Precision Matthews PM-728VT CNC conversion

[adamyq]

Status update:

Got the X and Y motors installed, and movement. Still not properly configured (I.e. 1 inch in y command results in 2 inch movement), but I assume easily remedied when I get into the configuration. Note that I just connected my two CP servos to the 8P connector. I had seen several folks cut the cables and splice but didn’t understand why. That said, there may be a good reason.


Sent from my iPhone using Tapatalk

 

[adamyq]

Questions on the Z axis:

1. The cast housing is pretty dang tall, so I could maybe get ~60% engagement on each axis. Note that this is the eBrake shaft. See photo.

2. The flange holes on the brake are just a shade too small for the screws that come with the cast mount. Did anyone else have this issue? I’m talking like ~5.5mm jokes vs the 6mm hex caps. Easy enough to bore out, but figured I’d ask.

Sent from my iPhone using Tapatalk

 

[koenbro]

Questions on the Z axis:

1. The cast housing is pretty dang tall, so I could maybe get ~60% engagement on each axis. Note that this is the eBrake shaft. See photo.

2. The flange holes on the brake are just a shade too small for the screws that come with the cast mount. Did anyone else have this issue? I’m talking like ~5.5mm jokes vs the 6mm hex caps. Easy enough to bore out, but figured I’d ask.

Sent from my iPhone using Tapatalk

Yes and yes. Engagement is about 50-60% and the flange holes need to be drilled out.


Sent from my iPhone using Tapatalk Pro

 

[rwm]

Could you not machine the casting down shorter? I am surprised to see a Lovejoy coupling rather than a flex disc coupling? More slop in the Lovejoy.

 

[Firebrick43]

Could you not machine the casting down shorter? I am surprised to see a Lovejoy coupling rather than a flex disc coupling? More slop in the Lovejoy.

I haven't seen a disc type coupling on a CNC machine newer than the early 90's??? Even bellows are not used as much any more. Alot of couplings are now the Rotex style, which are very similar to lovejoys. It depends on the accuracy and the durometer of the spider.

 

[rwm]

You are probably right Firebrick. Hey there is a good discussion of couplers in this thread that shows the various types:

CNC PM-728V-T

I have completed the first phase of my CNC conversion. I had a couple questions in another thread, but didn’t receive any feedback so I started a new thread to hopefully get a few answers and some feedback. First, a little bit about my setup. It’s a pm-728vt with PM’s CNC kit. G540 running...

www.hobby-machinist.com

 

[adamyq]

There was no easy way to attach the sensors to the casting; tried to drill a hole, but I learned very quickly that that's not going to work. So I decided to reuse the tapped holes for the locking levers on the saddle. I designed a simple bracket in Fusion.
View attachment 356606

Then cut in on the router out of scrap 1/16" 6061.
View attachment 356608


Here it is mounted on the mill. The left sided sensor is already on; have added the second (+ limit ) sensor since.
View attachment 356609


I plan to add two sensors for axis Y and maybe one for axis Z (upper limit). Still haven't figured out how and where I mount the y-axis sensors, as I want to use existing holes in the casting.

More importantly, will need to figure out how to configure the Acorn to stop the motors. Right now, it shows on the screen that the sensor is activated, but the motors do not stop.

I’m on to this phase now. Did you make those black sensor clamps or find them somewhere? Also curious on the cable you used.

I’ll likely mimic your x mounting strategy. Did you come up with a y solution?


Sent from my iPhone using Tapatalk

 

[koenbro]

I’m on to this phase now. Did you make those black sensor clamps or find them somewhere? Also curious on the cable you used.

I’ll likely mimic your x mounting strategy. Did you come up with a y solution?

I changed the mounts and am using 3D printed ones. They better match the movement range in x minus. This is the x minus:

And it has an Aluminum block mounted on the saddle (?):

The reason I gave it an overhang is that chips flying have stuck to the sensor and tripped it on the previous iteration.

This is Z:

It It gets tripped by an 8020 angle connector.

And this is y. It is the most primitive but works:

 

[adamyq]

Wow, those look great. I’ll likely try and replicate your setup to leverage your trial and error. Would you mind sharing the files for the printed parts?

I’ve been debating the Y-, what drove your decision there? I’m likely going with the same sensors (thanks again), where did you source the 90 deg connector cables?

Sent from my iPhone using Tapatalk

 

[koenbro]

The choice of placing the y at minus had to do with the convenience of attaching the sensor; more tricky to mount at Y-plus.

The cables came from Automation Direct. A word of caution: depending of where you place your control box, you might need longer cables.

Happy to share the 3D print files, but not sure how to upload the *.obj mesh files here. I have sent a DM with the link.

These are the renders should anyone care: