New PM-25MV Mill

The enclosure is about set now except for the doors. The windows have been installed, they are just .093" acrylic sheet and have 6 holes drilled, then 6 matching holes are drilled into the MDF and screws and nuts are used to secure it in place. I made a few parts with the machine and even without doors, I am containing well over 90% of the chips. There are only a few cuts in a few directions that manage to send chips out the front of the enclosure. I got my LEDs installed and that makes a huge difference. They are cheap ebay units, they cost $10 for the 2 of them. They are wired directly into an AC line and come on when the machine is powered up.

I had an issue with the Z axis driver cutting out and dropping the head, which obviously is not what you want when you are spinning a cutter... I lost a part to that and I am not 100% sure why it happened. The drive wasn't hot, but after leaving the machine off overnight and running it with an extra fan blowing on the electronics, the machine cut for probably 6 hours (spindle time, probably 10 hours total on time) with no problem.

For the spindle draw bar, I wired a button into the box where the chip guard interrupt was. I simply took that out, shorted it so that it assumes the chip guard is always closed, then the hole was the perfect size for a button I had on hand. Just wired it in to the solenoid and now when you press the button, the tool releases, release the button and the tool is secured. I forgot to take a picture of that, but you can see it easily in a few of the other pictures I have.

I am going to need to come up with a better work holding solution, because the vise I have isn't ideal. Using the vise, the max usable Y axis travel is a little over 4". In one direction, the vise runs into the way covers, and in the other direction it leaves 2" unreachable past the front jaw. I am considering either trying to find a vise that doesn't stick out as much in the front or making something suitable instead. There are other methods of workholding, but a simple and effective one that cuts your travels by 40% isn't ideal.

I finished making an encoder mount to give the controller some feedback and hopefully increase the reliability a little bit. I know a lot of people aren't fans of encoders since "properly designed" stepper systems shouldn't need them, but I am not an expert, and this system is far from properly designed. $30 for 3 encoders and a few hours of machine time to prevent lost parts due to stalls, drives overheating, etc is worth it for me.

Up next is some pulleys for the steppers and encoders, some motor covers, making the doors, a control panel, and continuing on the ATC. I have some of the parts machined already, now I need to figure out the air cylinder for it and the sensors. I got an inexpensive 7" usb monitor that I think should suffice for the control panel, along with some buttons and switches. I was looking into a motion controller for the machine, but I wasn't able to find one that would run linuxcnc for a decent price. Instead, I may just try to replace the control computer with something a little better.

Here are the LEDs I got. They are 10w each and are plenty bright.
Dmkbfkv.jpg


It makes it nice and bright inside. In this picture you can see the button on the side of the head for actuating the drawbar.
mCP8cM4.jpg


Here is a side view through the window. There is some glare off the LED and the cracked camera glass, but you get the idea. The wiring needs to be cleaned up, but in due time. I plan to fix all of that when I make a new electronics box (Not enough room in the current one) and control panel. Here you can also see the drawbar button. The wires and hoses will need to be run through the cable track once I figure out how I want to mount it.
bHLA9HJ.jpg


Here is the little USB monitor. It is only 7", but as you can see that is plenty to see linuxcncs interface.
aELSzvO.jpg


Here it is working on the encoder mount. It is using a 5/8" 2 flute indexible carbide endmill in a homemade holder. Everything is much easier to see with the lights.
mpKht4k.jpg


Here that part is after finishing everything but the backside. It is looking pretty good for the most part. There were some issues with chatter doing the corners during a finishing pass and there were some marks left by imperfect backlash compensation.
sTa3fts.jpg


Shiny.
n9VAHdk.jpg


Backside.
l8Iww42.jpg


Using super glue work holding for this part. It needs just about every inch of the 7" of travel the machine has (actually more like 6.8" due to an error on my end). The stock is 6.5" long. The first part will be done on the mill, then I will mount it on the lathe to cut the outsides down to size.
9m0Slnt.jpg


