Pm-940m-cnc (pre-assembled)

Hi Sal,

Great! It looks like there is nothing wrong with your equipment, the PM940M nMotion and VFD are working and so is Mach3! By my count you must have about everything working the way you want it!!! Sounds like you are ready to make something cool. Congratulations!

However, since you posted the two videos, and I see your G-code changed between the two runs, I think I realize some of the confusion. You must be trying to use two different pulley set ups in Mach3. This is as should be if you are trying to have your machine calibrated to the world i.e. other folks g-code.

Clearly, the gear box label on the PM940M is screwball! The label reads: position 1 and indicates a speed from ~90 to 980 rpm and for position 2 it indicates 1000 to 3200 rpm. This is incorrect! The reading for position 1 is ok, but for position 2 it should read something like ~290 to 3200 rpm. That is, the gear knob only changes the mechanical ratio by a factor of about 3.2x. So why would the lower limit not be ~90x3.2~290. As one turns the gear knob from position 1 to position 2 the intermediate gearing is changing to make the spindle turn faster for the same motor speed. It is not magically interacting with the computer! The motor is turning at the same motor RPM range (which is a min to max as the frequency changes from the min to the max). So I prefer to think of the Mach3/nMotion signal to the VFD as simply 0 to 10 volts DC. 0 volts signal from the nMotion causes the VFD to put out an almost zero frequency 220 volt sinewave and when at 10 volts from the nMotion the VFD puts out the 220 volt drive at its maximum frequency. (I think this is around 400 cycles per second).

So in the Mach3 settings one has the choice of spindle speed settings via the choice of pulley settings. However, all these do is to cause the DC voltage going to the VFD to range. The settings in my Mach3 set up only uses a single pulley setting for a maximum speed of 1000 RPM. So Mach3/nMotion sends 10 volts to achieve the maximum spindle speed and 0 volts to cause the minimum spindle speed. So when the G-code setting is S1000 the Mach3 knows that this is the maximum speed and sends out 10 volts. For S500 it sends out 5 volts and so the spindle speed is 5/10 =1/2 of the max motor speed. With out changing any of the Mach3 setting nor any of the G-code If one moves the PM940M gear knob to position 2 the speed would be 3.2x faster.

So when I write G-code, I like to think of the S1000 value not as an actual speed, but as a percentage x 10 of the maximum possible spindle speed. When I write my G-code, I know that S1000 will cause a spindle speed of 1000 rpm when on gear box knob setting #1, but the same S1000 code will cause a spindle speed of 3200 rpm when the gear box knob setting is #2.

Additionally, I am not for sure why PM puts the lower limit on the gear box 2 position to be ~1000 rpm. Clearly, one can run it at slower speeds than 1000 rpm while the gear box is on the high setting.

My only thought on this is that motors are designed to run at 60 Hz. Not 0 nor 400 Hz. So we know that this motor will not be as efficient or have as much torque at the extreme low or high frequencies of the VFD, but as long the motor does not burn out it is ok. The motor getting hot depends more on the load that you put it under than the speed. Clearly if you try to turn the motor very slowly while the gear box is set to the high position, 2, then the motor may stall and then motor will over heat quickly. However, my guess is that when you are trying to use the high spindle rpm values (gear setting 2) you are not going to be turning large diameter tools into steel anyway! You are probably cutting wood or plastic or other soft materials.

One last comment. You may not recall, but I measured my spindle speed verses the Mach3 programmed speed. At S0 setting the spindle turns slowly, a few rpm, maybe 50. The Mach3 speed control does not really start to kick in until spindle is going at about 100 rpm. When at S1000, full speed, the spindle speed was a few percent, 5-10% off from what it was suppose to be. Likewise, while not terrible, it was not linear between the Mach3 settings and the actual speed. So, one should not worry about the exact speed all that much.

Have fun and let me know how things are going.

Dave L.
 
Hi Sal,

Great! It looks like there is nothing wrong with your equipment, the PM940M nMotion and VFD are working and so is Mach3! By my count you must have about everything working the way you want it!!! Sounds like you are ready to make something cool. Congratulations!

However, since you posted the two videos, and I see your G-code changed between the two runs, I think I realize some of the confusion. You must be trying to use two different pulley set ups in Mach3. This is as should be if you are trying to have your machine calibrated to the world i.e. other folks g-code.

