Upgrade driver or stepper?

[cncuser1]

tbaudio: .5" lead and 2 starts... that is the ssource of the problem with lack of ooomph in Z. Any decent ball screw with .2" lead, 1 start! will give you everything you need in Z ( force and accuracy and speed).
Bonus points if you install ball screws in X and Y also.

[tjbaudio]

tbaudio: .5" lead and 2 starts... that is the ssource of the problem with lack of ooomph in Z. Any decent ball screw with .2" lead, 1 start! will give you everything you need in Z ( force and accuracy and speed).

So 300lb of friction does not sound excessive to you? According to the math I should be generating 400lb screw to head stock but only have 110 lb spindle to vice of down force.

[cncuser1]

300lb of friction does seem like alot. These mills have many sources of problems, this might be one. First off I'd ask if the ways get hot? That much friction in a peck drill operation must generate alot of heat?

But regardless of the friction, which you'll want to look into, the Z lead screw is just plain old wrong and will be the cause of continual grief.
If you do change the lead screw it will be a good opportunity to look into the fit of the head to the column, gibs etc and to verify if in fact you have the friction your numbers show. For example if you lay the column flat on a table and slide the head, do you in fact get 300lbs of friction? Is the gib jamming?

To my mind on the one hand you are not getting the performance oyu want or should be getting on Z. On the other hand you have the wrong lead screw. The shortest distance between two point is a straight line.

Good luck, curious how things turn out.

[tjbaudio]

How does this look? https://www.automationtechnologiesin...-nut-and-screw

While I am installing it I will clean and inspect the ways. I promise to report back.

[cncuser1]

tjbaudio: That ball screw has the specs I was referring to ( .2" pitch, one lead) . One part of the description says machined ends, another part says not machined, so you'll want to clear that up with the vendor. Prefer machined ends, to fit premade bearing blocks.

[tjbaudio]

Machined or not doesn't matter for me. I have a lathe that can cut it. All I have to do is drill and tap the end. The nut spins, not the screw.

[tjbaudio]

I'm getting ready to order the new ball screw. Assuming the machine will still have high friction, would going with the .2 lead and changing to a 570oz-in stepper be too much? I would keep the 2:1 step down pulleys. If my calculations are right, I could be near the maximum dynamic load of the lead screw.

[cncuser1]

If you do really have that much friction then you probably want to eliminate it, not design around it.
If you do really have that much friction then there will be too much wear related to the friction, so you want to eliminate it, not design around it.

Before you commit yourself to a design that my be with you for years, take the machine apart, remove the Z CNC motor and leadscrew. Lay the column flat on a table and constrain it. Adjust the gib and slide the head along the column. Get an idea of the friction by getting a feel of what it takes to slide it by hand. Or take the next step up and attach a 5 dollar luggage scale. Check for binding spots and gib fit.

If your head glides OK then check to see of the Z Axis motor and lead screw are racking your head( forcing it to one side or the other )

[tjbaudio]

One thing I have considered is that the Z assembly has about 55lb of spring counterbalance. I could reduce or eliminate that. IF that helps me to center up the ball screw and eliminate the friction from side loads I would have that extra down force along with the friction savings. I have an electronic dead man safety brake around if really needed.

[tjbaudio]

Just did some testing. Unhooked the counterbalance spring. Head moved fine under power. It added about 30 to 40 lb of down force at stall, about 150lb down total at stall.

Disconnected the belt from the stepper to the ball screw. With both the spring and belt off the head would mostly stay put, but was easy to push down. Pushed into the scale and released it would read about 30lb. It was hard to lift if I could even do it.

Here's the interesting thing. I could vary easily turn the ball nut by hand and lift the head-stock! There was some very minor resistance one one side of each turn. That tells me I have a minor issue in the ball screw bearing assembly. Not enough to cause my problems, more like one side is loose and the other has proper preload. Given how I made it with the tools at the time this does not surprise me. Now that I have a lathe I will re-build those parts but they are not my problem.