I need to work on my Fusion skills, because every program fusion gives to run, something like 30% to 50% of the time it spends just moving around and not cutting anything. On this program, it was 45 minutes in total, and 15 of those, it was just walking around doing random whisper cuts.
kW528Ld.jpg


Lotsa parts getting made. Some of them need some work on the backside done, but it is coming along. There is some corrosion on a few of the parts because I get my material from a recycling center and sometimes they sit outside for a while before I can rescue them. The discoloration usually comes right off with some sandpaper.
SZ4cr32.jpg
 
I ran into a snag with the enclosure doors, as tends to be the case. I was planning to re-use the draw slides from my old enclosure for the doors. Unfortunately, they are 18" long, and my smaller door is 14" long. That limits its opening by about half, which isn't going to do. I also didn't plan through how to account for the metal bending after machining, and there is a bit of twist in the door frames. I think I can force it to be relatively straight if I have stronger rails on the top and bottom. Also, I cut the long pieces I needed for the doors out of a big plate on the mill, and that took forever. I should have just ordered the stock in the size I needed and saved the time (at 5x the cost though...)

So right now I am trying to come up with new ideas for mounting the doors. I could try using some cheap round linear rails on the top and bottom, but I may run into space issues, since I didn't build enough in between the Y axis motor and the doors. I think I can slide the whole machine back a little bit to make room. I could also try some aluminum angle stock and make some plastic blocks to slide on those. It won't be as smooth, but I think it would be effective. Another option might be using some skate bearings and making some guide rails using an arrangement similar to how a bandsaw guides its blade.

I have a line on a new spindle motor that I am really excited for. It isn't final yet, but I may be getting a 2 HP 6000 RPM motor for it, and I will bump the speed up to about 7500. If that works out, I am looking into adding an encoder to the current motor and getting a servo drive to use that on a 4th axis/lathe attachment. That would be much better than making 2 4th axis attachments, one with a stepper for indexing and 4th axis cutting, and one with a BLDC for lathe work. If I can't make the spindle motor a servo, I might just forgo a true 4th axis, and instead do what Tormach does with their Rapidturn and have a motor + indexing plate.

Here is how I cut off the stock for the doors.
XFwc10N.jpg


Then they were held like this to cut them to size and such. I spent probably 15 hours with the spindle running to make the 8 brackets. There must have been better ways, but I made tons of chips and the machine ran well, so it works for me. Also, you can really see the button here. It is almost like that part was made for a power draw bar button.
tWLqtE7.jpg


I didn't take a picture of the finished parts because I wanted to solve the mounting problem first. They are going to need some clean up and a coat of paint though.

I also turned the tool platter on the lathe to remove all the excess material. That part is done now, which I figured would be the most difficult of all the tool changer parts. I have started on the air cylinder for it and I am waiting on stock for the tool fingers.
CmYnFe4.jpg


I could have sworn I took a better picture once this was done, but I don't have it now (anyone else have a galaxy S8+ that has disappearing images?) This is the best I have at the moment.
7wmGy76.jpg


The machine is coming along! Working out issues as I go seems to be working, and each little upgrade makes the machine (and me) much more productive. Hopefully that continues until it is making parts faster than I can keep up.
 
Also, I figured it was time to update the total costs so far. I am adding this info since I rarely see it about others machines which makes it hard to tell what you will spend to get a certain result. Hopefully this way, if someone else is researching the way I was and trying to figure out what it will cost for a machine like this, they will see this and get a much clearer idea. If it isn't useful to anyone else, at the very least, it can help keep me organized in case something breaks and I can't remember what it was when I go to replace it. If anyone wants links to the parts I used and where I got them, let me know.

I figured I would break it down by area so it isn't one huge list. All costs include shipping.

