New PM-25MV Mill

I think it is time for the direction of this machine to change. I have been in college for the last 4 years and for the most part, I have had an abundance of time and a very limited budget. Well I am graduating in 2 weeks and I have started a new job. As the stars have aligned, those two have reversed. I now have an abundance (well compared to what I am used to... which was no money) of money to spend and limited time. So now, I need the machine to work faster and more reliably, even if that means some things will be more expensive.

So its time to re do the machine I think. I would like to tailor it more towards production and reliability so I can design a part and output some toolpaths, then stick a piece of stock in the vise and press start, then come back however many hours later and have a finished part waiting for me. Ideally, I could get the machine up and running in 10 minutes or less, and not have to worry about a thing, even if that means I don't come back to the machine until the next day (or next several days). So what will I need to accomplish this?

1. Servos. I just don't trust steppers to run that reliably for long amounts of time without me checking in on them. I also would like the increased speed from servo motors and the fact that servos are much smoother is a big plus. Right now I am looking at DMM 640-DST-A6HS1 motors and DYN4-L01A2-00 drives for X and Y, then 86M-DHT-A6MK1 and DYN4-H01A2-00 for the Z. All in, those would run $1415 and would give me (theoretically) 1000 IPM rapids (more likely 400 IPM rapids) and cutting feeds in the 200 IPM range. Alternatively, I am looking at clearpath servos, which I haven't been able to decide on a model, but it looks like it will be roughly the same price, but I would get to try them out for 3 months to see if I like them first. I could also put servos on X and Y just for the running smoothness and rapids, then if the machine makes any money, add a servo the Z.

2. Flood coolant and new enclosure. My enclosure was meant to be quick and easy to contain some of the mess. It did that for sure, but I want something that will be water tight (ish) and allow for flood coolant. I sort of like mist coolant, but it doesn't do a great job with pockets and holes, and it is a bit unreliable for me. It is also tough to get it to work with different tool lengths without adjustment. I am working on a design for an enclosure made of steel instead of wood (I got a welder a few months ago and messed with the exhaust on my car to learn to use it, now the car is louder and I sort of know how to use a welder) and sheet metal instead of MDF. It will also be stronger and more rigid so it doesn't shake like crazy when the machine accelerates hard with the servos.

3. Auto tool changer. There is only so much the machine can do without me changing tools constantly. I made an auto tool changer and it had a lot of problems. So much so that I decided not to show it at all and keep it to myself. I have tweaked the design and changed a ton on it, and once the stock and parts come in, I will try tool changer take 2. It will need sensors out the hoorah to make sure that if it fails, the machine doesn't continue running and break everything. I have nightmares thinking about a partial index causing the machine drop down and break the tool changer, then have it continue running and break something each time it smashes into the tool changer. Don't want that nonsense.

4. Ground ballscrews. My ballscrews are rough and I don't love them. If I can find ground ballscrews for a semi reasonable price, I will buy them in a heartbeat. This doesn't need much explanation.

5. Remote control. My house is wired with security cameras, and I can check on them with an app on my phone. I also learned how to use remote desktop at my old job. I feel like these can be combined so I can keep an eye on the machine when I am away, and maybe even have it do stuff remotely. This is just a nugget of an idea, but who knows...

6. Probe. I would like to make a probe so I can fixture something and tell the machine to do its thing. I think the most time consuming part at the moment besides work holding is using a wiggler to touch off the workpiece. A probe would speed this way up. If I can't make a probe that is reasonably accurate, I will look into buying one.

7. A new control computer. The one I have is an old work surplus machine that isn't all that great. It works but it has its glitches (for example, 1 in 10 times when I remove the USB drive, the computer restarts) and it isn't all that reliable. I want one that will work all the time and minimize the time I spend redoing set up work because it conked out for some reason.

8. Spindle work. The spindle on the machine is like a 4 out of 10. It needs to be balanced and have a few other improvements made so it can run fast without vibrating. I want a better surface finish that it just can't handle.

9. Head spacer. The machines spindle isn't centered over the Y axis travel, so in order to use the whole thing, parts need to hang off the front of the table. This often makes them run into the column. You also lose a lot of travel when you use a vise. I would like to fix that.

I would love some feedback on all this. If you have ideas for things I could get to improve it, I am all ears. My idea is to make it out perform a Tormach 440 (which I consider the closest competition, though my machine has more travels) by far, for less than half the price of an equivalent machine ($14,500 for the machine with tool changer and enclosure).

Like everything for this machine, I hope it will only take a few months, but I am sure I will be halfway done by next Christmas...
 
I'm interested to see what you come up with... Servos and a encloser that can handle flood coolant are the things I dream about at night. If you and Jake both figure out servos, it should be a walk in the park for me!

