

It is working for now, but more as a matter of luck than anything else.

The test cut of a 50mm square in the middle of the bed was within+/- 0.5mm

[amcewen]

I cut another top of the museum in a box design, and it all seems good. I was going to close it, but @magman2112 reminded me that it'd still be good to rework the carriage to restore the full travel of the bed, so I've renamed it to better describe (I hope) what needs to be done.

I've spun the dodgy USB socket issue out as issue #1166.

[Sean-anotherone]

I have now (mostly) manufactured the required part (*), it is in the laser cupboard. I don;t see much point in fitting it as the machine actually seems to be working pretty well right now.

(*) the means of attaching the belt to the sled isn't built yet.

[amcewen]

Having just had to re-jig a bunch of design files because Sophia can't cut across a full 300x300mm sheet, logging my preference that we work out how to restore the full bed size. It's annoying to lose a ~5cm strip along our standard sheet stock.

[amcewen]

Copying @magman2112's comment from #1516:

As noted in #1508, the carriage on Sophia has been much modified and currently limits the x-axis travel due to parts catching the chassis.

I think we should design a brand new carriage that uses v-wheels and allows full travel on the axis as well as limit switches on the travel. We can then CNC cut this from a block of aluminium on the CNC machine.

I will look at doing some initial CAD work, using the existing carriage as a model, but I would also appreciate any thoughts, comments or wishes for improvements we can possibly include in this re-design.

[Sean-anotherone]

I have designed one and made it already, it's in the laser spares cupboard.  Needs some parts from the old carriage fitting, a spring to maintain 3rd wheel tension and it needs wheels etc, but it is mostly done Sean

On Tuesday, 3 November 2020, 22:44:12 GMT, Adrian McEwen <notifications@github.com> wrote:  

Copying @magman2112's comment from #1516:

As noted in #1508, the carriage on Sophia has been much modified and currently limits the x-axis travel due to parts catching the chassis.

I think we should design a brand new carriage that uses v-wheels and allows full travel on the axis as well as limit switches on the travel. We can then CNC cut this from a block of aluminium on the CNC machine.

I will look at doing some initial CAD work, using the existing carriage as a model, but I would also appreciate any thoughts, comments or wishes for improvements we can possibly include in this re-design.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.