El Beast - Initial design phase, comments and critique welcomed!

Looks very nice. I've heard good things about Misumi extrusions and they were one of the sources I had considered using in the past but I've never ordered anything from them. Thanks for the pics. They confirm my perception of Misumi's quality. Given the quality of the extrusion, do you think you really need the Thor Labs corner brackets? Looking forward to your build...

There is probably an argument to be made that I could live without them...

Maybe I'll go with just a single pair at the front and see how it comes out... easy enough to add a second pair at the back if required.


On a related tangent I've been working on a custom tool drawer insert for my CNC bench... still haven't placed all tools yet as I haven't fully decided what I want to put in... also need to make the handle in the front.


http://i.imgur.com/9yTQaYT.jpg

http://i.imgur.com/64vAP8O.jpg

http://i.imgur.com/2wrVWJA.jpg


https://www.youtube.com/watch?v=EmsdlI6nCVQ

So as promised, something more scientific....

I used my touch probe to measure the length of the cross braces. Fairly impressive cutting tolerance really I suppose, the nominal length was 344mm and the longest was 344.34mm while the shortest was 344.28mm so they are within 0.06mm of each other. Would have been even more impressed if they were all 0.3mm shorter :playful:

http://i.imgur.com/iEia1AN.jpg

http://i.imgur.com/rnzPGoB.jpg

http://i.imgur.com/1MUxB8U.jpg


So my options are to use some shim material to make them all up to an effective 344.34mm or I can try to nip them all down to 344mm on the CNC... Shimming is probably the least risk route.

Got home today to find FedEx had left me a nice treat:

http://i.imgur.com/xyLmFI0.jpg


My linear motion stuff arrived, and initial inspections it all arrived intact. HG25 rails and carriages could probably be used for seal clubbing if one were so inclined... didn't realise quite how beasty those are.

Quality of the TBI ballscrews looks to be fantastic - finish is excellent, as is the end machining and they are as straight as an arrow with absolutely no discernible slop or play on the ballnut. I intend to do some tests with the glass scales just to see what that tells me but from a visual inspection stand point they seem top-dolla!

some linear motion porn (NSFW?)


http://i.imgur.com/Wtpg5po.jpg

http://i.imgur.com/MLR2J4A.jpg

http://i.imgur.com/8BJ9979.jpg

http://i.imgur.com/s74dYYQ.jpg

http://i.imgur.com/PAsXLcU.jpg


Had a bit of an order snafu with Thorlabs - got an order receipt and they reserved funds on my bank account but the order didn't get processed. I have spoken to them and re-ordered so those should be arriving shortly. Other than that I'm just waiting on an order from motedis for various extrusion related fixings and then I should have most of what I need for the bottom frame build.

Out of curiosity, where did the linear motion bits come from?

Topper on aliexpress... I was dealing with a guy that goes by the name of John (I would assume a western friendly name!) - very helpful chap. Just emailed him asking for a quote for all parts with details of what I needed and then they can create a special order for you on aliexpress so you still get the buyer protection afforded by them.

I was one HD16 short (can't really complain since they threw those in for free) and I've sent an email to enquire about that so we'll see what they say. They do both TBI, HIWIN and they do their own factory Chinese made stuff too which would work out quite a bit cheaper. For this machine I went with the HIWIN rails and C3 TBI screws, for my other project I went with some of their own factory manufactured ground screws and nuts which were significantly less than the TBI equivalent (C5 ground 1204 screws are very expensive it seems!). On first impressions those factory own screws look well made too, and while they don't have any discernible play on the nut they aren't quite as smooth to spin by hand as the 16mm TBI ones for El Beast.

Was very happy to see the end machining has been done to a very high standard - bearings are a snug fit and the machining is concentric. All the other parts like rails, carriages, MBA-12C mounts, FK and FF bearing sets seem high quality - So far so good.

As mentioned I want to check the screws against the glass scales just to check lead error and consistency more out of interest than anything else, but having had a good look at them I would be surprised if they were anything other than excellent.

