What's new
What's new

HSK A-63 on a HAAS, Anyone own/driving this spindle (yet) ?

cameraman

Diamond
Joined
Nov 24, 2014
.
.

Like the title says, lol
.
HSK A-63 on a HAAS, Anyone own/driving this spindle (yet) ?
.
.

" Haas Automation is now offering two new tool holding options: HSK-A63 and HSK-F63.

The HSK-A63 option has been integrated into our previous 40T 15k inline spindle design, and this will be built on our production floor. The HSK-A63 utilizes clamping unit style B by OTT-JAKOB. The HSK-63A spindle option is currently available on the UMC-1000/SS, the VM-2, VM-3, and VM-6 with optional max speeds of 12,000 or 15,000 RPM.

The HSK-F63 option will be purchased as a complete spindle from HSD Mechatronics. It will have a max speed of 25,000 RPM. The HSK-F63 option will only be available on future machines, the GR-712 5-axis, and the GM-2-5ax which are not yet available. This information is accurate as of 6/26/2019 and will be updated as more options become available.
"

Taken ^^^ from HSK Spindle Basics
on Haas' website.


+ UMC 500 option. (I think it's pegged to 15K rpm on the HAAS UMC 500.). 15,000-rpm Spindle, HSK-A63

- Anyone using / running this spindle option on any HAAS machines ?

- Any good ?

- Does it meet HSK "Expectations"

- Wished you got something else instead (option wise) or happy you did ?

- If so what brands / tool holders you using etc. materials cuts and applications etc. etc. ?

TIA.

________________________________________________________________________________________


Will Edit / add if needed later etc. :cheers:
 
I too am interested, I've always thought their spindles were kinda soft. A few months back we had the baldor motor seize tight and was replaced with a yaskawa and new spindle due to coupling changes. We run the full 12k on a regular basis.
 

Bueller, Bueller...

Anyone ?

No worries, early days, very few of these out in the wild yet. Seems HAAS have been testing these for at least a couple of years I think ?

Maybe a year from now some sort of picture will gradually emerge.


Long term reliability takes (obviously) years to map out so if one is taking a chance on the HAAS HSK option it's still going to be a bit of a risk in terms of how the thing will actually perform and will be slightly different from the kinds of tweaks and improvements they may make over a decade.

I guess kinda depends on how much someone is really into HSK and HSK + HAAS idea.

OR (maybe) like plastikdreams looking for a viable spindle replacement ?
 
It seems like a total waste of money.

HSK at 15k seems a bit pointless as none of the advantages over CAT-40 will really present themselves at those speeds. What you do get is a very lightened wallet - $7k for the HSK 15k, plus another $14k for the 50 tool changer. If you were buying the 15k + 50 tools, you're only looking at $7k additional... but!

Now you can go from paying Frank $150 for very high quality CAT40 tool holders, to paying $300+ for way more exotic HSK, and I really don't see how you'll be getting the most out of any of that investment when you are only at 15k on the RPMs.

What this machine is screaming for is a Big Plus CAT40 spindle. At this spindle speed, and target audience? That would be perfect for dramatically increasing rigidity and utility.

OR if the HSK option brought with it a 20k spindle. That would be a nice performance bump and start justifying the increased machine and tooling cost.

Alas, the way Haas is doing it is neither fish nor fowl.
 
It seems like a total waste of money.

HSK at 15k seems a bit pointless as none of the advantages over CAT-40 will really present themselves at those speeds. What you do get is a very lightened wallet - $7k for the HSK 15k, plus another $14k for the 50 tool changer. If you were buying the 15k + 50 tools, you're only looking at $7k additional... but!

Now you can go from paying Frank $150 for very high quality CAT40 tool holders, to paying $300+ for way more exotic HSK, and I really don't see how you'll be getting the most out of any of that investment when you are only at 15k on the RPMs.

What this machine is screaming for is a Big Plus CAT40 spindle. At this spindle speed, and target audience? That would be perfect for dramatically increasing rigidity and utility.