Clearly, the gear box label on the PM940M is screwball! The label reads: position 1 and indicates a speed from ~90 to 980 rpm and for position 2 it indicates 1000 to 3200 rpm. This is incorrect! The reading for position 1 is ok, but for position 2 it should read something like ~290 to 3200 rpm. That is, the gear knob only changes the mechanical ratio by a factor of about 3.2x. So why would the lower limit not be ~90x3.2~290. As one turns the gear knob from position 1 to position 2 the intermediate gearing is changing to make the spindle turn faster for the same motor speed. It is not magically interacting with the computer! The motor is turning at the same motor RPM range (which is a min to max as the frequency changes from the min to the max). So I prefer to think of the Mach3/nMotion signal to the VFD as simply 0 to 10 volts DC. 0 volts signal from the nMotion causes the VFD to put out an almost zero frequency 220 volt sinewave and when at 10 volts from the nMotion the VFD puts out the 220 volt drive at its maximum frequency. (I think this is around 400 cycles per second).

So in the Mach3 settings one has the choice of spindle speed settings via the choice of pulley settings. However, all these do is to cause the DC voltage going to the VFD to range. The settings in my Mach3 set up only uses a single pulley setting for a maximum speed of 1000 RPM. So Mach3/nMotion sends 10 volts to achieve the maximum spindle speed and 0 volts to cause the minimum spindle speed. So when the G-code setting is S1000 the Mach3 knows that this is the maximum speed and sends out 10 volts. For S500 it sends out 5 volts and so the spindle speed is 5/10 =1/2 of the max motor speed. With out changing any of the Mach3 setting nor any of the G-code If one moves the PM940M gear knob to position 2 the speed would be 3.2x faster.

So when I write G-code, I like to think of the S1000 value not as an actual speed, but as a percentage x 10 of the maximum possible spindle speed. When I write my G-code, I know that S1000 will cause a spindle speed of 1000 rpm when on gear box knob setting #1, but the same S1000 code will cause a spindle speed of 3200 rpm when the gear box knob setting is #2.

Additionally, I am not for sure why PM puts the lower limit on the gear box 2 position to be ~1000 rpm. Clearly, one can run it at slower speeds than 1000 rpm while the gear box is on the high setting.

My only thought on this is that motors are designed to run at 60 Hz. Not 0 nor 400 Hz. So we know that this motor will not be as efficient or have as much torque at the extreme low or high frequencies of the VFD, but as long the motor does not burn out it is ok. The motor getting hot depends more on the load that you put it under than the speed. Clearly if you try to turn the motor very slowly while the gear box is set to the high position, 2, then the motor may stall and then motor will over heat quickly. However, my guess is that when you are trying to use the high spindle rpm values (gear setting 2) you are not going to be turning large diameter tools into steel anyway! You are probably cutting wood or plastic or other soft materials.

One last comment. You may not recall, but I measured my spindle speed verses the Mach3 programmed speed. At S0 setting the spindle turns slowly, a few rpm, maybe 50. The Mach3 speed control does not really start to kick in until spindle is going at about 100 rpm. When at S1000, full speed, the spindle speed was a few percent, 5-10% off from what it was suppose to be. Likewise, while not terrible, it was not linear between the Mach3 settings and the actual speed. So, one should not worry about the exact speed all that much.

Have fun and let me know how things are going.

Dave L.

@B2 you've been our guide and savior! @chocadile and I can't thank you enough for all the help you've given us.

Only thing left was to convert the inches to mm for my motor tuning and voila we've had successful passes on our wood stock.

Slowly getting used to FreeCAD for using the Mill and imagine it will take some time.

Again, thanks a bunch. We're very excited to have this running!

 
Hi Sal,

Great to hear from you and to see your movie. You are welcome for any assistance I may have provided. I look forward to hearing about your builds and tricks to make my machine run better. Also, I also am still feeling my way with FreeCad. Any FreeCAD tricks you find that are handy I would like to know about.

Regards,

Dave

PS.
I think I may have mentioned it, but I am now using the Linuxcnc, FreeCad Path converter. Several of them seem to work to some degree ... with slightly different results, but Linuxcnc seems to work as well as any of them. Some do not work at all.

