[bluegrasschips]

That's looking killer! Keep up the work and the updates. Excited to see how this performs.

[petruscad]

I started finalizing the column design so I can start making a mold for it and I'm faced with an interesting design decision. Should I support the weight of the mill's head or not?
The head's weight is made up of

Casting - ~40kg
Spindle - 35kg
Motor - 10kg
Other - ~5kg
--------------------------
Total 90kg

The options are as follows:

1. Counterweight. Add a 90kg counterweight on the back of the column connected with chains and pulleys. The advantage is that I could use a smaller diameter screw and a smaller motor. The disadvantage is having a bulky weight which makes everything ugly and complicated.

2. Air cylinders. In some commercial VMCs they add two air cylinders under the head. They are pressurized in such a way as to support most of the weight of the head. Not as bulky as a counterweight but they probably work as well. The disadvantage is needing air all the time and I'm not sure what pressure control you need for that air.

3. Air springs. There are air springs (like the ones used in the trunk of cars) that can support 45kg. Two of those can be used to support the head. The advantage is that they are really thin. The disadvantage is that the force exerted by them in not constant but depends on the compression somewhat.

4. No support! This is the simplest! The screw I was planning to use is 36mm in diameter with a double nut; it should support 90kg forever with no problems. The only issue is how much torque the stepper/servo motor will need to easily move the head up. Another advantage of this is that it may improve cutting since the weight will not allow the cutting forces to move the head up.


What do you think? What would you do?

[bhurts]

If you use option 1 cast a hollow in the column for it to travel up and down in. This will keep it looking clean. It's also what the big boys do.

Ben

[Hezz]

I started finalizing the column design so I can start making a mold for it and I'm faced with an interesting design decision. Should I support the weight of the mill's head or not?
The head's weight is made up of

Casting - ~40kg
Spindle - 35kg
Motor - 10kg
Other - ~5kg
--------------------------
Total 90kg

The options are as follows:

1. Counterweight. Add a 90kg counterweight on the back of the column connected with chains and pulleys. The advantage is that I could use a smaller diameter screw and a smaller motor. The disadvantage is having a bulky weight which makes everything ugly and complicated.

2. Air cylinders. In some commercial VMCs they add two air cylinders under the head. They are pressurized in such a way as to support most of the weight of the head. Not as bulky as a counterweight but they probably work as well. The disadvantage is needing air all the time and I'm not sure what pressure control you need for that air.

3. Air springs. There are air springs (like the ones used in the trunk of cars) that can support 45kg. Two of those can be used to support the head. The advantage is that they are really thin. The disadvantage is that the force exerted by them in not constant but depends on the compression somewhat.

4. No support! This is the simplest! The screw I was planning to use is 36mm in diameter with a double nut; it should support 90kg forever with no problems. The only issue is how much torque the stepper/servo motor will need to easily move the head up. Another advantage of this is that it may improve cutting since the weight will not allow the cutting forces to move the head up.


What do you think? What would you do?

I think what you do depends on what you want to do with the machine. If you want to have high accuracy with fast rapid speeds it would seem to me that the air cylinders would be the best but of course they will add complexity. But i don't think the complexity will be that great. Mostly I think it will just reduce X axis travel a bit. But maybe the air rams could be put on each side of the column out of the way.

The counterweight would probably be easier to do but I don't think that it would allow for really fast rapid movements. Still it could probably be pretty fast and doing rapid movements in the Z positive axis may not be as important as dong them in the other axis.

If the machine is going to be used all day long several days of the week it would probably be a good idea to do counter balancing just to save some asymetrical wear on the ballscrew nuts and keep the machine accurate over time.

I think that with a strong servo motor you could go without any support as long as you don't need fast rapids in Z positive direction. But there is also the issue of wear. IF you want the machine to wear a long time without having to replace the Z axis ballscrews it would probably be wise to have somekind of counter balance. If the machine is only going to be used a couple of hours a week it could probably go without.

If it's purely a hobby machine you could go without. But I think since you are going to all the trouble to build such a nice machine you should at least design for the future by making it easy to add in the future.

[pippin88]

If you do nothing the head will fall when power is off.

