PM932 CNC Build

[John Hasler]

John - Yes it is on XP. It was a clean install of XP pro. The PC only runs Mach3 and is not connected to a network, no anti virus etc. I have disabled as many unneeded processes as possible but there could be some I missed. Not sure what I could do to challenge the software while homing.

Copy a large file.

 

[jbolt]

Copy a large file.

Thanks, I will give it a try.

 

[JimDawson]

OK, I checked the repeatability of my limits. As close as I can tell they are dead on 100% of the time using a 0.001 reading indicator. I'm using about 20% speed when I'm homing. My motor speeds are set at max 70 IPM, so 20% speed = 14 IPM. I'm using about 15 degree cams to actuate the switches.

 

[jbolt]

Project Update:

I had a mishap with the mill where during a 2" deep contour run a piece of material that should have fallen away got sucked into the cutter (3/4" end mill). The 1/2 thick piece of aluminum jammed between the part and the cutter stalling the spindle. Running at 20 ipm it moved off the part before I could e-stop. The spindle is fine, the x-axis (the moving axis at the time) is fine but the Y-axis, which was not moving now has a dead spot in the motor. The drive also makes the motor very noisy. At low rpm the motor hesitates as it goes through the dead spot, at moderate speed it vibrates like crazy and at full power it runs like normal. I swapped the X & Y drives and the Y drive makes the X motor sound noisy. The Y motor on the X drive is less noisy but still has the dead spot. I'm not really sure what was damaged but for now I just plan on replacing the motor and drive.

I pulled the table off to inspect the ball screws and everything there is good. I did discover that coolant was getting into places I didn't think it could. I use Rustlick ws-505o at about 7.5:1. It's just amazing where this stuff will work itself into let alone its ability to remove paint and adhesive.

While I have the machine apart I extended the notch for the Y-axis in the base to pickup 2 more inches of Y travel and also machined some reliefs in the bottom of the table to gain 3 more inches of X travel. The 20 mm ball nuts are too tall to fit under the thickened casting under the end wells on the table. I was able to safely remove enough material to get the extra travel I had originally lost. It seems you can never have enough travel.

Other thing I discovered is that the Z-ball nuts were not getting oil through the one-shot oiler. Once primed the system drains back through the X and Y ball nuts. To fix this (hopefully) I have installed in-line check valves on all the supply lines to the ball nuts and ways.

One of the sheet metal sponsors for the high school robotics team I help mentor is making me a new one piece splash shield with drain gutter to fit around the front and sides of the table. In addition to that I will be replacing the factory way covers which have not held up well to the flood coolant, with a single sheet of 1/8" rubber. One end will be attached to the cross slide and the other to the Z dovetail plate. I'm going to start with an 18" wide piece to see how it works. My only concern is how it will bunch when the Table and Z are at their closest to each other. The rubber sheet is cheap compared to high quality way covers so its worth a try.

Also in the mix is:

Changing out the Chinese BOB and UC100 USB controller for a PMDX-126 BOB, Eithernet SmoothStepper.
Upgrading the spindle bearings.
Changing to a belt drive and VFD.
Adding a no-mist system
Re-doing the control box that houses the motor drive, power supplies and BOB so the components can be removed/replaced without having to completely dismantle the box. (poor original planning)
Trying out some mechanical slow acting switches for my Home/limits to see if the work better than the proximity sensors.

Here are a coupe of photos of the underside of the table being milled on the trusty old Smithy Granite 3-1.

Jay

 

[jbolt]

OK, I checked the repeatability of my limits. As close as I can tell they are dead on 100% of the time using a 0.001 reading indicator. I'm using about 20% speed when I'm homing. My motor speeds are set at max 70 IPM, so 20% speed = 14 IPM. I'm using about 15 degree cams to actuate the switches.

Jim, what type of plunger/actuator arm is on your limit switches?

Jay

 

[JimDawson]

They are roller lever type.

 

[jbolt]

The mill is once again operational.

The Y-axis motor and drive have been replaced. The head was knocked out of tram by 0.080" which surprised me.

I was able to reconfigure my control box that houses the drives , BOB and controller so all the components can be individually removed without having to disassemble half the box.