These also seem to work: Centroid, grbl, Phillips (some differences, uses X,Y, theta, R and seems to interchange Y with Z), smoothie (somewhat, speed seems screwed up). Dynapath works but has some problems in pre and post. Did not work at all for me: openshop, rml, comparams.

None of them are good at some functions and the more complex the shape, the more wasted cuts they seem to make. Lots of jumping around on the part. I tend to make a first past at a drawing and path generation with FreeCad and then go into the G-code and fix it up .... sometimes a lot.
 
Circling back to the conduit. I just found some on Amazon that comes in reasonable lengths.
Would you be able to tell me the CL-to-CL distance on the holes at the bottom of the cast iron base of the PM940? I am trying to attach a flat bar to these holes and would love to confirm the distance. My rough measurement is 16".

Thanks
 
When I got my 940M-VS-CNC the base was not mounted. I flipped it over and tried to measure the base holes and locations. My general impression was that they were not on a square and that they were crooked compared to the outer edges of the base. They were just rough holes in the casting. I.e. they were just some holes/sticks left in the poor quality sand casting mold that the iron was poured into. Anyway, after a lot of worry about how to fasten a base plate to the stand holes I made a compromise and made the holes in my mounting base plate to be on a rectangular pattern with a spacing for the center to center hole spacing at 15.8" x 17.8" I drilled 5/8" diameter holes . However, since these bases are probably made from individual sand molds and the holes in the sand mold were probably made by someone just shoving in some sticks there is no reason that they should all be the same. Anyway, it has been several years now, but if I recall I drilled 5/8" through holes in my base plated and then used 1/2" bolts to go through the stand holes and the base plate holes and this was enough extra space that it all fit.

On to my base plate I mounted casters to be able to wheel the mill around. Once it is in place I block it to prevent any motion during operation.
You can see a picture of the final casters and base plate with the mill stand etc mounted here:


Yes, I too purchased a PM-940M CNC VSD (Pre-assembled) with all 4 axes.
If you really want I can send you some drawings I made for my base plate and how it was assembled for the casters. I made the base plate from 4 flat bars and bolted them together via the casters. I did all this drilling using the CNC mill before it even came off of the pallet and it was sitting on the pallet on the garage floor! However, if I were to do it over I would just purchase a sold steel plate of the right size and then have someone plasma cut it with a void in the center to reduce the weight and the holes. I think @ptrotter did something like this after we talked about it.

Dave L.
 
When I got my 940M-VS-CNC the base was not mounted. I flipped it over and tried to measure the base holes and locations. My general impression was that they were not on a square and that they were crooked compared to the outer edges of the base. They were just rough holes in the casting. I.e. they were just some holes/sticks left in the poor quality sand casting mold that the iron was poured into. Anyway, after a lot of worry about how to fasten a base plate to the stand holes I made a compromise and made the holes in my mounting base plate to be on a rectangular pattern with a spacing for the center to center hole spacing at 15.8" x 17.8" I drilled 5/8" diameter holes . However, since these bases are probably made from individual sand molds and the holes in the sand mold were probably made by someone just shoving in some sticks there is no reason that they should all be the same. Anyway, it has been several years now, but if I recall I drilled 5/8" through holes in my base plated and then used 1/2" bolts to go through the stand holes and the base plate holes and this was enough extra space that it all fit.

On to my base plate I mounted casters to be able to wheel the mill around. Once it is in place I block it to prevent any motion during operation.
You can see a picture of the final casters and base plate with the mill stand etc mounted here:



If you really want I can send you some drawings I made for my base plate and how it was assembled for the casters. I made the base plate from 4 flat bars and bolted them together via the casters. I did all this drilling using the CNC mill before it even came off of the pallet and it was sitting on the pallet on the garage floor! However, if I were to do it over I would just purchase a sold steel plate of the right size and then have someone plasma cut it with a void in the center to reduce the weight and the holes. I think @ptrotter did something like this after we talked about it.

Dave L.
I would love to take a peek at your drawings for this machine. I have a design, but it would help to see what you did.
Thanks for taking the time to answer.

Salah
 
Here are some snapshots from solidworks for the PM940CNC cast iron base.