With the counter balance spring back on I could still move the head-stock just with my fingers on the ball nut and minimal turning force. Grabbing the ball nut and giving it a good hard twist, like you would use to turn a nut driver on a stuck bolt, I could generate 300lb of down force.

Having said that I'm pretty sure I was twisting harder than I would expect a 283 oz-in stepper to be able to do, even with a 2:1 belt on it. My calibrated fingers may be off! The spec sheet for the ball screw says .088 lb/in (lb-in?)to raise 1 lb. SO .088 *16 to get 1.408 oz-in. 283*2 = 566. 566oz-in /1.4 oz-in/lb = 400lb! I'm pretty confident in the math that says I should be getting 400lb of force. I'm not confident in the spec sheet because the picture does not match even tho the number on it does. Also the feel of pushing parts around don't ball park well with the math.

CONCLUSIONS?
The friction in the Z axis is not excessive. Perfect system, no, but not zapping 100's of lb of force. Them math or spec is wrong or there is something wrong with the stepper/driver side of things. The experience and opinions in this thread appear to match my hands on feel.

I'm going to order a new .2" lead ball screw since I need to remake a part anyway. I'm going to contemplate adding a 570oz-in stepper and driver to the order. I want to do an electronic lead screw on my lathe so even if I end-up not needing it for this it will be good for that.

Also tomorrow, check stepper wiring and driver functions.

[cncuser1]

tjbaudio: when turning the ballnut by hand you had a strange friction, take a look at the ball screw it may be bent or maybe the bearings in your mounting may have problems. When you remount the new ball screw you really want it to be aligned with the axis. When the head moves along the column you don't it to be pulling the ballscrew off axis.

[tjbaudio]

tjbaudio: when turning the ballnut by hand you had a strange friction, take a look at the ball screw it may be bent or maybe the bearings in your mounting may have problems. When you remount the new ball screw you really want it to be aligned with the axis. When the head moves along the column you don't it to be pulling the ballscrew off axis.

I agree everything in the system needs checking. It will be interesting to see how close I got.

When I made the parts I did not have a lathe and had never turned anything. I used the mill as a lathe to make the aluminum core of the rotating assembly. To hold the work piece I drilled and tapped a large hole at the center and used a big bolt as an arbor. I clamped a parting tool to the top of a 123 block and hacked away at my blank. The blank was way over sized and it took days to cut it down. Final fitting of the parts was a mess because my mounting arbor was not repeatable. I am amazed this conversion turned out as well as it did.

Today I have a lathe with a 4 jaw chuck to hold the work. It's an old Atlas 10" with a bed long enough to put the ball screw between centers for measurement. Another advantage is a 3D printer. I find them excellent for making alignment jigs for assembly. Add to that years of watching some great machinist and teachers on Youtube. This time around will be much easier.

[tjbaudio]

I got the bearing assembly out and apart. Ball screw is straight and the nut feels good. The parts I made are another story. TIR at the pulley teeth and hub was .02"

The preload screws were not very tight or even. I think the rubbing I felt was from the concentricity errors. The bearings look and feel great. I give them a shot of way oil every time I start up.

I'm embarrassed to show these but here they are. I will true up or make new parts.

[tjbaudio]

I did some more testing, the electronic side comes out fine. The DC resistance and current drive as measured with a fluke clamp on amp meter are the same for both phases and the same as the other identical parts in the same machine.

Unfortunately I did find the pulley on the Z is at the extreme limits of tolerance for concentrically. Also the Stepper shaft has .002" TIR bend in the shaft. The two problems appear to mostly balance out but I may as well fix it while I'm at it. Got to love it when stuff get's more expensive!

[cncuser1]

tjbaudio: Mount the pulley on the stepper and put an indicator where the belt contacts and also on an upper or lower face. Measure the runout at those two locations and re-evaluate if the belt can accommodate the wobble it actually sees..

