Haas Mini Mill spindle/tool holder/tool balance issue?

[MiniMillMachinist]

I don't know if this has been covered yet, but do you get the same results in all directions? Like if you cut around a square block, do all faces look the same? Then rotate the square 45° (diamond shape) and cut again? If one or two sides look significantly different, this would indicate a machine rigidity issue, like slop in ways, ways or trucks not tight, etc.

It looks basically the same, the problem might be slightly worse in X, but the same cutting characteristics are there when you zoom in with the microscope - the cutting lines cross over one another and you can see the fish scales. I assume the machine is more rigid when cutting in the Y axis. I have tried pushing and pulling on the X and Y as hard as I can and letting go and I can't get more than about a micron or two of slop to show up between pushing and pulling, with the dial indicator referenced against the spindle head. When I push on the spindle head and reference the table, it's about the same story, but perhaps slightly less free play. Hard to say because the numbers are so small!

 

[MiniMillMachinist]

I don't know if this has been covered yet, but do you get the same results in all directions? Like if you cut around a square block, do all faces look the same? Then rotate the square 45° (diamond shape) and cut again? If one or two sides look significantly different, this would indicate a machine rigidity issue, like slop in ways, ways or trucks not tight, etc.

Thinking about this a bit more, is it possible for a spindle to have more slop in one direction than the other? The problem does seem to be amplified slightly in Y, but I don't think it can be a rigidity issue as my other machine has a lot more flexibility to its frame (it's a router) and doesn't have the same characteristic to the cut.

I know the vibration analysis showed a decent result, but I wonder if the vibration that is there is amplified in Y because the machine is less rigid this way... if that makes sense?

 

[Houdini16]

Problem is, its a Haas, like mine, good luck getting them to fix it. Let us know if they do, they don't like throwing money at it, but seems that's where its at.

 

[MiniMillMachinist]

Problem is, its a Haas, like mine, good luck getting them to fix it. Let us know if they do, they don't like throwing money at it, but seems that's where its at.

I think they're going to look at replacing the belt first, so we will see... ultimately I expect them to resolve this one way or another, at the end of the day I've paid good money for this. I'm otherwise quite impressed with the machine, it seems really well built.

 

[mhajicek]

but I don't think it can be a rigidity issue as my other machine has a lot more flexibility to its frame (it's a router) and doesn't have the same characteristic to the cut.

What I'm thinking is maybe during assembly some bolts didn't get tightened all the way. So you could have a rigid frame, but some bolts between the frame and the ways, or the trucks and the table, or somewhere else might not be fully tight, allowing things to move in ways they shouldn't. This wouldn't show up as vibration while running under no load, but as play in one or more directions under load.

 

[Houdini16]

I think they're going to look at replacing the belt first, so we will see... ultimately I expect them to resolve this one way or another, at the end of the day I've paid good money for this. I'm otherwise quite impressed with the machine, it seems really well built.

I have 2 lemon machines, and one of them they just stop communicating with me pretty much blew my off. The other they had excuses why they weren't going to fix it/swap things out. So don't underestimate them not willing to fix it, and just blaming it on other things. Its a good time trust me.
Watch this video!

 

[MiniMillMachinist]

What I'm thinking is maybe during assembly some bolts didn't get tightened all the way. So you could have a rigid frame, but some bolts between the frame and the ways, or the trucks and the table, or somewhere else might not be fully tight, allowing things to move in ways they shouldn't. This wouldn't show up as vibration while running under no load, but as play in one or more directions under load.

Makes sense, but surely this would show up when I'm pushing and pulling on the table more so than in a very light cut? My router has WAY more flex in it for comparison, as you'd expect.

 

[MiniMillMachinist]

I have 2 lemon machines, and one of them they just stop communicating with me pretty much blew my off. The other they had excuses why they weren't going to fix it/swap things out. So don't underestimate them not willing to fix it, and just blaming it on other things. Its a good time trust me.
Watch this video!

I'll let you know if my experience with Haas UK is better... so far I have high praise for them, but I suppose it is when problems happen that you find out the true nature of a company. Like I say, to this point everyone has been really friendly and helpful.

 

[MiniMillMachinist]

After reading up a little more on this, is it possible that what I may be facing is 'non-repetitive runout', with the spindle running only?

 

[Milland]

After reading up a little more on this, is it possible that what I may be facing is 'non-repetitive runout', with the spindle running only?

You should be able to check that with a micron-accurate measuring pin gauge (like Deltronic: https://deltronic.com/class-x-pin-gages/) that you put into your best holder, then run the spindle slowly (10 RPM or whatever) and map the indicator deflection over multiple revolutions.

That sort of multi-rotation error can occur as a result of non-concentric bearing races and improper ball matching, but I think that's unlikely in your case. If it is the spindle, I'd guess it's bent, or loose preload, or bad belt sending "pulses" as the motor/spindle rotates. Heck, if the motor drive and/or spindle pulleys are eccentric that could possibly do it.

Does your dealer have a similar machine at their facility? Take your material/cutter/holder/program and run some test cuts on their tools, see what you get...

 

[MiniMillMachinist]

You should be able to check that with a micron-accurate measuring pin gauge (like Deltronic: https://deltronic.com/class-x-pin-gages/) that you put into your best holder, then run the spindle slowly (10 RPM or whatever) and map the indicator deflection over multiple revolutions.

That sort of multi-rotation error can occur as a result of non-concentric bearing races and improper ball matching, but I think that's unlikely in your case. If it is the spindle, I'd guess it's bent, or loose preload, or bad belt sending "pulses" as the motor/spindle rotates. Heck, if the motor drive and/or spindle pulleys are eccentric that could possibly do it.

