[rs4race]

Thanks, I emailed their tech support this morning. Depending on the cost of components I might make a signal converter myself. I found a schematic online. Its in this document http://www.cjh.com.au//PWM%20Spindle...ng%20Mach3.pdf

I will be waiting to see what leadshine says about this. It would be nice to have a quick easy solution.

[rs4race]

Sorry no updates on spindle speed control, Leadshine stopped talking to me once I asked them about getting the MX3660 repaired. I have been working on a bunch of parts and havent had time to look more into it.

I did figure out how to get a decently quick and easy tool change though. I use tool holders on a straight shank similar to tormach. I set up tool height offsets in mach and then in my G code I use the following when changing tools:

T0 M6 (Change to tool 0 in mach)
G43 (Adjust DRO for tool height offset)

M03 (Start spindle)
TOOLPATHS (G-Code..)

G0 Z2.0 (rapid to safe Z height for tool change)
X-3 Y1.0 M05 (Move to tool change position, M05 Stops Spindle)

M00 (Pauses G-Code until you press 'Run' again)

T1 M6 (Change to tool 1 in mach)
G43 (Adjust DRO for tool height offset)

M03 (Start spindle)
TOOLPATHS (G-Code.. Repeat tool changes if necissary)

M02 (End of G-Code)


This has made things pretty easy since I dont have to manually change tool number in mach and then load another g-code file. I hope this might help some others that have run into this problem. I had a hard time finding out how to do this, and the code I had found was backwards, G43 T0 M06, and didn't adjust the DRO as needed.

[SCzEngrgGroup]

Sorry no updates on spindle speed control, Leadshine stopped talking to me once I asked them about getting the MX3660 repaired. I have been working on a bunch of parts and havent had time to look more into it.

I did figure out how to get a decently quick and easy tool change though. I use tool holders on a straight shank similar to tormach. I set up tool height offsets in mach and then in my G code I use the following when changing tools:

T0 M6 (Change to tool 0 in mach)
G43 (Adjust DRO for tool height offset)

M03 (Start spindle)
TOOLPATHS (G-Code..)

G0 Z2.0 (rapid to safe Z height for tool change)
X-3 Y1.0 M05 (Move to tool change position, M05 Stops Spindle)

M00 (Pauses G-Code until you press 'Run' again)

T1 M6 (Change to tool 1 in mach)
G43 (Adjust DRO for tool height offset)

M03 (Start spindle)
TOOLPATHS (G-Code.. Repeat tool changes if necissary)

M02 (End of G-Code)


This has made things pretty easy since I dont have to manually change tool number in mach and then load another g-code file. I hope this might help some others that have run into this problem. I had a hard time finding out how to do this, and the code I had found was backwards, G43 T0 M06, and didn't adjust the DRO as needed.

For G43 to work, you must also give an H-word that tells it WHICH tool length offset to use. For example: "G43 H6" says use the Tool 6 length offset. The tool number in the H-word will typically be the same as the one in the M6, but it doesn't have to be. Use G49 to turn off tool length offsets.

M6 T3 G43 H3
T3 H3 G43 M6
G43 M6 T3 H3

or any other permutation of those same four "words" will all perform identically. The order is immaterial, as the execution order follows strict rules of precedence, so there is no "backwards".

You also don't need to put the G0s in the G-code. There is an option under Config->GeneralConfig to do manual toolchanges, which will cause Mach to do an M0 automatically whenever it executes an M6. There is also a setting for toolchange location, so you don't really need to include that in the G-code either.

Regards,
Ray L.

[rs4race]

Thanks Ray, I will check general config for those settings.
The weird thing is though that when trying to include g43 on the same line it will not change the DRO. Adding it to the next line works, with no H value either. Reading about g code definitions corresponds to what your saying. Maybe I need to go back and double check.

I did try setting tool change positions before, and because I don't have home switches hooked up yet I crashed a couple tools, then disabled it. I recently found the reference all home button so I might give it another chance. At least I will give it a try with a dry run.

In your example with "G43 H6" does the H6 mean 6 inches or will that reference tool 6 height offset?

[jseehorn]

H6 references tool 6. It will load whatever you have saved in the table for the tool # you use.

[rs4race]

Thanks for the info. I use NX to generate my tool paths if I don't do it by hand. I want to start using something else as NX seems to waste alot of time, but thats another story. It has a section of code that always would want to crash my tools that looks something like G43 Z0.0 H00. I thought the H00 was a offset in inches. I guess I will go back and plug in the tool numbers and see how that works. Then I don't really have to edit any code. I'm going to be back out in the garage working on more parts today I think I will try the tool change position and the H phrase in my code. It's weird that it worked without it.