OR if the HSK option brought with it a 20k spindle. That would be a nice performance bump and start justifying the increased machine and tooling cost.

Alas, the way Haas is doing it is neither fish nor fowl.

Pretty much other than Z consistency (as you know) esp. for 5 axis.

I don't have any clear idea why HAAS didn't dooooooo the Dual contact Big Plus @ 15K rpm ?
 
Pretty much other than Z consistency (as you know) esp. for 5 axis.

I don't have any clear idea why HAAS didn't dooooooo the Dual contact Big Plus @ 15K rpm ?

If you are so fussy about tool length repeatability to the micron, to the point where the HSK advantage here has a material difference on your requirements... you aren't buying a Haas.
 
I don't have any clear idea why HAAS didn't dooooooo the Dual contact Big Plus @ 15K rpm ?

I can think of a few reasons.

Speculatively, they may have been refused a license, or it may be that they didn't like the licence costs. It's plausible that BK would prefer not to be associated with Haas, seeing that as a dilution of their premium brand.

Haas machines AFAIK are all CAT40, and Big Plus is BT-MAS, so offering BBT on only a portion of their machines would require a split of their production lines while moving to BBT across the board would drive their prices up and probably not be seen as a great benefit from the perspective of most Haas buyers.

On the other hand, HSK is an easy out for Haas as there are a number of sources of quality bolt on HSK spindles that they can easily pull into their supply chain.
 
If you are so fussy about tool length repeatability to the micron, to the point where the HSK advantage here has a material difference on your requirements... you aren't buying a Haas.

I have to admit HAAS + HSK is a trifle paradoxical.

So I would like to see dynamic testing and parts and tooling comparisons on the HAAS platform - specifically new UMCs.


^^^ I posted this on SVFeingold thread Okuma or Spedio / jamming a HAAS UMC 500 into a small garage thread.

@2:30 two minutes and 30 seconds in you see a part that I would not think in a million years you could do on HAAS UMC... There they talk about use of HSK. Is the part really a shambles and they only show 1/2 seconds footage so you can't get a close look.

HAAS themselves say (on their website) that the HSK interface provides better radial concentricity (especially for lonnger tools) and more consistent Z (inherent to dual contact). I assume as compared to their regular CT 40 spindle nose...

Better is Better … if it is "Better" ?

You can map out a lot of behaviors on a machine to get things dialed in on multiples of the same part, I think that's somewhat easier to achieve with a dual contact interface. Removing one level of "Randomness".

Is it worth it for a HAAS ? TBD.

That's why I'm throwing that out there and thread starter "Anyone driving a HAAS HSK ?" ~ I have to HAASK ?
 
If you are so fussy about tool length repeatability to the micron, to the point where the HSK advantage here has a material difference on your requirements... you aren't buying a Haas.

5 axis Yasda I worked on was 40 taper non-BigPlus and it had no issues with tool length variability. 15k RPM spindle. It was not used for heavy cutting though, mostly holding tight tolerance positional relationships between features on multiple faces. That may have made a difference.
 
5 axis Yasda I worked on was 40 taper non-BigPlus and it had no issues with tool length variability. 15k RPM spindle. It was not used for heavy cutting though, mostly holding tight tolerance positional relationships between features on multiple faces. That may have made a difference.

A YASDA 40 taper spindle nose vs. a HAAS 40 taper spindle receptacle ?

The problem is HAAS is not going to go out of their way to show how 'Deficient" their regular 40 taper interface is in respect of their new HSK interface and vice versa.
 
The only reason I can see for a Haas having an HSK spindle would be to allow a company with high grade machines and a big investment in HSK toolholders to not need to buy a bunch of 40 taper holders when they step down to a Haas.

It also levels the comparison when the bean counters get involved with a machine purchase decision. "Look, same spindle at half the price"
 
The only reason I can see for a Haas having an HSK spindle would be to allow a company with high grade machines and a big investment in HSK toolholders to not need to buy a bunch of 40 taper holders when they step down to a Haas.

