Deckel vertical spindle - the 2 thrust bearings- what is the theory behind..

not using one of the ball raceway spacers supplied by the manufacturer (Ina) but rather, using the standard raceway spacer (with groove to match the ball diameters) on one side but using a flat hardened spacer with no groove on the other side ?

My guess ? More ball deformity via preload with less contact surface for less heat and possibly more "open" at the sides for better grease penetration ? What say ye ?



FP2 Activ, in this case

Good question. Deckel did similar with the FP2NC flip-head horizontal quill thrust bearing arrangement, having just flat precision washers for the balls to ride on. There, I am guessing the idea is to have the thrust bearings handle only thrust, and let the needle bearings handle only radial forces. With conventional thrust washers, the grooves provide radial restraint, which perhaps Deckel wanted to avoid. I am pretty sure it's not the grease exposure, and I am certain there should not be preload on the thrust bearings in the FP2NC. I doubt there is supposed to be thrust preload in the Fp2 Aktiv; it would be way too hard to control.

An engineer would certainly not use flat washers when load capacity was an issue.

Rich is exactly correct...don't want the ball thrust bearings to fight with the needle rollers for spindle truth....

As to preload...NO NO NO.....no preload on plate style thrust bearings be they needle roller or balls....Because they are working in direct compression, one element directly over the next, you can't tolerate any preload.
Always have to maintain some end float here....Deckel spindles call for about .0001-.0002 of end free play....Make then tight and the thrust bearings will fail..

AlfaGTA said:

Rich is exactly correct...don't want the ball thrust bearings to fight with the needle rollers for spindle truth....

As to preload...NO NO NO.....no preload on plate style thrust bearings be they needle roller or balls....Because they are working in direct compression, one element directly over the next, you can't tolerate any preload.
Always have to maintain some end float here....Deckel spindles call for about .0001-.0002 of end free play....Make then tight and the thrust bearings will fail..

Click to expand...

Right....I should have put "preload" in quotes...but did imagine some deformity of balls in extreme milling or drilling circumstances. But mostly I was just grasping at straws for possible reasons they did it that way....to not "fight with needle bearings for spindle truth" makes the most sense. Good to be reminded of the .0001-.0002 free play...you mentioned that in older posts concerning spindle rebuilds, but I forgot !

So it it my understanding that some Deckel spindles have hardened all flat trust spacers for both sides of the thrust bearing ? (in contrast to the FP2 Activ, which is one normal grooved spacer and one flat one)

AlfaGTA said:

Rich is exactly correct...don't want the ball thrust bearings to fight with the needle rollers for spindle truth....

Click to expand...

In practical terms, however, I seriously doubt there would be much fighting between the two when using P4 or better thrust washers, and any fighting would not amount to a hill of beans at the working end of the spindle, due to the long lever arm. Deckel may have been going for bragging rights, that's all.

Actually, its because they are "German" and just can't help it.....any lesser approach is simply "Not Permitted"
Cheers Ross

Milacron said:

Right....I should have put "preload" in quotes...but did imagine some deformity of balls in extreme milling or drilling circumstances. Good to be reminded of the .0001-.0002 free play...you mentioned that in older posts concerning spindle rebuilds, but I forgot !

Click to expand...

No doubt about deformity. The contact pattern progresses from essentially a point to a circular patch as the load increases, the ball flattens, and the race indents. In the case of a conventional curved bearing race, the contact patch is elliptical, because the groove envelopes more of the ball. Hence the contact stresses are smaller in the latter, and the bearing capacity is higher. That said, I have not seen any evidence whatsoever that the thrust capacity of Deckel's thrust arrangement is inadequate for the application.

For reference, the Aciera F4 and F5 use opposing tapered roller bearings at the rear of the spindle to handle both radial and thrust loads. Their axial clearance setting when cold is 5 µm (0.0002"). The nose bearing, which is a two-row cylindrical roller bearing with tapered bore, only handles radial loads.

AlfaGTA said:

Rich is exactly correct...don't want the ball thrust bearings to fight with the needle rollers for spindle truth....

As to preload...NO NO NO.....no preload on plate style thrust bearings be they needle roller or balls....Because they are working in direct compression, one element directly over the next, you can't tolerate any preload.
Always have to maintain some end float here....Deckel spindles call for about .0001-.0002 of end free play....Make then tight and the thrust bearings will fail..

Click to expand...

But just for discussion
A thrustbearing is designed to take a load It does not know the difference between a preload or a load does it ?
Setting a preload is difficult anyhow You just use the elasticity of the whole contrapsion and preload can be set too big very easily
( i have to be carefull and pick my words right otherwise you may put me on ignore like richard king did )

Peter

Peter...your never on ignore here.
Problem i think is that the thrust can run at zero clearance...but that will provide no place for the changes like thermal expansion....
They are sort of like needle roller bearings...gotta have some clearance....
Cheers Ross

Peter from Holland said:

But just for discussion
A thrustbearing is designed to take a load It does not know the difference between a preload or a load does it ?
Setting a preload is difficult anyhow You just use the elasticity of the whole contrapsion and preload can be set too big very easily
( i have to be carefull and pick my words right otherwise you may put me on ignore like richard king did )

Peter

Click to expand...

