A2Z CNC monster mill X and Y setup ordered with Kerk Screws

[Don-in-Japan]

OK, expanding on that logic a bit...A 2 pitch screw would be better than a 4 pitch screw which would be better than a 20 pitch screw. So why do manufacturers waste all that time making higher pitch screws and reduction units?

Remember that power is the product of torque and speed. If you reduce the speed with a gear reduction by 20:1 you get 20 times less speed but twenty times more torque. (But the same power.)

You also have to account not just for moving the table but for developing enough torque to push the metal past your bit.

I was only talking about the Sherline unit. Is it common to use high pitch screws and reduction units on them? If so, I went the complete opposite by purchasing the A2Z recommended leadscrews.

[Jeff-Birt]

Sherline (and Taig) used 20 TPI screws to get all the drive reduction needed in one step. The Kerk screws are really good screws. The combination of size and pitch used makes for a much more efficient screw but you are still loosing all that mechanical advantage (reduction) from the higher pitch stock screw.

It is very common to see commercial machines use a belt reduction on both stepper driven and servo driven machines. In our DIY CNC community some folks are opposed to the perceived added complication (and cost) of adding a belt reduction to each axis and try to compensate by adding huge motors.

Taking the Sherline as an example the stock 20 TPI screw will give you 20 times the torque of the stepper: 20 * 135 = 2700 oz-in. With the same motors and a 4 TPI screw you have: 540 oz-in. (I using holding torque numbers here for simplicity). So through the entire operating range you have less torque available to push the stock past the bit. That is why you said you had to take light cuts.

[escott76]

Sherline (and Taig) used 20 TPI screws to get all the drive reduction needed in one step.

I would guess that this has more to do with the fact that you get a nice even division into .050 per handwheel turn on the manual version with an off the shelf common screw than having anything whatsoever to do with getting reduction. While it does give a reduction (and not IMHO a needed one for this application, but our opinions differ on this subject) it was not a choice made specifically over a different thread on the manual version.

[Jeff-Birt]

I would guess that this has more to do with the fact that you get a nice even division into .050 per handwheel

That could very well be, but they could just have easy used 10 TPI screws and made you turn the handle half as much. But then the handles are teeny-tiny and that would have been really tough to turn, it would have been much harder to be accurate with the graduations half the distance apart too. (Smaller diameter handle = less applied torque.)

Take a look at this article: http://www.arceurotrade.co.uk/projec...cle_Part_1.pdf , pay particular attention to how he chose the pitch of the leadscrews specifically with direct drive in mind, and he chose direct drive to make the conversion easier.

The point is that is takes a certain amount of torque to push the material past the bit, and move the table, etc. You simply can't use any old pitch screw and get the same results.

[hoss2006]

You should read thru the entire pdf more carefully.
As you said he chose direct drive to make the conversion easier which is why it's the most popular way to do it.
After all his calculations he chose larger motors than necessary,

I like
a very good margin of safety in matters such as this.

His math didn't account for everything,

As my work invariably uses small cutters, I chose to ignore cutting forces.

Have a look at the last pic on page 3, he uses a very small vise that isn't anywhere near what most people use on an X3.

the torque estimated above does not take into account the forces generated during machining or the additional weight of a vice.

Pity he did not give any specs on how it performed after the conversion.
That's what matters, results.
There have been more than enough conversions done on this site of every type of mill
that people can see what works well and what does not.
I didn't see anyone here suggesting putting a N34 640 on a sherline or taig.
A 282 is not a big motor.
Choices have to logical.
The point of switching to a lower pitch screw is to increase rapids.
If someone is intent on keeping lower torque motors, then they'll have to either increase the voltage or use a drive reduction to keep from losing steps.
Pulleys will cost them rapid speed which was the reason for the switch in the first place.
Cheaper than upgrading drivers and power supply would be to increase the motor size so long as it's a good match to the existing drivers and PS.
I run a 270 direct drive on a 5tpi screw at 40V, it's more than enough torque to exceed
the endmills chip load recs.
A drive reduction would do nothing but cost rapid speed.
Direct drive is popular for very good reasons.
Hoss

[escott76]

That could very well be, but they could just have easy used 10 TPI screws and made you turn the handle half as much.

No, that is my point exactly, they couldn't just as easily have used 10 TPI screws, as that is not a commonly available off the shelf part in those thread sizes. Making these mills is a compromise between many things, but ultimately cost to the consumer has to be balanced. Trying to assert that they "chose" these screw pitches to make the "correct" reduction for CNC is reaching.
I run 5TPI screws direct drive in my homebuilt. I did a bunch of math and got the performance I wanted. I've got enough power to snap endmills. Why is it I need more reduction again?

[Jeff-Birt]

All right I give up. Then all the folks that make NEMA gear reduction units are idiots and we really don't need those pesky transmissions on our cars! Believe what you want, physics is physics.

Trying to assert that they "chose" these screw pitches to make the "correct" reduction for CNC is reaching.