General Machine Parts:
PM-25 mill - $1649
4 inch mill vise - $100
4 TTS holders +1 ER20 collet - $126
6 TTS knock offs (non atc collar) - $90
5 Homemade TTS holders - $30
Total: $1995

CNC Conversion Parts:
2 400 oz in Nema 23 steppers - $67
1 1000 oz in Nema 34 stepper - $69.79
2 DM542T stepper drivers - $67.90
1 DM860T stepper driver - $53.95
1 550 mm SFU1605 ballscrew - $65
1 750mm DFU1605 ballscrew - $0 (Details below)
1 350mm DFU1605 ballscrew - $0 (Details below)
6 7001 AC bearings - $15.54
3 flexible couplers - $15
350 watt power supply - $30
240 watt power supply - $20
Break out board - $10
Aluminum stock - $10
Total: $424.18

Enclosure:
2 48x96x.75 MDF Sheets - $72
5 2x4s - $15
1 box of screws - $9
3 Acrylic Sheets - $36
2 can of Epoxy Paint - $30
2x 10 watt LEDs - $11
Total: $173

Automatic Tool Changer:
Air Cylinder Stock - $50
Air Cylinder Seals - $20
Air Cylinder Connectors - $5
5 Way Pneumatic Solenoid - $20
Fasteners - $30
400 oz in Nema 23 - $33
6205 Ball Bearing - $3.11
Linear Guides and Bearings - $34
Aluminum stock - $25
Total: $220.11

Grand Total - $2812.29
 
I realized I mentioned I would document the deal with the ballscrews, then forgot and never did.

I ordered a pair of double nut ballscrews from linearmotionbearings2008 on ebay. I was quoted $100 for the 2 screws and 2 double nuts, shipped to me, machined to my specifications. I knew that price was low, but others had reported good experiences and received a good product, so I said what the hell, and took a shot. They were made and shipped, I received them about 2 weeks after I placed the order. First thing that was a problem was the packaging. It was packaged in a flimsy cardboard box, and the ballscrews were wrapped in bubble wrap. There was nothing to protect them from being bent.

The second problem was pretty obvious from the picture below.
49pUMvj.jpg


He had sent single nuts, not double nuts. Oh well, it happens. I contacted him and he apologized and sent a pair of double nuts the same day. I received those about a week later. In the mean time, I inspected the screws. They were indeed not straight, with the long one being much worse than the short one. The actual bend wasn't so bad, it was probably around .05" for the long one, and .03" for the short one. The real issue was the end machining. First, none of my stuff fit. The bearing journals were oversized (by about .01") so it took a lot of sanding to get the bearings to fit. That is where I found just how straight the end machining was. When I chucked it up in the lathe, holding one end in the 3 jaw and the other end in the steady, while both ends had runout less than .005", the middle was wiggling at .375", if not .5". It was way off.

Then the double nuts came.
The first thing I saw was the grinding job done to make the screws and keys sit flush. Looks like someone just put them together and used a bench grinder to grind it flat. In the process, taking the head off the screw making it very difficult to remove. When I tried to thread them onto the screws, neither one would thread past the first nut.
JgpdqfG.jpg


1JVgoqV.jpg


Removing the screws was rather difficult, because they were soft, and had almost no head left. I eventually did get them removed and put them on the ballscrews. Then it was pretty clear why I couldn't get it to thread on. With the keys removed, this is where the two nuts met, with no pressure added.
3FTGJbw.jpg


Taken apart all the way for further inspection found that the key ways were not milled straight and they weren't on center. The faces that met each other also were not ground, and there were some nasty burrs. Both were also full of grinder dust.
YpQXPrN.jpg


A closer look at the screws showed some of the threads were stripped out. I got some replacements which showed that 3 of the 4 holes were not tapped straight.
Jl3tmtV.jpg


All of these issues meant the screws and nuts were basically unusable without significant work.

