[CFLBob]

You can heat the tubing up with a hotair gun for example to give it a 90 degree bend or 2 so it clears the saddle, it will hold that shape when it cools.
Hoss

Attachment 347890

Thanks, it fits better, but the slide is still over that piece of tubing. If you say it's cool, I'm good with it.

Attachment 347896

Onward to putting the bearings on.

[joemusso]

Maybe someone can give a little help here.

I have a C11 BOB and cant get mach3 to see it.

When I got to tune the motors they are all grayed out I can't click on them.
Not sure what i did wrong.


Thank you
Joseph

[hoss2006]

Can't tell but if it rubs it will eventually rub through the tubing.
Here's another view of mine from the plans.
Hoss

[CFLBob]

Apparently the sliding part of the table is wider or something, because it still goes over the fitting. In the picture I just posted, it seems like it's right over the bore of fitting. I pulled the tube and it's really easy to see this way.

Attachment 347906

This is with the table slid as far as it goes in that direction. While working, I wondered if a smaller tube I have could work there. It's 3/32 ID, so it has to be stretched 1/32 to fit on a 1/8 fitting, so I heated it and tried. Worked fine.



This 1/32 smaller than the 1/4" tube in both OD and ID. I have fittings for it all.

[pippin88]

Can you use a small bracket to pull the tube down onto the alu part towards the bottom of the picture so it doesn't rub?

[CFLBob]

Maybe someone can give a little help here.

I have a C11 BOB and cant get mach3 to see it.

When I got to tune the motors they are all grayed out I can't click on them.
Not sure what i did wrong.


Thank you
Joseph

Not sure I can help, I don't have a C11, but are the lights on? I have the CNC4PC C35 board and it has LEDs for every axis. When I was setting it up, I didn't know I had to wire an enable line on the card to get it work, and a quick check of the C11 manual says it needs that, too. I wasted a few hours before finding that.

Just a couple of things to look into.


Bob

[CFLBob]

Can you use a small bracket to pull the tube down onto the alu part towards the bottom of the picture so it doesn't rub?

I think it always rubs, since the table goes over the middle of the fitting. I haven't checked really closely.

[Wiggles84]

@CFLBob

you could machine away some of the table

Attachment 347916

Andrew

[CFLBob]

@CFLBob

you could machine away some of the table

Attachment 347916

Andrew

Well that would fix it for sure, but I don't have another machine big enough to do that.

[CFLBob]

@CFLBob

you could machine away some of the table

Attachment 347916

Andrew

After sleeping on this, I think I see another way to fix this problem besides cutting away that corner of the table.

Cut away the ballnut mount. Specifically this web of metal. (Crude Microsoft Paint drawing)

Attachment 347974

This makes the 3/8 wide notch in the top go through the full thickness of the aluminum. That makes it easy to put in a right angle fitting, which pushes the tube coming out well inside of the edge of the table.

The only downside is that it takes my system completely down to the smallest subassemblies again. I'll have to break LocTite in lots of places, put the ballnut and mount back on the ballnut removal tool, and take it off the screw. A lot of work. On the other hand, if that tube is abrading, it's going to require a lot of disassembly, too.

[CFLBob]

Well that would fix it for sure, but I don't have another machine big enough to do that.

I have a tendency to suffer from what they put in that old saying, "when all you have is a milling machine, every problem looks like milling operation". (or something like involving hammers and nails).

I resolved this with a couple of little things. First, I glued the bent over tube to the ballnut mount, taped it down with duct tape, and when that didn't seem like a good enough clamp, I stacked some weights on it. The tube came out about 10 mils under the slide, as seen here. That white thing is a business card that I measure at .010. I eventually tried a feeler gage and found a 12 mil strip would go in there, but not a 15. 12 mils clearance is good, but since this is totally unobservable when the mill is running, I wanted some more clearance.

Attachment 348112

Talking with a friend, he suggested it would be easy to make my slide look like that one with an angle grinder, and grinding that had honestly not crossed my mind. A couple of hours later, I had this:

Attachment 348114

So now I'm ready to try to put the table back in position and get X/Y moving. While I figure out what to do about Z.

[kuyote]

Hoss, I'm going back over my phase 2 build from your plans, and I'm wondering, in older versions of the balls screws, was the x axis tension on the balls screw set by the end caps of the bed? The ballscrews weren't threaded on the end without the motor, but I see in the videos now, that you screw both sides in to prevent backlash. There isn't any play when the ends are tightened, I just want to make sure I'm not leaving something incomplete causing backlash.