That is NOT what I said. I said that pitch gives you the reduction needed. I suspect you could get by with a bit less. The point I have been trying to make and will soon shut up about is that you simply cannot slap a stepper motor on any old screw and think it will work well.

Seriously. When every you go through all the trouble to actually engineer something and so all those useless calculations you wind up with a good estimate of what 'should' be required. Then you have to factor in the unknown things like how tight will the end user actually tighten the jibs, how 'sticky' will the ways get between cleaning, etc. Then you add some safety factor. Have you ever driven across a bridge? Aren't you glad some engineer actually designed it and bothered with all those calculations, found out what 'should' be required, and then made it 2-3 times as strong? Beats the hell out of pulling a number out of your butt and building it the 'easy' way.

The gentleman in the article gave a good explanation of the calculations needed if you actually try and engineer something. He admittedly ignored certain things and even gave ideas on how other things could be measured in order to make the calculations better. You can find more detailed explanations of the same process in many places. This is one I found for a customer of mine who was looking at doing an X3 conversion. Do what you want, the is the great thing we can all do it our own way.

[g29cc]

So being new to all this , and I love info. Going with a reduction setup reduces the rapid speeds , over a direct drive setup ?

[Don-in-Japan]

Don't give up Jeff.. this info is what keeps me going. I understand what you are saying about the reduction drives, and leadscrew pitch.. what I want to understand is about the A2Z/Sherline model in particular.

I upgraded to the 4TPI leadscrews thinking mainly on having zero backlash, and thought that since the motors will be turning at a lower speed, the torque would be greater.. apparently, I am losing torque by using a coarser thread. At what point would a reduction drive with smaller steppers be better than a direct drive system with larger steppers (say.. 400 oz.) on an A2Z upgraded Sherline?

I want to upgrade this thing correctly, and don't want to waste any more money doing it. I cut mainly aluminum, and have had issues from time to time..software and hardware.

What would the ideal leadscrew pitch/motor size/reduction (or not) be for an average use A2Z mill?

[hoss2006]

All right I give up. Then all the folks that make NEMA gear reduction units are idiots and we really don't need those pesky transmissions on our cars! Believe what you want, physics is physics.


That is NOT what I said. I said that pitch gives you the reduction needed. I suspect you could get by with a bit less. The point I have been trying to make and will soon shut up about is that you simply cannot slap a stepper motor on any old screw and think it will work well.

Seriously. When every you go through all the trouble to actually engineer something and so all those useless calculations you wind up with a good estimate of what 'should' be required. Then you have to factor in the unknown things like how tight will the end user actually tighten the jibs, how 'sticky' will the ways get between cleaning, etc. Then you add some safety factor. Have you ever driven across a bridge? Aren't you glad some engineer actually designed it and bothered with all those calculations, found out what 'should' be required, and then made it 2-3 times as strong? Beats the hell out of pulling a number out of your butt and building it the 'easy' way.

The gentleman in the article gave a good explanation of the calculations needed if you actually try and engineer something. He admittedly ignored certain things and even gave ideas on how other things could be measured in order to make the calculations better. You can find more detailed explanations of the same process in many places. This is one I found for a customer of mine who was looking at doing an X3 conversion. Do what you want, the is the great thing we can all do it our own way.

No one suggested slapping any old stepper on any old screw so you're reaching there too.

The only thing pulled out of a butt is your attitude.

Engineers can get it wrong too so you know, the I-35W bridge in Minnesota comes to mind.

Your just stuck in the past.
Look back at the beginnings of this forum around the same time as the article, 04/05.
You see plenty of slow mills built with weak steppers and belt reductions.
Here in 2009 we are lucky that stepper/driver availability and affordability is much better than then,
so we can build machines that have power AND speed and not have to break the bank.
It's called progress.
That's why Syil, Novakon, Tormach, etc sell mills with direct drive, low pitch screws and steppers that match the driver and power supply which is what
many of us do and pass on to the new guys.
People want power AND speed, that's just the way it is.
I'm sure there was an engineer or two behind all these advances ( Mariss and the G540) but you don't have to be one
to build a mill.
You can browse forums and see what others have already done and replicate it if you like their results.
Move forward.
Hoss

Not trying to be mean or anything here, just giving a different point of view.

[escott76]

All right I give up. Then all the folks that make NEMA gear reduction units are idiots and we really don't need those pesky transmissions on our cars! Believe what you want, physics is physics.


That is NOT what I said. I said that pitch gives you the reduction needed. I suspect you could get by with a bit less. The point I have been trying to make and will soon shut up about is that you simply cannot slap a stepper motor on any old screw and think it will work well.

Seriously. When every you go through all the trouble to actually engineer something and so all those useless calculations you wind up with a good estimate of what 'should' be required. Then you have to factor in the unknown things like how tight will the end user actually tighten the jibs, how 'sticky' will the ways get between cleaning, etc. Then you add some safety factor. Have you ever driven across a bridge? Aren't you glad some engineer actually designed it and bothered with all those calculations, found out what 'should' be required, and then made it 2-3 times as strong? Beats the hell out of pulling a number out of your butt and building it the 'easy' way.