[ChrisAttebery]

I just skimmed through this thread so I may have missed something.

Anyway I have a G0704 with 5 turn per in ball screws and 570in/oz steppers running on 45V. Even with a 2hp 56 frame motor I can move the Z at 180-240IPM.

IMHO You should have decent torque with the 5 turn screws and a 282oz stepper. It may not be lightning fast but it should be usable.

[tjbaudio]

I just skimmed through this thread so I may have missed something.

Anyway I have a G0704 with 5 turn per in ball screws and 570in/oz steppers running on 45V. Even with a 2hp 56 frame motor I can move the Z at 180-240IPM.

IMHO You should have decent torque with the 5 turn screws and a 282oz stepper. It may not be lightning fast but it should be usable.

Current ballscrew is 0.5 in lead or 2 turn per in. The belt drive is 2:1 so a total of 4 turns of the stepper per in of travel.

I ordered a 570oz stepper and a leadscrew with a.2in lead.

[ChrisAttebery]

You should have 10x more torque available then.

[tjbaudio]

Got the new parts in today, I think the new ball screw will make a world of difference. If the original motor had a strait shaft I would have been just fine with it. It will take me a while to get it installed. Have to make a whole new set of brackets and such to make it fit.

I am very seriously considering going without counterbalance if it all works out. Will see how testing goes.

[tkalxx]

Having said that I'm pretty sure I was twisting harder than I would expect a 283 oz-in stepper to be able to do, even with a 2:1 belt on it. My calibrated fingers may be off! The spec sheet for the ball screw says .088 lb/in (lb-in?)to raise 1 lb. SO .088 *16 to get 1.408 oz-in. 283*2 = 566. 566oz-in /1.4 oz-in/lb = 400lb! I'm pretty confident in the math that says I should be getting 400lb of force. I'm not confident in the spec sheet because the picture does not match even tho the number on it does. Also the feel of pushing parts around don't ball park well with the math.

.

Your answer of 400lbs lifting force is correct, but it's also not... You're calculating this force from the motor's maximum rated holding torque. I'm not sure where you got the ball screw "spec sheet numbers" or what they even mean but there is a mathematical way to determine the lifting force of a screw. The most commonly used equation for describing the relationship between torque and force of a screw is:

T = F[Tp + Tt + Tb]; where:

T = total input torque applied to the screw
F = tensile force generated in the screw as a result of the input torque
Tp = component of torque that creates the force in the bolt through the "wedging" action of the threads
Tt = component of torque that must overcome friction between internal and external threads
Tb = component of input torque that must overcome friction between the screw head and bearing surface

Since we're considering ball screws in this application, for simplicity we'll just eliminate the friction between the threads (since they're ball bearings) and the friction between the screw head and bearing surface (since you have bearing mounts).

So we're left with T = F*Tp; where Tp = P/2*pi and P = thread pitch

You have 283oz-in motors with a 2:1 belt reduction so the very maximum torque you can produce is 566oz-in or 35.4lb-in. You have to remember, this 35.4lb-in is the maximum holding torque... this is not the torque your steppers are putting out when moving your machine at 100ipm. I pulled the torque curve for the kelling 270oz-in 2.8A stepper and found that at 100ipm with your 2:1 belt reduction, 30V power supply and a 0.5" thread pitch ball screw, your maximum input torque to the screw is just over 400oz-in or 25lb-in. Using the equation above we can find the magnitude of force put out by the screw at that input torque.

F = (2*25lb-in*pi)/(0.5in) = 314lb of force. ---> This is the absolute maximum force your screw and motor combination can put out at 100ipm neglecting every single source of friction in the system - in reality it will be lower because friction exists everywhere (especially in the z-axis of every machine).

Changing to a 0.2" thread pitch ball screw will make a massive difference (however not 10 times more like ChrisAttebery suggested):

F = (2*25lb-in*pi)/0.2 = 785lb of force