We should compare spindles. Do you have a model of yours that I could compare to mine? I've been doing a lot of dreaming of a spindle that has stacked AC bearings and can turn 10k rpm... but this means a whole new spindle, and probably quill. I've been thinking about how to do this in a smaller lathe and hold everything as concentric as we would need it... if our spindles are the same, I'd be willing to split some of the development effort with you.

PZ
 
I'm interested to see what you come up with... Servos and a encloser that can handle flood coolant are the things I dream about at night. If you and Jake both figure out servos, it should be a walk in the park for me!

We should compare spindles. Do you have a model of yours that I could compare to mine? I've been doing a lot of dreaming of a spindle that has stacked AC bearings and can turn 10k rpm... but this means a whole new spindle, and probably quill. I've been thinking about how to do this in a smaller lathe and hold everything as concentric as we would need it... if our spindles are the same, I'd be willing to split some of the development effort with you.

PZ
The DMM and clearpath servos are supposed to be very easy to set up and use. I anticipate clearpath being much easier then steppers from what I have seen. The reason I have never used them is just because of cost.

I have a model that I can send to you tonight. I don't think the two are very similar though, yours is a whole lot bigger. Mine is really quite tiny and unimpressive. Ive considered making an entire new spindle, but I don't know that it would be worth it for this particular machine. I would need to replace the head as well, and I would probably want to change it to a BT30 instead of R8. I think I can get this spindle to run acceptably at 10,000. I will need to remachine the threads at the top and replace the r8 collet it is currently using (threads arent super straight on those either), then see if I can use a phones vibration sensor to get it balanced better. If that all works, it should run much smoother and vibrate less.
 
I have been visiting my parents for Christmas, so I have been away from the shop for a few weeks. That means plenty of time to model and design stuff. I made a number of changes to the design of the tool changer because there were a few things that weren't quite working for me.

First was the tool fingers. They were a little bit over complicated, had very small parts that needed to be made, couldn't get enough spring tension in the space that I had left, and they required a very accurate tool holder geometry. The diameter of the groove had to be very specific and even when they were close, it would either be super tight, or somewhat loose. I didn't like that, so I changed the design to a 2 piece finger design that rotates on a shaft and are forced apart at one end by a spring so they grab the tool holder on the other end. I saw a very similar design on Haas tool changers so I figure it has to be a decent design. I started making some of those parts before I left and I think it should work just fine.

Here are is the updated version of the tool changer, from the bottom. You can also see some cut outs in the tool platter to try to reduce the weight.
ddN5CMn.png


I also decided to just buy an air cylinder instead of making one. I was having a heck of a time boring out a bar to use (accurately) and an already finished one made of stainless steel was only $24 brand new. Screw it, I bought it instead. It is a 2 stage cylinder with a 4" stroke. That will be plenty for the tool changer. It has a little over a 1" bore which will let me push it with 150lbs of force if I need to. I don't anticipate needing to use more than 30ish lbs.

I ran into some trouble with the motor that is used to rotate the tool changer. I was concerned that it wouldn't have enough torque to rotate the tools. The tools rotate on an axis with a diameter of 4.5". Considering only the weight of the tools, estimating 1 lb per tool (conservatively, more likely each tool will weigh less), that gives me 10 lbs (160 oz) rotating at 4.5". To move them, I would need 720 oz/in of torque (if my math is right, it is very possible that I am miscalculating completely). I am using a 400 oz in stepper, so it might struggle. To mitigate this, I made a 2:1 gear box that will double the torque and hopefully let me rotate the tool changer without trouble.

SseNxKE.png


I am planning to make some changes to the pneumatic draw bar as well. Right now I am using a 3 stage air cylinder with 3" pistons. The springs I am using have a working load of 900 lbs and a flat load of 1200 lbs. I am running them in pairs giving me 1800 lbs of working force and 2400 lbs to release. At 115 PSI, that gives me enough force to release the tool. The plan now is to add another spring to give me 2700 lbs of holding force and 3600 lbs to release. This will let me push the tool a bit harder without worrying about it pulling out. To get the 3600 lbs of release force, I will add another stage to the air cylinder and push the pressure up to 130 psi. That will give me 3675 lbs of force to release the tool.

Last, I got a deal on some vises on ebay. I got 2 4" machine vises for $85 each that seem to be good enough for what I want to use them for. I now have 4 vises, and they should all fit on the table of the machine, letting me run 4 parts at the same time, or maybe run OP1 on 2 vises, then OP2 on the second 2 vises. I think this and the tool changer should let me really ramp up the production capabilities of the machine.

Here is a look at the 2 vises I received. They have the flat sides which should let me get them in much closer to each other and save some space. For what I paid, I am very pleased.

yFr4hnI.jpg


Here is how it should look when I get all 4 mounted with the tool changer.

ZMsyt89.png
 
Awesome job on the ATC! I can't wait to see the real deal. I have mine 90% built and laying in a corner. You're making me want to go finish it up.
 