You'll need a brake (brakes that mount to steppers or servos are not cheap) or similar

[Miguel-]

How about gas springs? Like these: Alko Gasfeder Gasdruckdämpfer 495mm 200mm Hub 100N 900N Wählbar | eBay. If the travel is not long enough, you can add pulley ontop of spring (guess you'll have to support it also) and pull a wire around it to double the travel. These wont need any power, are quite durable and cheap also.

Edit: I guess you mean these in option 3. But anyways the force is almost constant all the way because these have high pressure gas inside..

[petruscad]

If you do nothing the head will fall when power is off.

You'll need a brake (brakes that mount to steppers or servos are not cheap) or similar

Yeah, I'll need a brake for sure to prevent back driving of the ball screw. And I happen to have both motor mounting breaks as well as large standalone ones. I have to incorporate that into the design as well.

[CPK_Jim]

go with the counter weight. hanging from roller chain and use a sprocket arrangement that allows you to double the lift capacity by trading for distance that counter weight moves. this helps with accell and decell times because theres less mass flying around. If that makes sense? I did this on my mill, i can take a picture if ya need a reference. The weight doesnt seem to affect rapid travel speeds at all.

Nice looking build so far!! Keep going you will get it moving in no time.

[petruscad]

go with the counter weight. hanging from roller chain and use a sprocket arrangement that allows you to double the lift capacity by trading for distance that counter weight moves. this helps with accell and decell times because theres less mass flying around. If that makes sense? I did this on my mill, i can take a picture if ya need a reference. The weight doesnt seem to affect rapid travel speeds at all.

Nice looking build so far!! Keep going you will get it moving in no time.

Yeah, a picture would be nice. And thanks for the encouragement.

[CPK_Jim]

Attachment 293300

pic is kinda dark so i highlighted the chain path in orange and circled the sprockets in green. Chain is tied to the column at one end, then goes down loops around a sprocket attached to the head and then up and over a sproket on the column then back to the rear sprocket and down to the weight. My setup is a lil crudely thrown together but has been working flawlessly. with this arrangement i was able to remove half of the counterweight appose to just an up and over setup wich is nice cuz the head on this machine is heavy.

Also no brake required!! its balanced pretty well and doesnt move much if any when you power off. But if you have the brakes i would prolly thrown one on there anyways.

[rocketflier]

Yeah, a picture would be nice. And thanks for the encouragement.

Not me/mine

Attachment 293318




Counterweight BF30 CNC - YouTube
Mill Counterweight Test - YouTube
Pneumatic counterweight on Z axis in my CNC router. - YouTube
Setting Up The Counterweight for the CNC Head - YouTube

Attachment 293316
Attachment 293322
Too big to load
http://modelengineeringinthailand.co...-Assembly1.jpg

[rocketflier]

Chain is tied to the column at one end,

Yeah, that's pretty clever!

[petruscad]

Ok, I think I'm staring to make up my mind. I will embed two plates on top of the column for the chain sprockets to be mounted to so that the chain will drop on the back side of the column. At first I will complete the mill without a counterbalance but later on if weight will be an issue I will add it. I can decide then whether to go with a dead counterweight or whether to use a pneumatic cylinder to generate the counterbalance force (I will add some way of mounting the cylinder to the back of the column). I like the pneumatic cylinder method because it allows you to easily control how much of the head's weight is balanced out and there is no bulky dead weight to move whenever the mill needs to be moved. Thanks guys for all your input.

[petruscad]

With the saddle cast its now time to start working on the precision surfaces. I began with the bottom mounting pads which go on the carriages of the Y axis. Those pads are made of steel so some scraping is needed! And for scraping one needs three things: a reference surface, a scraper and a lot of sweat! Here's how the saddle underside looked in the beginning:

Attachment 293844

The large dark slab is a grade A granite surface plate coated with Prussian Blue. Here's how the first rub against the surface plate looks like; you can see the plate touches only the outside edges of the four pads. Remember the goal here is to make all four pads lie in the same plane (as much as possible).