PM940CNC-Casters-2.png
PM940CNC-Casters-1.png
PM940CNC-Casters-3.png


PM940CNC-Casters-4.png


I am using CarryMaster ALC-1000-FB

CarryMaster-4..jpg
CarryMaster-2.jpg


My current measurement for CL-to-CL of the mounting holes is 16". I can check more accurately once I jack the machine up and remove the old leveling feet. I might slip a 2x4 under the 2 holes (2 front and 2 back) and center-punch it with a custom-made transfer punch 1/2" in diameter.

The machine is already mounted on top of the base making it hard for me to take that measurement.


IMG_0633.JPEG
IMG_9064.JPEG
 
Last edited:
Hi Salah, Are you related to Sal_the_Man who I was working with on this string some time ago?

Nice photos and CAD sketches. Did you repaint the base. Mine is green! I will come back to the 940 base plate and casters, but I was curious about when you got your 940M-CNC so I looked at some of your other posts and found your old comment regarding a PM1340 lathe:
I am trying to figure out what the calculation is. It is not the reciprocal of the TPI. I seem to be missing a ratio of some sort to get from the TPI to the feed rate. Can you help?
If you are interested in TPI and Feed rates for your 1340 you might want to check out the spread sheet tool I spent quite a bit of time building. I actually works for any lathe that I have seen by setting up a template for the particular lathe. However, I already did this for my 1440 and the 1340 and a few other similar lathes. From this template the excel VBA macros will generate a table of ALL possible TPI, Feed, x-Feed rates plus the metric versions. You can add or remove external gears that you might have or not. For my 1440 with the supplied external gears there turns out to be over 6000 combinations and so a lot of TPI and feed values. Anyway, these all appear in a spread sheet table. I also supplied macros to search as well as sort the table etc. You can even search with wild cards so that if the TPI value you are after cannot be exactly made you can see how close you can get.... Anyway, if you have seen my posting then maybe you have already tried out the spread sheet.

I am not getting much feed back on the excel tool and would like some so if you try it out please let me know what you think.

Now, wrt to the 940M base plate. I am afraid I did this back in 2016-17 time frame and the only cad I had at the time was Microsoft VISIO, which was only 2D. Do you have access to Visio? If so I could send the drawing. For now I will just generate a pdf file of the over view page for you. Since all I was doing was drilling lots of holes plus tapping I did not need 3-D. Since I was moving the mill into my basement, I took 4 bar stock (two widths) of about 3/8 thickness steel and laid them out on a square such that the ends overlapped. I then drilled lots of holes to mount to. At the corners I drilled the holes for the casters and these then both held the casters and held the four plates together. "These were all 1/4-20 tapped straight through so that the bolts were essentially flush where they were protruding out. Of course I drilled other holes as well. I was moving the mill into my basement, so I let this "dolly" do double duty. I drilled extra holes and tapped them so that I could first mount the mill directly to the dolly and bring it in. Then I took this off and them mounted the dolly to the mill stand before placing the mill on the mill stand. Since the mill and the mill stand do not have the same foot print there are a double set of drilled and tapped holes in the dolly. The ones that were used with the mill are no longer being used. This required that I move the casters around so instead of 4 tapped holes for each caster there were 8 etc.... So this was my first CNC program to drill all of these holes. Of course the whole base plate would not fit on the mill table at one time so I did each of the 4 pieces of metal separately and let the CNC get the holes perfectly placed!

Anyway, a pdf file is attached.

This is not rocket science and your sketches look good. However, I think that ptrotter's approach is far simpler than either of ours. One big plate that sticks out on the sides like your design to accommodate the casters. I think he used 1/2 plate steel. Had them, where he purchased the steel, cut the OD via plasma an had them also cut out the center to eliminate weight. He then drilled holes, but he told me that as good as the plasma cuts were he probably would have them drill the holes with the plasma cutter as well.

Good luck

Dave L.

PS. I am assuming that you got this about the same time I purchased mine. Have you had any problems with the head being too heavy for the small z-axis motor to hold it up when the power is off? Unless I over tighten the gib mine walks down until the cutting tool or head rests on the table etc. Not good as it also looses the CAD position.
 

Attachments

  • PM940 Mill Caster Mount design HC020_1300.pdf
    496.9 KB · Views: 159
Hi Salah, Are you related to Sal_the_Man who I was working with on this string some time ago?