I told the seller about my concerns, and the response was very unprofessional. I won't go into detail, but he said that was the first complaint, these screws and nuts are actually very high quality, the double nuts are precision adjusted for backlash under .0006", and he has already lost money on the transaction because he had to send the double nuts that they messed up the first time. After some back and forth, he tried to get me to send them back to him on my dime, I refused because that cost was huge, and it was their problem to fix. Then they offered to remake the screws if I paid the $46 shipping, I again refused stating that I did not want to pay extra to get what I was promised the first time. Then they offered a full refund asking me to ship them back if the cost isn't too great so they can check them and try to improve. The cheapest quote I got for the slowest shipping option was $296. I have no idea how he was shipping them here so cheap, but I did not send them back.

It took a lot of work trying to get the preload set somewhat well, but these screws are still very rough and because of the bend and poor end machining, there is some oscillation as the axis moves. As a note, the nuts and screws were significantly lower quality than the ones I got for the X2, and even after I spent a lot of time trying to get them preloaded, the backlash is still about the same as with single nuts. So I am using these screws for now while I look for replacements. I check ebay often looking for some ground options I can afford. Some have come up that are close, but nothing is just right.
 
I realized I mentioned I would document the deal with the ballscrews, then forgot and never did.

Sorry to hear about your experiences. I originally purchased Roton screws, but wasn't the most pleased with the backlash and lead error (although the screw and nut quality was quite good). Really this isn't their fault as they are commercial "transport" grade screws, not precision ground units. I had looked at your supplier for screws and was still actively considering them, however based on your experiences I may pass them over. Unfortunately there aren't many offerings for low cost precision screws for us hobbyists.
 
Sorry to hear about your experiences. I originally purchased Roton screws, but wasn't the most pleased with the backlash and lead error (although the screw and nut quality was quite good). Really this isn't their fault as they are commercial "transport" grade screws, not precision ground units. I had looked at your supplier for screws and was still actively considering them, however based on your experiences I may pass them over. Unfortunately there aren't many offerings for low cost precision screws for us hobbyists.
Given the price I was quoted, I went in expecting them to be crap, hoping they would surprise me. They did not surprise me though so I got a decent deal and some day, I might try to fix them. I have thought about pressing a sleeve onto the bearing journal and threads that are off, then trying to machine it in line with the ball races and stepping up to the next size AC bearings. At the very least, I don't think I could make it any worse.

You are right about the screws. The options are few and far between unfortunately. I found another place that sells ballscrews that look a lot nicer, but they are priced accordingly. Anahiem automation is the site, and they have 1605 and 1610 DFU screws available for $236 for 1000mm max length screws. But they are still rolled and C7 accuracy. As soon as you step up to ground, the cheapest I have found start around $485 each. For 3 axes, that costs almost as much as the entire mill, which just doesn't make sense.
 
Anahiem automation is the site

I've looked at their products and have been impressed. They seem to have quality products at less-than-industry prices, but still higher than a hobby machine deserves. Another thought would be to work with Automation Technologies to get the screws you need. I'm pretty sure they come from the same place as your screws, but as a US based company, you can probably have a little more luck with making sure your screws come straight and concentric. Worth a call at least.
 
I have had the chance to play with the mill a bit more and get started on the CNC parts. I forgot just how much fun it was turning handles and how fast simple operations can be.

I modeled up the conversion parts I will be using. I am going to need to make 15 total parts. I am currently shopping for a new X axis ballscrew with double nuts and a nema 34 for the Z axis.
View attachment 256296

The machine is very capable and leaves a very nice finish in aluminum. It has been able to hold .001" without much effort on my part. If I tried a little harder, it would certainly do better.
View attachment 256297