[hoss2006]

Hoss, I'm going back over my phase 2 build from your plans, and I'm wondering, in older versions of the balls screws, was the x axis tension on the balls screw set by the end caps of the bed? The ballscrews weren't threaded on the end without the motor, but I see in the videos now, that you screw both sides in to prevent backlash. There isn't any play when the ends are tightened, I just want to make sure I'm not leaving something incomplete causing backlash.

The non stepper end of the x screw just calls for a single roller bearing just lightly snugged to prevent whipping of that end of the ballscrew, the stepper side with double bearings is where to adjust for backlash.
Hoss

[CFLBob]

Another question, Hoss. I noticed while getting my table to clear the hoses that sometimes it felt like it hit something hard while I was sliding the table back and forth. I saw some debris on the top of the X-axis ballnut, maybe scraped grease or something else black.

Now I find that when I put all the hardware together, it seems to be binding somewhere and not moving well. I noticed this discrepancy in the X-axis ballnut from Chai. Do you think this is OK, or do you think I should grind that off a bit?

Attachment 348534

Thanks

[hoss2006]

Yeah the top of the flange will likely need a little more removed.
You can wrap the nut and screw with tape just leaving the little bit to grind exposed, grind a little at a time to keep it from getting too warm.
Hoss

[CFLBob]

Yeah the top of the flange will likely need a little more removed.
You can wrap the nut and screw with tape just leaving the little bit to grind exposed, grind a little at a time to keep it from getting too warm.
Hoss

In case anyone stumbles across this, the ballnut flange wasn't my issue. I found the binding in my motor mount and once that was resolved, my machine moves just fine.

[CFLBob]

It's done! My machine is fully built up and running just fine on all three axes.

Attachment 348796

Couldn't have done it with your help.

At this point, the whole project isn't done, but I'm moving from building to tweaking and optimizing. I checked accuracy this afternoon, at least on the small scale. When I tell the mill to move 0.100", it goes .100. When I tell it to go 0.500, it goes 0.500. I need to check it over longer distances, once I get my calipers I can trust.

On the downside, I'm disappointed with the backlash. The whole purpose of changing from phase1 to phase3 was to reduce backlash to the smallest amounts I could get. I measured .012 on X, .010 on Y, with Z coming in at a more respectable .004. I was hoping for .001 or .002 on all three. I'm not sure where to start looking for those thousandths.

I'm not getting the same kinds of speeds on rapids. I think I get around 150 IPM on X and Y, and didn't get up to 100 on Z. It's faster going down than up. On the other hand, I'm trying to run LinuxCNC and the motor tuning is bizarre. Or worse, so it might just be that the motor tuning really sucks.

Anyway, a good time to take a day off and do a little celebration.

Then it's on to completing the oiling distribution, building up the enclosure, adding cooling system, limit switches, and other tweaks.

[kuyote]

In case anyone stumbles across this, the ballnut flange wasn't my issue. I found the binding in my motor mount and once that was resolved, my machine moves just fine.

How was the motor mount causing binding? I'm having some binding right now and trying to find where its coming from

[CFLBob]

How was the motor mount causing binding? I'm having some binding right now and trying to find where its coming from

The corners of jam nuts would hit the bored hole on the bottom half. I could see that the mount was offset compared to the ballscrew. I think I shaved like .010 or .015 off the bottom part of the bored hole.

Hope that's clear.

[CFLBob]

I have an odd question that I hope someone has seen. It's in the electronics, and I think I've ruled out everything except for the C35 interface card from CNC4PC.

The KL-6050 drivers have a switch to chose between 1/2 and 1/8 microsteps. I find that the driver stays in 1/2 microsteps regardless of what that switch is set to.

Here's the really odd part. Sounds like the motor driver is acting wrong, right? I have four of the KL-6050 drivers; three purchased at once in June of '15 and one purchased a few weeks ago in December. All four step in 1/2 microsteps no matter what that switch is set to. I'd rather have the 1/8 microsteps. I suppose it doesn't really matter because with the controller software set up, if I tell it to go some distance, that's how far it goes. I just expect more steps.

The C35 interface is the only thing that touches the drivers, and I don't see how that could change the function of that switch. It's the only thing left, though. On the one hand, the motor drivers pretty much have to be good; all four can't be bad. On the other hand, the C35 board apparently can't affect that switch. It just turns the step and direction signals into the different levels that the motor needs.

So nothing is wrong, but something is wrong, which is a bad place to be.

Has anyone ever seen this? I asked Automation Technologies on Wednesday, 2/8, and they have yet to answer.


Bob