It also levels the comparison when the bean counters get involved with a machine purchase decision. "Look, same spindle at half the price"

I was wondering that too, so if you have a bunch of German machines (HSK'd up to the teeth) then the HAAS HSK "Machine" isn't "cock blocked" right form the start by being CT 40. And as you say more compatible with the bean counters by using existing tooling.

It is possible that the HSK interface on the HAAS has a very minor / infinitesimally small technical advantage on a HAAS machine, but that is relative to the regular 40taper offering on HAAS. Unless you buy both and run them side by side + rigorous testing , hard to know ?
 
Watching this thread with some interest, disappointed that these haven't been out in the wild enough for any real data. I do have a few thoughts it'd be nice to get comments on.

1. As far as I see it, from the info that's freely available, HSK63 should be considerably more rigid than standard 40-taper (except for torsional stiffness). If it weren't, I wouldn't be able to read dozens of threads praising dual contact spindles. I've seen very little quantitative data on this one way or the other that wasn't sourced from a manufacturer of HSK. None, actually. I've seen a few opinions in other threads along the lines of "40 taper has a taller shank than HSK63, so it must be more rigid." To me that sounds kinda like "That car has bigger wheels, it must be faster." Intuition isn't always correct. Understanding stiffness in structures was one of the tougher things for younger engineers I've mentored in the last couple years to "get." E.g. "That big fuckoff 2" solid steel plate isn't a quarter as stiff as you think it is, and you could achieve 4x the result for 1/4 the cost and 1/10 the weight in aluminum." Not even getting to the fun stuff like what a preloaded interface (bolted connection for instance) does in response to applied forces. Something almost nobody gets quickly because it's non-intuitive.

2. The actual z-stability/growth of a standard 40-taper toolholder (with speed or temperature or whatever else) over, say, a 4 hour toolpath is...what? I've seen references to ~25 microns or thereabouts being possible. Far more than the 1-micron number thrown out above. And 25 microns is more than enough to destroy a surface finish. Moreover, probing at 0 RPM is all well and good. What happens at 15k RPM for a dual-contact vs standard-taper tool? Surely this can be measured with laser toolsetters but I haven't seen anyone mention it yet. I can totally buy that the effect is minor. But unless it's nonexistent then going from, say, a 20um delta to a 15um delta is still better. As good as a 1um delta on a higher class machine? No, but still better than before. Whether that's worth it is then up to the end user.

3. Given that tolerance/rigidity are cumulative, I don't really get the mindset of "it won't matter because Haas." Maybe it's negligible but, see above comments about the dual contact praise. It seems like many people think that Haas machines are so floppy that the extra tool/holder flex won't matter at all. I'm skeptical. A tighter tolerance laid over a looser tolerance still results in...a tighter tolerance overall, compared to two loose tolerances overlaid. More rigidity is more rigidity. It's certainly nice to have everything be rigid, but if that's not the case then adding rigidity where you can is usually better than doing nothing. Is it just not worth it, despite the benefit?

4. Going back to data, it's exceedingly hard to find any. One-off finished parts don't really prove much as data points, forsaking all other factors. I've seen trash come off of expensive machines and beautiful tight-tolerance parts come off Haas machines. Showing a one-off incredible (or terrible) surface or part doesn't really prove anything. Except of course for the moon-machines like Kern putting out optical mirror finishes. The folks in here know as well as anyone that process is important, and the machine is a part of that process. An important one no doubt, but still only a part. When production equipment would make bad parts we didn't just scrap the cell and start over. I mean we still cursed it, but then we root-caused the problem using actual data and measurements and brought it into spec.

5. Final point about data...it's a rare thread here (or anywhere) where someone used a tenths indicator to probe a column (or holder, or the end of a tool) and then applied a known force to it, then repeated that in multiple directions, then did the same on another machine to compare. Nor anyone using an instrumented hammer or accelerometer to check the transient response of a casting or some such. Surely tons of people out there in shops have access to these tools but I've literally never seen it done, or at least the results shared. MTBs most definitely do those things but they like to play all secret squirrel about it, so it ends up being mainly just conjecture and repetition of the same points over the years, sometimes using data or anecdotes that are from 10-20 years ago when X machine might have been very different from today.
 