Nice photos and CAD sketches. Did you repaint the base. Mine is green! I will come back to the 940 base plate and casters, but I was curious about when you got your 940M-CNC so I looked at some of your other posts and found your old comment regarding a PM1340 lathe:

If you are interested in TPI and Feed rates for your 1340 you might want to check out the spread sheet tool I spent quite a bit of time building. I actually works for any lathe that I have seen by setting up a template for the particular lathe. However, I already did this for my 1440 and the 1340 and a few other similar lathes. From this template the excel VBA macros will generate a table of ALL possible TPI, Feed, x-Feed rates plus the metric versions. You can add or remove external gears that you might have or not. For my 1440 with the supplied external gears there turns out to be over 6000 combinations and so a lot of TPI and feed values. Anyway, these all appear in a spread sheet table. I also supplied macros to search as well as sort the table etc. You can even search with wild cards so that if the TPI value you are after cannot be exactly made you can see how close you can get.... Anyway, if you have seen my posting then maybe you have already tried out the spread sheet.

I am not getting much feed back on the excel tool and would like some so if you try it out please let me know what you think.

Now, wrt to the 940M base plate. I am afraid I did this back in 2016-17 time frame and the only cad I had at the time was Microsoft VISIO, which was only 2D. Do you have access to Visio? If so I could send the drawing. For now I will just generate a pdf file of the over view page for you. Since all I was doing was drilling lots of holes plus tapping I did not need 3-D. Since I was moving the mill into my basement, I took 4 bar stock (two widths) of about 3/8 thickness steel and laid them out on a square such that the ends overlapped. I then drilled lots of holes to mount to. At the corners I drilled the holes for the casters and these then both held the casters and held the four plates together. "These were all 1/4-20 tapped straight through so that the bolts were essentially flush where they were protruding out. Of course I drilled other holes as well. I was moving the mill into my basement, so I let this "dolly" do double duty. I drilled extra holes and tapped them so that I could first mount the mill directly to the dolly and bring it in. Then I took this off and them mounted the dolly to the mill stand before placing the mill on the mill stand. Since the mill and the mill stand do not have the same foot print there are a double set of drilled and tapped holes in the dolly. The ones that were used with the mill are no longer being used. This required that I move the casters around so instead of 4 tapped holes for each caster there were 8 etc.... So this was my first CNC program to drill all of these holes. Of course the whole base plate would not fit on the mill table at one time so I did each of the 4 pieces of metal separately and let the CNC get the holes perfectly placed!

Anyway, a pdf file is attached.

This is not rocket science and your sketches look good. However, I think that ptrotter's approach is far simpler than either of ours. One big plate that sticks out on the sides like your design to accommodate the casters. I think he used 1/2 plate steel. Had them, where he purchased the steel, cut the OD via plasma an had them also cut out the center to eliminate weight. He then drilled holes, but he told me that as good as the plasma cuts were he probably would have them drill the holes with the plasma cutter as well.

Good luck

Dave L.

PS. I am assuming that you got this about the same time I purchased mine. Have you had any problems with the head being too heavy for the small z-axis motor to hold it up when the power is off? Unless I over tighten the gib mine walks down until the cutting tool or head rests on the table etc. Not good as it also looses the CAD position.
Hello Dave,

No, I am not related to Sal-the-Man. Now you got me curious about this thread and I intend to read it all from the begining.

I got my PM940 CNC in 2018. I did have the problem of the head moving down on its own without power and I also felt the z-axis motor is underpowered. So, I changed it to a higher torque one. I think this machine will benefit from a counter-weight or a strut balancer as the head is quite heavy. I did many fixes: changed the controller to a centroid, I completely rebuilt the head gearbox, and rewired the cabinet. I have designed a tooling quick-change system similar to Tomarch's but I have not completed all the parts for it. Right now, I want to move the machine to a different location and build a full enclosure as I intend to run flood coolant.

At some point, I will do a write up on this machine and will post it here. I will let you know when this happen.

Thank you for sharing your ideas and the drawing. I do not have visio, but the pdf you provided gave me a good idea of what you did. I understand it.

I am very interested in the TPI calculator. I will for sure take a look and will try to use it. I will send you feedback once I do. Tomorrow, I will start reading your thread and will post some updates as I start the drilling and tapping of my contraption.

Thanks again for taking the time to explain what you did and for sharing your work.

Salah
P.S. I did repaint the base as I hated the blue color it came with.
 
Back
Top