I'm almost finished with these parts.

[rs4race]

It's been awhile since I posted here but I have a few updates.

First off I recently got a bunch of tormach tts tool holders. This is a really nice upgrade since they are much more rigid than my previous tool holders. I also finally got a clamping kit. And a few cutting tools.

I also moved from NX to HSMWorks. Oh wow how nice it is to model and generate tool paths in the same program and have it all in one file. Combined with HSMAdvisor I have been saving a lot of time running bigger DOC, smaller WOC and higher feed speed.

That kinda leads into my real update. I have been trying to get my speed up on the z axis, also I want to drop the backlash there. Well after some inspection it appears the machined end of my ball screw isn't concentric with the screw. I bought a 20mm screw on eBay with a double nut which seems to be much nicer. It's a little long so I will have to cut 50mm or so off the end. The machined end on this is 15mm instead of 17mm so I also ordered some AC bearings to replace the thrust bearings I have been using.

I will need to make a new ball screw adapter, and bearing mount. Good for me, I can bore out the bearing mount that came with my original ball screw kit. I kinda want to use steel though, but not sure if there's enough benefit.

I will post some pics of the z axis conversion. Also if anyone has dimensions for the ball screw adapter do you mind sharing?

[SCzEngrgGroup]

I kinda want to use steel though, but not sure if there's enough benefit.

Steel will make absolutely no difference whatsoever.

Regards,
Ray L.

[Fastest1]

Call Leadshine, They have tech support out of California.

[rs4race]

I didn't have much time to work, but I fixed the bearing pocket for the z axis. I took it from 35mm to 42mm. I also cut a short pocket on the bottom for the bottom bearing and bolted the ball screw to the bearing block.

I used a dial indicator to find the center of the hole. Which gets interesting when the hole isn't round.



I didnt get a picture of the new bearing hole but I got a couple of it all together. This ball screw and nut is a lot more beefy than the previous.





I might end up making a new bearing block soon depending on how this one works out.. I might just make a new one just because this one isn't so pretty. Next up is some motor mounts, and a ball nut adapter.

[rs4race]

Hey I figured out the mx3660 0-10V output. Turns out my pin 14 had a loose connection. I pushed on it and the voltage showed up. I'm not sure if I set it up wrong but at 0 rpm its full voltage and full rpm its 0V. I might have a chance to test it tomorrow after work.

I also finished the ball nut adapter for the 20mm screw and will be working on the bearing block soon. I will post up pics and drawings for it after I test it.

I had been doing some research, looking at the Tormach mills. I found the 1100 uses a 1.5 hp , 90 framed 230v 3PH AC motor, and 770 uses a 1 hp , 71 size , 230V 3PH AC motor. I think the g0704 is probably closer to the 770 as far as size so I'm leaning towards the 1hp motor. I found a source in europe, but not sure if shipping is going to kill that for me.

I also made the mistake of buying a 1200 oz in NEMA 34. I have to see about returning it in exchange for another. After some research I found a 640oz in NEMA 34 that will work with my drives and better suited for the feeds I want. To go with that I'm thinking about bumping up the power supply to 60V. The 640oz Nema 34 has roughly the same torque below 400rpm (~80ipm) , but 2-3 times more above that. Sounds good to me.

I also really want to put some home and limit switches on and working on getting that together.

[Fastest1]

I didn't have much time to work, but I fixed the bearing pocket for the z axis. I took it from 35mm to 42mm. I also cut a short pocket on the bottom for the bottom bearing and bolted the ball screw to the bearing block.

I used a dial indicator to find the center of the hole. Which gets interesting when the hole isn't round.



I didnt get a picture of the new bearing hole but I got a couple of it all together. This ball screw and nut is a lot more beefy than the previous.





I might end up making a new bearing block soon depending on how this one works out.. I might just make a new one just because this one isn't so pretty. Next up is some motor mounts, and a ball nut adapter.

Judging from this top picture. I would guess you have little backlash in your Y axis. Do you run backlash comp? Or did you at this time?

[rs4race]

Judging from this top picture. I would guess you have little backlash in your Y axis. Do you run backlash comp? Or did you at this time?

That's the part that came in the conversion kit I bought. Such bad quality. After repacking the ball nuts on the x and y I have 1-2 thou backlash on those and compensate with backlash comp. I probably should recheck. After sitting all winter and summer I just oiled the ways and started cutting the ball nut adapter. There's slight signs of backlash right now but not nearly as bad as that!