Watching this thread with some interest, disappointed that these haven't been out in the wild enough for any real data. I do have a few thoughts it'd be nice to get comments on.

1. As far as I see it, from the info that's freely available, HSK63 should be considerably more rigid than standard 40-taper (except for torsional stiffness). If it weren't, I wouldn't be able to read dozens of threads praising dual contact spindles. I've seen very little quantitative data on this one way or the other that wasn't sourced from a manufacturer of HSK. None, actually. I've seen a few opinions in other threads along the lines of "40 taper has a taller shank than HSK63, so it must be more rigid." To me that sounds kinda like "That car has bigger wheels, it must be faster." Intuition isn't always correct. Understanding stiffness in structures was one of the tougher things for younger engineers I've mentored in the last couple years to "get." E.g. "That big fuckoff 2" solid steel plate isn't a quarter as stiff as you think it is, and you could achieve 4x the result for 1/4 the cost and 1/10 the weight in aluminum." Not even getting to the fun stuff like what a preloaded interface (bolted connection for instance) does in response to applied forces. Something almost nobody gets quickly because it's non-intuitive.

2. The actual z-stability/growth of a standard 40-taper toolholder (with speed or temperature or whatever else) over, say, a 4 hour toolpath is...what? I've seen references to ~25 microns or thereabouts being possible. Far more than the 1-micron number thrown out above. And 25 microns is more than enough to destroy a surface finish. Moreover, probing at 0 RPM is all well and good. What happens at 15k RPM for a dual-contact vs standard-taper tool? Surely this can be measured with laser toolsetters but I haven't seen anyone mention it yet. I can totally buy that the effect is minor. But unless it's nonexistent then going from, say, a 20um delta to a 15um delta is still better. As good as a 1um delta on a higher class machine? No, but still better than before. Whether that's worth it is then up to the end user.

3. Given that tolerance/rigidity are cumulative, I don't really get the mindset of "it won't matter because Haas." Maybe it's negligible but, see above comments about the dual contact praise. It seems like many people think that Haas machines are so floppy that the extra tool/holder flex won't matter at all. I'm skeptical. A tighter tolerance laid over a looser tolerance still results in...a tighter tolerance overall, compared to two loose tolerances overlaid. More rigidity is more rigidity. It's certainly nice to have everything be rigid, but if that's not the case then adding rigidity where you can is usually better than doing nothing. Is it just not worth it, despite the benefit?

4. Going back to data, it's exceedingly hard to find any. One-off finished parts don't really prove much as data points, forsaking all other factors. I've seen trash come off of expensive machines and beautiful tight-tolerance parts come off Haas machines. Showing a one-off incredible (or terrible) surface or part doesn't really prove anything. Except of course for the moon-machines like Kern putting out optical mirror finishes. The folks in here know as well as anyone that process is important, and the machine is a part of that process. An important one no doubt, but still only a part. When production equipment would make bad parts we didn't just scrap the cell and start over. I mean we still cursed it, but then we root-caused the problem using actual data and measurements and brought it into spec.

5. Final point about data...it's a rare thread here (or anywhere) where someone used a tenths indicator to probe a column (or holder, or the end of a tool) and then applied a known force to it, then repeated that in multiple directions, then did the same on another machine to compare. Nor anyone using an instrumented hammer or accelerometer to check the transient response of a casting or some such. Surely tons of people out there in shops have access to these tools but I've literally never seen it done, or at least the results shared. MTBs most definitely do those things but they like to play all secret squirrel about it, so it ends up being mainly just conjecture and repetition of the same points over the years, sometimes using data or anecdotes that are from 10-20 years ago when X machine might have been very different from today.

Sandvik have published data on the relative stiffness of the various spindle interfaces. The objective obviously was the tout the advantages of capto, but the general trend was steep taper > dual contact steep taper > HSK > Capto.
 








 
Back
Top