Did a little machining today... found some nice little extrusion connectors at motedis which will work nicely at pulling the two side extrusions together:

http://i.imgur.com/i1mWQS9.jpg


There is a rather pricey drill bit and jig that you can buy to put the hole in the extrusion at the right spot, but I figured it was just as easy and much cheaper to use the CNC instead. I just 3D printed some little work alignment jigs so that I didn't have to re-zero between each setup:

setting up:

http://i.imgur.com/53xS3dV.jpg


Milling using a fairly long 8mm bit to a depth of 36mm

http://i.imgur.com/X6Xwd9m.jpg

http://i.imgur.com/MFIeInr.jpg


Fits nicely:

http://i.imgur.com/Agk8lcS.jpg

http://i.imgur.com/Htp75AJ.jpg


Oh and Thorlabs brackets arrived last week - I've just gone for the front ones at the moment given that they are certainly not cheap. Will wait to see whether I think the rear ones are a good idea or not.

http://i.imgur.com/kNE73g5.jpg

Looks like it will be an extremely high quality machine!

I have a question which you will probably find quite noobish but I am interested in your glass scales. Specifically, how do you use the glass scales in a feedback loop to control your position? Does the position from the scales get fed back into the motor driver and that corrects any error? Or is the position fed back into your break out board and on to your motion controller?
The reason I ask is if you have position feedback​ from a linear scale any backlash in your drive system wouldn't matter?? Then theoretically one could use c3 screws instead of c7 for instance. Or, standard gearboxes instead of low backlash units. Am I thinking correctly here??

Sent from my Pixel XL using Tapatalk

Quote:

---

Originally Posted by 1Jumper10

Looks like it will be an extremely high quality machine!

I have a question which you will probably find quite noobish but I am interested in your glass scales. Specifically, how do you use the glass scales in a feedback loop to control your position? Does the position from the scales get fed back into the motor driver and that corrects any error? Or is the position fed back into your break out board and on to your motion controller?
The reason I ask is if you have position feedback​ from a linear scale any backlash in your drive system wouldn't matter?? Then theoretically one could use c3 screws instead of c7 for instance. Or, standard gearboxes instead of low backlash units. Am I thinking correctly here??

Sent from my Pixel XL using Tapatalk

---

I simply use the glass scales for calibration purposes. They were mainly the answer to my current machine's less-than-optimal screws - I simply adjusted the steps per unit to match the scales. In mach3 you can actually go deeper than that and map the ballscrew to compensate for variation along the screw itself too, but I never went that far.

Certainly it is possible to link a positional encoder such as a glass scale and a CNC controller to give absolute position feedback, but I haven't looked into it too far as for my needs the calibration was sufficient. I think linuxCNC has a way of using it.

It certainly would be a good way of compensating for lead inaccuracies and variation - I don't think it would do a great job of controlling excessive backlash though, but then most semi-decent ball screws don't have much backlash anyway. If you had a lot of backlash it would have to be very quick to arrest movement to stop chatter and poor edge surface finish, there is never really a good solution for backlash other than to remove it.

BTW I think you have the C's backwards, the lower the C number the more demanding the specification... so in your example it would more be a case of it allowing you to use c7 rather than more expensive c5/c3/c0. Of course good glass scales aren't cheap either so it's debatable as to how much saving you would make.

Edit - EMC2 can do it too apparently.

Edit 2 - also came across this quote which pretty much describes what I was thinking regarding it not being a cure for backlash:

Quote:

---

But, a caution: Knowing position via the glass scales doesn't
eliminate the backlash problem. The real problem is that the
position of the table is not constrained in both directions by
the servo. it is only constrained in one direction at a time,
and the motor cannot hop from one side of the backlash to the
other instantly, therefore cutting forces and inertia can flip
it from one side to the other faster than the motor can
compensate. This can lead to messed-up parts, broken tools and
general foul language around the shop. So, don't think the
glass scales are some kind of panacea that allows you to do
precision work with sloppy leadscrews.

---

Got it. Thanks!

Sent from my Pixel XL using Tapatalk