The gentleman in the article gave a good explanation of the calculations needed if you actually try and engineer something. He admittedly ignored certain things and even gave ideas on how other things could be measured in order to make the calculations better. You can find more detailed explanations of the same process in many places. This is one I found for a customer of mine who was looking at doing an X3 conversion. Do what you want, the is the great thing we can all do it our own way.

Transmission in a car is a far cry from what we are talking about, and there are MANY different NEMA frame 23 motors for MANY different applications, so I'm sure there is cause for one in some applications.
It's nice that you've chosen to patronize me by explaining how engineers calculate things, but I'm pretty well versed in safety factors (you know like the ones I use when I design press tooling at work so it doesn't fracture under load/overload). I am also aware of how to do the relevant calculations for a mill, did so, and chose direct drive steppers and a 5 TPI screw. According to you this must work like crap because I didn't add some sort of reduction. It doesn't.
Yes, you need to consider specifications and choose motor/screw/driver/ps appropriately, but a reduction system with steppers is more complex than needed much of the time with commonly available equipment today.

[g29cc]

So given the setup mentioned, what steppers would you guys recommend ?

[scudzuki]

You also have to take into account the efficiency of the screw thread itself, as there will be less frictional loss with ground threads vs machined and even less with ball screws. An accurately ground 5 TPI screw could approximate the mechanical advantage (under load) of a much higher TPI screw (8 TPI? 9?) with rougher chased threads.

Joe

[GeneK]

I find that the 10k spindle works very well with the smaller end mills, 3/32" to 1/4". I run about 8 inch per minute feed with a .003-.006" depth of cut in steel. the chips are so small and don't pile up to break the bits.

GeneK

That is exactly what I am doing. I have a manual Sherline 4400, and so ordered the X, Y and Z bits from Tim. I plan to just switch the headstock back and forth until I get bored or need 10K on the spindle.

Thayer

[GeneK]

Sorry I just noticed this question today. The stock steppers perform better with the 4 tpi screws. As Tim said there is more torque available at the lower step rate hence I now get rapids of 40 ipm instead of the 20 ipm I had before. Actual feed during cutting seems to be limited by the material and the cutter bit.

GeneK

GeneK,
Did you use the stock Sherline steppers for the conversion ? If so , how well do they work ? I was told with the replacement screws being 4 tpi vs stock 20 tpi that the I might have torque issues and have to go to a better stepper setup to achieve max IPM settings. Which A2Z claims is around 90-100 ipm rapids.
Anyways I might just bite the bullet , and get the z column, some reduction steppers assemblies, a new controller and spindle control , maybe even a new spindle motor ( something with a little more power ). Not sure yet. I am thinking of doing this so I have a extra machine at my disposal. But I am not sure if its worth the cost ( around $3400 is what I ll have in the entire machine when I am done ).

If your happy with your steppers ( if they are stock ) then I might just run my machine the way it is and see how it goes. I am using it more for a small business light light light production of parts.

Thanks

[g29cc]

Sorry I just noticed this question today. The stock steppers perform better with the 4 tpi screws. As Tim said there is more torque available at the lower step rate hence I now get rapids of 40 ipm instead of the 20 ipm I had before. Actual feed during cutting seems to be limited by the material and the cutter bit.

GeneK

A few people I have talked to said they are getting close to 100 ipm with bigger steppers. With you saying that you get 40 ipm rapids, does your setup miss steps if you go faster then that ? I am just curious, I ll be happy with double the speed, but if I can get closer to 100 ipm , then I would be extremely happy.

Thanks !

[GeneK]

I really don't remember why I didn't go for higher rapid speed. In most of the cutting I do there is much more time at the feed rate and minimal time at the rapid rate. The real limit I run into is the acceleration on the Z axis. There is just so much mass in the head stock and motor that it is hard to not have missing steps.

GeneK

A few people I have talked to said they are getting close to 100 ipm with bigger steppers. With you saying that you get 40 ipm rapids, does your setup miss steps if you go faster then that ? I am just curious, I ll be happy with double the speed, but if I can get closer to 100 ipm , then I would be extremely happy.

Thanks !

[g29cc]

I am going to do a little research and if I can get the machine setup to run close to 90-100 ipm on rapids, I would be tickled pink !

[g29cc]

Well I bought a Z column today and looking at steppers Keling

8C: KL23H284-35-4B (1/4” Dual shaft with a flat);387 oz-in $49 each. for all three axis.

Question , should I replace the motor on my rotary table too ? Since I am not really changing anything with that and its not really used that often I am thinking I could just leave it alone.

But hopefully I'll have my little monster mill going soon and then I think end of the year I ll pick up the Novakon NM 200 for my real production machine.

[g29cc]

Still accumulating parts. But I hope to have this running soon !