I installed a new PMDX-126 BOB, PMDX-107 spinde control card and an Eithernet Smooth Stepper (ESS) in place of the old generic BOB and UC100 controller. This was much more complicated to setup. The PMDX suffers in places from too much information in the manual and a little too cut up for my liking but I managed to fiddle my way through it. The ESS suffers from too little documentation. Hurray for online forums and Google! Anyway I really like how the ESS nests on top of the PMDX-126 as a daughter board. Very tidy setup. I am having an odd issue where if a limit switch is triggered the keyboard jog function no longer works unless the system is power cycled. Additionally, if I plugin the xbox controller the keyboard jog and fly-out jog become inactive? Fortunately the xbox controller works fine which is what I normally use for jogging anyway.

I installed the new coolant shields. The shield around the table is really slick and does a great job of directing coolant off the table without it getting into everything. The rubber way cover is also working out nicely. One end connects to the cross slide and the other end to the Z dovetail. To keep the rubber sheet from bunching behind the table I added an intermediate support that is suspended from the column. It's a little Rube Goldberg-ish but so far it is working well.

Next is to machine the motor mount for the belt drive conversion.

 

[Boswell]

When you hit a limit switch, are you selecting Limit Override in Mach3 before you try to jog back off the limit switch? Wait, are you using Mach3 ?

 

[TomS]

Project Update:

I had a mishap with the mill where during a 2" deep contour run a piece of material that should have fallen away got sucked into the cutter (3/4" end mill). The 1/2 thick piece of aluminum jammed between the part and the cutter stalling the spindle. Running at 20 ipm it moved off the part before I could e-stop. The spindle is fine, the x-axis (the moving axis at the time) is fine but the Y-axis, which was not moving now has a dead spot in the motor. The drive also makes the motor very noisy. At low rpm the motor hesitates as it goes through the dead spot, at moderate speed it vibrates like crazy and at full power it runs like normal. I swapped the X & Y drives and the Y drive makes the X motor sound noisy. The Y motor on the X drive is less noisy but still has the dead spot. I'm not really sure what was damaged but for now I just plan on replacing the motor and drive.

I pulled the table off to inspect the ball screws and everything there is good. I did discover that coolant was getting into places I didn't think it could. I use Rustlick ws-505o at about 7.5:1. It's just amazing where this stuff will work itself into let alone its ability to remove paint and adhesive.

While I have the machine apart I extended the notch for the Y-axis in the base to pickup 2 more inches of Y travel and also machined some reliefs in the bottom of the table to gain 3 more inches of X travel. The 20 mm ball nuts are too tall to fit under the thickened casting under the end wells on the table. I was able to safely remove enough material to get the extra travel I had originally lost. It seems you can never have enough travel.

Other thing I discovered is that the Z-ball nuts were not getting oil through the one-shot oiler. Once primed the system drains back through the X and Y ball nuts. To fix this (hopefully) I have installed in-line check valves on all the supply lines to the ball nuts and ways.

One of the sheet metal sponsors for the high school robotics team I help mentor is making me a new one piece splash shield with drain gutter to fit around the front and sides of the table. In addition to that I will be replacing the factory way covers which have not held up well to the flood coolant, with a single sheet of 1/8" rubber. One end will be attached to the cross slide and the other to the Z dovetail plate. I'm going to start with an 18" wide piece to see how it works. My only concern is how it will bunch when the Table and Z are at their closest to each other. The rubber sheet is cheap compared to high quality way covers so its worth a try.

Also in the mix is:

Changing out the Chinese BOB and UC100 USB controller for a PMDX-126 BOB, Eithernet SmoothStepper.
Upgrading the spindle bearings.
Changing to a belt drive and VFD.
Adding a no-mist system
Re-doing the control box that houses the motor drive, power supplies and BOB so the components can be removed/replaced without having to completely dismantle the box. (poor original planning)
Trying out some mechanical slow acting switches for my Home/limits to see if the work better than the proximity sensors.

Here are a coupe of photos of the underside of the table being milled on the trusty old Smithy Granite 3-1.

Jay

View attachment 96392 View attachment 96393

Jay,

I like what you are doing, especially the belt drive upgrade and upgraded spindle bearings. Waiting anxiously to see how you do these mods.

After extending the Y axis notch and removing the bellows how much total travel did you end up with?

Tom S

 

[jbolt]

When you hit a limit switch, are you selecting Limit Override in Mach3 before you try to jog back off the limit switch? Wait, are you using Mach3 ?

Yes and yes!

Jay