A couple of weird things I have experienced so far:
1. When I lock the X axis gibs, the Y axis tightens up and becomes harder to turn. The only thing I can think of that would cause that is the saddle distorting when the gibs are tightened.
2. There is a point on the X axis where it starts chattering a bit in deep cuts. It is only about half an inch that it happens so I figure it must be something with the leadscrew, or maybe a low spot in the ways.
3. There is a slight delay from the spindle motor when you start a cut. When the cut starts, it pauses for maybe half a second while the RPMs drop, then it jacks up the torque and gets the spindle back up to speed. It takes way longer for this motor to respond than the brushed motors I have used.
hey shooter, now that you have many pounds under your belt on the pm-25, do you have any regrets not going with the PM-727? I am right on the edge of ordering the 727 (my first mill) but is the 25 capable enough? I assume that since you purchased it to immediately upgrade it to CNC, ( and your parts look great!) The 25 must be a pretty strong contender for a first mill?
 
hey shooter, now that you have many pounds under your belt on the pm-25, do you have any regrets not going with the PM-727? I am right on the edge of ordering the 727 (my first mill) but is the 25 capable enough? I assume that since you purchased it to immediately upgrade it to CNC, ( and your parts look great!) The 25 must be a pretty strong contender for a first mill?

When I was looking at a new mill, the PM-727M was also one that I considered. For me, the PM-25 won for a few reasons.
1. Factory belt drive with the spindle going to 2500 RPM. The PM-727M uses gears and a single speed motor, which meant it would be a little less flexible, and speeding up the spindle would take a lot more work. The PM-727V has a 1.5 HP BLDC motor, but still had the gear head and was $600 more than the PM-25. Bumping up the PM-25 spindle only required making 2 new pulleys. The 727 would probably need to have the head taken apart, the gears removed, potentially draining the oil, then redesigning the motor mount, maybe replacing the motor (not sure if it can be run variably with a VFD), buying a VFD to get variable speeds, then making the new parts. Just a lot more work.
2. The extra X axis travel. The PM-727M and the PM-25 have the same table (I think, not 100% sure, but same size) but the PM-25 has 20.5" X travel while the PM-727M has only 15:. This means that the 727 will have a stronger table since the saddle is longer and supports an additional 5.5", but having only 1 milling machine, that extra travel was worth the loss in strength.
3. Being similar to the G0704 meant that there would be more info related to the CNC conversion out there. That meant less design work from scratch for me.

While I don't doubt that the 727 is a stronger and more capable machine, I don't regret my choice at all. The PM-25 is more flexible (and I mean that like being adaptable, not flexible like flimsy) in my opinion which is what I needed. I can say for sure that the machine is strong and capable, and very well made. So far, the limiting factor in terms of heavy cuts has been the spindle motor. It usually starts struggling before rigidity in the machine becomes a problem. So far, the heaviest cut I have taken is 1.25" wide, .05" deep, at 52 IPM. It was with my 2" 4 flute face mill, which gave me an MRR of 3.25 in^3/min. I think the motor still had some strength left, but the belt started slipping when I went higher.

I hope that helps make your decision! If you have any specific questions, I would be more than happy to try to answer them.
 
hey shooter, now that you have many pounds under your belt on the pm-25, do you have any regrets not going with the PM-727? I am right on the edge of ordering the 727 (my first mill) but is the 25 capable enough? I assume that since you purchased it to immediately upgrade it to CNC, ( and your parts look great!) The 25 must be a pretty strong contender for a first mill?

I also wrestled with the mill size question. I purchased my G0704 from Grizzly when I was just starting out in the hobby and had outgrown my Sherline "mini mill" (micro). The machine out of the box was really quite nice, and I did eventually CNC it because it was a fun project. Today, (5 years later) I am feeling like I am just running out of space and power on the machine. I would love a larger machine (RF45 or larger) and will probably upgrade. That being said, I think it all about the kind of parts you want to make, how much you want to spend, and how capable you are to design your own CNC conversion (if you choose to do so).

I will say that the extra 1.5" of Y travel on the 727 would go a long way as that is where I am the most limited on the G0704. I rarely need the full 18" of travel in X. The 727 also weighs quite a bit more would would likely lead to better rigidity (the biggest issue with all <2000lb machines). In the end, you will have a great time with either machine, just understand everything will have its limitations.
 
Back
Top