Attachment 293846

Attachment 293848

After some number of scraping iterations it started looking like this:

Attachment 293850

Attachment 293856

You can see how more of the pad's surfaces are touching the surface plate now. And after several days of on and off scraping sessions I got to this point...

Attachment 293858

Attachment 293860

This is much better that before, the pads are almost co-planar over their entire surfaces. More iterations will be needed but at some point I will call it done!

It's pretty painful to scrape these pads because of several reasons. First, I use a scraper of my own making which is basically a carbide insert bolted to a steel bar. Second, when I scrape, the tool sometimes slides off the steel pad and lands in the cast itself. The stone obviously tends to dull the tool and I have to sharpen it pretty often. Scraping steel is definitely harder that grinding EG and this validates my choice of mounting the rails straight on the EG surface.

[lcvette]

Looks great! Alot of work but when its done you will have a hell of a machine! Keep up the great work!

Sent from my XT1080 using Tapatalk

[Hezz]

Hey Petruscad,

It's nice to see some progress on your machine. I wonder how much an electric scraper costs. I wonder if you could just wrap a large piece of wet/dry sand paper around the surface plate and tape in tight and then just use the weight of the surface plate to lap the mounts flat. Would be faster than scraping I think. Looks like a smaller surface plate might do for lapping to.

By the way, I'm just curious how much time scraping you spent on the first round shown in these pictures. It's obvious that your contact with the surface plate has improved a great deal on this round of scraping. The one plate looks to have near a 50 percent contact area.

[rocketflier]

I don't quite understand. Scraping the base in this manner will make the 4 plates coplaner, but do nothing towards making them parallel with the upper surface?

[lcvette]

Those are the pads for the y axis trucks. They need to be coplaner and flat for a good mounting point and to not deflect when heavy loads are put on the table. Once done and the saddle is mounted on the base he will need to work the rail mounting points to ensure they are parallel to the base rails. One step at a time to get it all dialed in but you need to start somewhere.. He could do it the other way. But its about the same thing.. One side becomes the datum for the other. Depending on how far out the two planes are from parallel. He said that grinding the EG was simpler then scraping and I would agree. Less surface area to contend with the way he is doing it and once done it will be simpler for him to grind the rail mounting pads parallel.

At least that would be my interpretation of what the pictures tell me.

Chris

I imagine once he finishes the truck mounting pads

Sent from my XT1080 using Tapatalk

[petruscad]

I wonder how much an electric scraper costs.

Anywhere between &500 and $800 on eBay it seems.

I wonder if you could just wrap a large piece of wet/dry sand paper around the surface plate and tape in tight and then just use the weight of the surface plate to lap the mounts flat. Would be faster than scraping I think. Looks like a smaller surface plate might do for lapping to.

I haven't tried this but I don't think it will work very well. The surface plate is fairly heavy and the paper will tend to get worn away.

By the way, I'm just curious how much time scraping you spent on the first round shown in these pictures. It's obvious that your contact with the surface plate has improved a great deal on this round of scraping. The one plate looks to have near a 50 percent contact area.

You can achieve that initial progress with aggressive cuts it 4-5 hour of scraping; maybe 10-20 iterations.

[petruscad]

Those are the pads for the y axis trucks. They need to be coplaner and flat for a good mounting point and to not deflect when heavy loads are put on the table. Once done and the saddle is mounted on the base he will need to work the rail mounting points to ensure they are parallel to the base rails. One step at a time to get it all dialed in but you need to start somewhere.. He could do it the other way. But its about the same thing.. One side becomes the datum for the other. Depending on how far out the two planes are from parallel. He said that grinding the EG was simpler then scraping and I would agree. Less surface area to contend with the way he is doing it and once done it will be simpler for him to grind the rail mounting pads parallel.

At least that would be my interpretation of what the pictures tell me.

Yes, your understanding of the process is pretty much on the spot. If the top surface gets flattened first god knows how much steel I'd have to scrape from the bottom pads to make the two surfaces parallel. If I scrape the bottom first, I'll get away with the minimal possible effort. Grinding EG is a lot faster: if I'm aggressive I can remove 0.001" per pass; there is not way to do that while scraping cold rolled steel (cast iron - maybe).