Does your dealer have a similar machine at their facility? Take your material/cutter/holder/program and run some test cuts on their tools, see what you get...

Some really good suggestions there, much appreciated! From my testing, there may have been a spot where the belt felt slightly tighter, but it was hard to say. Yes, I may make that request - that would tell me for certain if this is a general Mini Mill thing or a my Mini Mill thing.

A bent spindle - Jesus! How on earth would that happen?

 

[Milland]

A bent spindle - Jesus! How on earth would that happen?

It's a Haas - they could have used some spindles as rollers for moving a machine, then placed them back in the assembly line. Or the company softball team was missing some bats. Or...

 

[MiniMillMachinist]

It's a Haas - they could have used some spindles as rollers for moving a machine, then placed them back in the assembly line. Or the company softball team was missing some bats. Or...

Point taken!

 

[memphisjed]

A bent spindle - Jesus! How on earth would that happen?

never under estimate day shift.
Is there a difference between x only movement and y only? I am still leaning an axis issue, magnified by belt.

 

[MiniMillMachinist]

never under estimate day shift.
Is there a difference between x only movement and y only? I am still leaning an axis issue, magnified by belt.

Yes, slightly - it's a little better on Y

 

[Houdini16]

Also did you order your machine with a lot of $$ options, It seems that machines ordered without a lot of options, that have a lower profit margin, may actually be assembled with factory second parts.

 

[mottrhed]

You should be able to check that with a micron-accurate measuring pin gauge (like Deltronic: https://deltronic.com/class-x-pin-gages/) that you put into your best holder, then run the spindle slowly (10 RPM or whatever) and map the indicator deflection over multiple revolutions.

That sort of multi-rotation error can occur as a result of non-concentric bearing races and improper ball matching, but I think that's unlikely in your case. If it is the spindle, I'd guess it's bent, or loose preload, or bad belt sending "pulses" as the motor/spindle rotates. Heck, if the motor drive and/or spindle pulleys are eccentric that could possibly do it.

Does your dealer have a similar machine at their facility? Take your material/cutter/holder/program and run some test cuts on their tools, see what you get...

These are all true and good points, but measuring runout synchronous or non-synchronous at 10 rpm with an indicator is certainly not telling the whole story. To do this right you need non-contact sensors on 3 axes to measure this error at speed and see what's really going on. The tools exist, I have 2, and I've always said the oem or machine repair guys should have these units, but does the oem REALLY want to know how good/bad their product is... Nope, especially Haas, they don't make money on spindles, they make money on labor so getting to the bottom of the problem quick leaves money on the table and would show the customer what quality they really received. Plus these units likely cost more than the machine in question..

 

[Houdini16]

Do you have a X-Y accelerometer?

Haas has got a bad rep for delivering incomplete/unfinished machines, and that your settings were off at the beginning is almost certainly tied to the bad finish. But you need to know your beans if you start playing with servo settings on your own, I wouldn't on a new machine. But you'll likely have to lean on your support contact to get this done right, they'll try to half-ass it with less competent personnel otherwise.

Agree, on my DM2 with servo tuning issues, the standard tech., the higher end ball bar tech, neither had skills to adjust servos,
they actually sent out the service manager from another state to do it, they never fully dialed it in, like using a a o-scope, but did get it usable then bailed,
So you may need to get a tech who actually can tweak these servo settings, not just factory defaults, but since the issue increases with rpm, not feed, I am leaning toward spindle.

 

[MiniMillMachinist]

Also did you order your machine with a lot of $$ options, It seems that machines ordered without a lot of options, that have a lower profit margin, may actually be assembled with factory second parts.

I did spec a fair bit, including the Renishaw probe. I think I only really left the 4th axis on the table, so to speak.

These are all true and good points, but measuring runout synchronous or non-synchronous at 10 rpm with an indicator is certainly not telling the whole story. To do this right you need non-contact sensors on 3 axes to measure this error at speed and see what's really going on. The tools exist, I have 2, and I've always said the oem or machine repair guys should have these units, but does the oem REALLY want to know how good/bad their product is... Nope, especially Haas, they don't make money on spindles, they make money on labor so getting to the bottom of the problem quick leaves money on the table and would show the customer what quality they really received. Plus these units likely cost more than the machine in question..

Well, it's under warranty so it's up to Haas how quickly they want to fix it - I have had good service so far, but I will keep everyone posted. I hope this is resolved soon. They seem to want to look at the drawbar, but this was measured at 1050lbs, which is within the 780-1440 ish spec.

Agree, on my DM2 with servo tuning issues, the standard tech., the higher end ball bar tech, neither had skills to adjust servos,
they actually sent out the service manager from another state to do it, they never fully dialed it in, like using a a o-scope, but did get it usable then bailed,
So you may need to get a tech who actually can tweak these servo settings, not just factory defaults, but since the issue increases with rpm, not feed, I am leaning toward spindle.

While I'm pretty clueless at the moment, my gut tells me you are probably right. A Haas member told me the 500 value for D-Gain is correct, so it's likely the other settings are too. A bit annoying that I'm not able to look at this though!

 

[MiniMillMachinist]

Haas are wanting to replace the draw bar as they have seen this issue before, so hoping this fixes it. Now that I think about it my Renishaw probe originally came with a CAT40 holder and pull stud instead of a BT40 one, and neither myself nor the tech realised this and attempted a tool change with it in. It wasn't pretty and jammed in the umbrella, but I wonder if the pull stud could have damaged the draw bar. It kind of makes sense, when the tool is running faster and there is presumably more load on the draw bar the finish is worse. We will see...