Also, from having designed and built an ATC from the TTS system, be aware that the slot on the tool holder is very finicky and intolerant of misalignment. It has caused me a lot of problems in the past. I'm happy to share any design files for my ATC which might help you.
 
welcome to the land of tax payers!

I saw that your posting on the coolant lines that occasionally stopped flowing, that you had to turn the needle valve open and when it started flowing again you turned it back down. If you want this to run automatically that cant happen.

Couple of thoughts...
1. on the inlet side before the needle valve put an inline filter. You can get these at Tractor Supply or Rural King (or Ebay) for about $5-$8. Look for the filters that are used for home spraying systems (I have one I pull behind my garden tractor when I want to spray weed killer in the yard). I would start with the smallest screen mesh they have since the viscosity of what you are dealing with is just above water. This will eliminate any very small particles that get into the coolant as it wont take much across that needle valve to plug it enough that requires manual intervention. Another filter that would work would be a paint filter, like is used for a manual spray gun. This might be a better way to go since they may make them with smaller mesh size. At any rate you can try one of them.
2. After putting the filter in place, if you still have problems, get a more accurate regulator. if you are using 5psi, make sure that you are using a regulator that is 0-15psi....you want that regulator to not be sitting too close to off...that will raise hell with you too.
3. Not sure what you are using for a needle valve but the same issue here. If you cant control the pressure feeding it the needle valve will have to take the pressure drop meaning it will be hardly open, again, it wont take much to choke off the flow requiring you to open it and then after what ever it was cleared, then setting it back again....you might need a better needle valve....

Having built systems before in my work life, these were the kinds of things that often tripped me up. These systems were going into automated systems where there were no human eyes so they had to work. That is what you want.

Let me know your thoughts....
 
That really looks great.

As far as torque is concerned for rotating the ATC when there is no external force being applied to the assembly (is spins free save friction): Torque=Angular Acceleration * Moment of Inertia (τ=αI). Moment of Inertia is a term takes into account size, shape, mass ect. of the object that is rotating. τ=αI is actually just the rotational equivalent of F=ma for linear motion. Except F is replaced with τ, m is replaced with I, and a is replaced with α.

I could go into more detail, if you like, but long story short, you won't need the gear box to spin the ATC. I think there is a way to get Fusion to tell you your moment of inertia, if not I think I can get it in SolidWorks. Or we can roughly calculate it and get close enough. OR I'll beat my inner nerd back to where he belongs.

As always, inspiring work. Your video "Pushing PM-25 CNC Faster" actually got me off my ass and working on my machine!

PZ
 
welcome to the land of tax payers!

I saw that your posting on the coolant lines that occasionally stopped flowing, that you had to turn the needle valve open and when it started flowing again you turned it back down. If you want this to run automatically that cant happen.

Couple of thoughts...
1. on the inlet side before the needle valve put an inline filter. You can get these at Tractor Supply or Rural King (or Ebay) for about $5-$8. Look for the filters that are used for home spraying systems (I have one I pull behind my garden tractor when I want to spray weed killer in the yard). I would start with the smallest screen mesh they have since the viscosity of what you are dealing with is just above water. This will eliminate any very small particles that get into the coolant as it wont take much across that needle valve to plug it enough that requires manual intervention. Another filter that would work would be a paint filter, like is used for a manual spray gun. This might be a better way to go since they may make them with smaller mesh size. At any rate you can try one of them.
2. After putting the filter in place, if you still have problems, get a more accurate regulator. if you are using 5psi, make sure that you are using a regulator that is 0-15psi....you want that regulator to not be sitting too close to off...that will raise hell with you too.
3. Not sure what you are using for a needle valve but the same issue here. If you cant control the pressure feeding it the needle valve will have to take the pressure drop meaning it will be hardly open, again, it wont take much to choke off the flow requiring you to open it and then after what ever it was cleared, then setting it back again....you might need a better needle valve....

Having built systems before in my work life, these were the kinds of things that often tripped me up. These systems were going into automated systems where there were no human eyes so they had to work. That is what you want.

Let me know your thoughts....
I hope uncle sam enjoys living large on my hard work :)

I will keep your thoughts in mind. I think a filter is probably a good idea, but the issue has largely gone away on its own. I switching to koolmist from WD-40 helped a bunch, and I think the WD-40 may have been clogging the lines a little bit for a while after. It is definetely a cheap thrown together system. The needle valve is one I found in the garage, probably form an old refrigerator and the mist system parts are almost completely from an RO water filter. The regulator is an inexpensive one and I am not sure what the rating on it is. If I redo the mister, I am going to follow your advice and add a filter and a better regulator.
 
If you stated earlier, I am sorry if I missed it. What are you building this machine to make? Are you trying to get in the machine shop business? Make your own products you plan to sell?
 
Back
Top