Here's my new ball nut adapter. This is quite a bit beefier than it needs to be but not too heavy. This is the first time I milled a chamfer. I was pretty excited when the chamfer started, very cool! I did this in three fixtures and didn't quite get it zeroed when I flipped it. It worked out though since I had to open the center bore to give the ball nut some clearance. Nice slip fit.

[lcvette]

How thick is the portion the bearing is resting on? Looks very thin, the entire head will be pulling on the thickness of that bearing ledge.. I could imagine a bit of deflection if it's only as thick as it appears in the picture.

As affordable as actual bearing supports are I decided to try them and was rewarded with improving backlash from .00235" to .00113" with a standard single nut on the Y axis, and similar on x and z. I also have a set of double nuts I will be installing in a few weeks along with a z axis linear rail conversion.

I also noticed my acis movement were significantly quieter with the actual bearing block supports. It is possible that the bores were off center slightly on each side which created the issues. Looking at the plate you posted I imagine you may have a similar situation happening.

I should say also I made the parts that were off myself so no one to blame but me, but I wanted to share because if others have the same issue then there is some more accuracy to be had from your machines. I look forward to seeing how low I can get the backlash with the double nuts and seeing if they can be further tweaked to get it down to zero. Also hoping the linear rails add some stiffness and assist in the getting that backlash number down. I look forward to being able to make some nice round interpolated holes without the transition oddities from axis reversals.

By the way the rest of the build looks really good!

Sent from my XT1080 using Tapatalk

[rs4race]

So that plate originally had a 1/16 in flange the bearing was resting on. Really thin. When I bored it out for the larger bearings I left a 1/8 flange between the bearings. It's deeper in the part too so it should be stiff enough. I dont really like the finish of the outside of the part so I'm thinking about making a new one, just waiting for a piece of scrap to come by the bin at work.

I have some AC bearings for the X and Y axis also that I'm planning to get to after I get the Z done and spindle speed done in mach.

When your talking about bearing supports are you talking about the z axis? The only axis I don't have supported both ends is the z axis. I thought of possibly supporting that but was going to check the improvement from the double nut first.

[lcvette]

I was referring to Precision bearing blocks such as BK, BF, FK, FF and the following number designator is the bearing ID. They have a very good Precision finish that lines the bearings up perfectly and a nice preload nut with lock so you don't have to fight with a double but setuo which I found difficult to get just right and often times "unlocked" at inopportune moments ruining parts.

If you heck Hiwin's website, go to the ballscrew section and look at the peripherals section, they have 3 variations of ballscrew end supports. They come preloaded. For the Z axis i used an FK17 on a plate along with keeping the cast iron top plate. I was amazed at how much better these performed then the home built version I had made (again mine were misaligned but for this to work correctly they should be perfectly aligned with no movement possible for the bearings).

I ordered mine from china along with my ballscrews. They are relatively cheap for the benefit they provide. Think I paid like $26 for the FK17, I used BK and BF 12's for the X axis a set was only $17, and I used FK12 for the Y axis $24. Not sure why the FK series are higher, but they fit my support blocks design the easiest and was well worth it.

The fact that they are in housing that are perfectly square is very helpful. They also allowed me to use larger diameter Oldham couplers. I was never a fan of the 1/4" shafts to begin with. I switched to metric motors also which had 10mm shafts which were much beefier and less prone to move in the coupler on such a small diameter shaft.

Sent from my XT1080 using Tapatalk

[rs4race]

Those are actually pretty close to what I was thinking of. I was going to use shims to preload the bearings to one another. I think I would save a few bucks going that route though.

Do you have a build thread lcvette? I wouldn't mind checking out your setup.

[Fastest1]

The ballnut mount came out very well.

It is fun to watch a successful run of a new operation. Chamfering is essential to pretty parts. I had a CAD instructor years ago say"if you have to touch it by hand, your being lazy". Have you tried threadmilling yet? Quite unreal to witness.

[lcvette]

Yes.. I love thread milling... It's ma favorite...lol

Rs, I don't have a build thread, but I do have some vids of the machine in action making a new stepper mount for a nema 34 for a buddy.

Speaking of thread milling... Here is a little G0704 thread milling after chamfering some hole entrances..

https://youtu.be/8QItjpeTwRE

Sent from my XT1080 using Tapatalk

[deuce217]

I like the Eboard you are building! I am not into skating but want to build my own motorized longboard doing a similar setup. Thanks for sharing.