Peter,
You are correct. It is extremely difficult to control the preload for straight-loaded thrust or radial bearings. There is also the danger that a temperature difference would increase the preload to dangerous (to the bearing) levels due to differential thermal expansion. It is possible to design a spindle so that if it expands axially due to heating, the thrust bearing preload decreases, but what happens if the whole machine warms up and then the spindle cools faster than the housing? Then, the preload would increase.
RKlopp

rklopp said:

In practical terms, however, I seriously doubt there would be much fighting between the two when using P4 or better thrust washers, and any fighting would not amount to a hill of beans at the working end of the spindle, due to the long lever arm. Deckel may have been going for bragging rights, that's all.

Click to expand...

Not to mention the fact that one of the spacers (or ground washers, whatever one calls them) still has a groove /ball raceway, which would tend to "fight spindle truth" in thrust situations. I wonder if that one might be slightly smaller in OD, such that the entire spacer can move a bit in it's bore ? I'll try to remember to check that and compare the two.

Ross,

Could you explain in a bit more detail here for me?

I always thought that thrust bearings (conventional ones with curved races on both sides) needed preload. For example the support structure for the bevel gear above the Deckel vertical head is a needle bearing with thrust bearings above and below it. I was told that those thrust bearings should be lightly preloaded.

Cheers,
Bruce

Bruce,

The issue is that a ball or needle pure thrust bearing is too stiff for practical preload. With high stiffness, only a tiny change in movement of the races leads to a huge change in force. On the other hand, angular contact bearings and tapered roller bearings move the balls or rollers up a slope, and the stiffness is much smaller. You can still preload the latter kinds of bearings to high levels, but they are much less sensitive to changes in preload with small changes in movement, due to the reduced stiffness. Stiffness is the slope of the movement-force curve, in simple terms. The same problem occurs when trying to preload a pure radial roller bearing, like a Deckel needle roller bearing spindle. It is OK to use a spring to preload a ball or needle thrust bearing, because the spring limits the force.

RKlopp

ballen said:

Ross,

Could you explain in a bit more detail here for me?

I always thought that thrust bearings (conventional ones with curved races on both sides) needed preload. For example the support structure for the bevel gear above the Deckel vertical head is a needle bearing with thrust bearings above and below it. I was told that those thrust bearings should be lightly preloaded.

Click to expand...

Yeah, the bevel gear thrust bearing uses standard issue grooved ball race spacers on both sides. That one I can see needing a light preload since it shouldn't be subject to the same sort of forces put upon the spindle itself. Although it does rotate at the same speed obviously, since it in fact is what imparts the RPM to the spindle.... but most thrust forces should float independent of it via the keyway slot.

As to the solid versus grooved thrust bearing spacers, I measured them today and they are both exactly 60mm OD. What I had not noticed before is the bore they sit in is quite a bit larger than 60mm...more like 60.3mm.

That throws a kink in the "spindle truth" theory IMHO....so now I can only guess Deckel perhaps felt the grooved spacer receiving the most load was simply not heavy duty enough (Ina or Fafnir), thus the stronger all flat spacer was substituted for the normally used ball race spacer.

Have I opened a hornet's nest here or what ?

rklopp said:

The issue is that a ball or needle pure thrust bearing is too stiff for practical preload. With high stiffness, only a tiny change in movement of the races leads to a huge change in force.

Click to expand...

This makes sense to me, as does the idea that if you want a preload for this style bearing, use a spring. My issue is that this contradicts what I have been told in several cases, where there is no spring. For example Franz Singer told me to lightly preload the thrust bearings around the bevel gear which drives a FP2 vertical head spindle. It's hard for me to imagine that he's wrong about this, though of course everyone makes mistakes from time to time.

Cheers,
Bruce

I finally got around to taking the FP2 spindle apart, cleaning/regreasing the needle bearings and replacing the thrust bearings. I found nearly microscopic fracturing in small sections of both thrust grooved bearing ball raceways....in other words just barely visible to naked eye but I could feel the rough spots with my fingertip. Needle bearings and raceways appeared/felt as new after cleaning with carb cleaner (a more expensive, supposedly "no residue" carb cleaner)

Now with new thrust bearings and Kluber....OMG...what a difference.....will run at full 2,500 RPM so quiet and with almost no heat even after 20 minutes straight.

So, one wonders if the grooved spacers were either thicker or not grooved, might that fracturing never have occurred ? Or perhaps previous owner didn't allow for the .0002" play necessary to avoid too much ball stress when overheated ?

======================================================================

As an aside, when you order these thrust bearings from Singer, it is confusing at first because they send you the full thrust bearings with 2 grooved spacers per bearing, plus the two all flat spacers. In other words you don't use two (one per bearing) of the grooved spacers that are included anyway.

Further fun is the fact that although each is 60mm OD, the grooved spacers have slightly different ID's, such that one will fit snug on the spindle whereas the other is too loose ! At first I installed the one that was too loose and had a momentary freak out where I thought they sent me the wrong bearing......but then tried the other spacer and it was a perfect fit.

Thats on all thrustbearings The smaller ID is the rotating race fitting tight around the spindle The other is the staitionary one That needs to be otherwise both races spin around Not ment to be
Peter

Peter from Holland said:

Thats on all thrustbearings The smaller ID is the rotating race fitting tight around the spindle The other is the staitionary one That needs to be otherwise both races spin around Not ment to be
Peter

Click to expand...

Yeah, that makes sense. Didn't occur to me on the Deckel as the rotating spindle OD is turned various diameters such that it could use thrust bearings with the same ID as long as they were both the smaller ID. But then one doesn't use the second one regardless, but rather a flat spacer/washer....of which we still have yet to have a definitive answer as to why they did it that way ! Theories abound, but noting definitive....another Deckel mystery...

Talking about Deckel mysteries
I talked to a guy yesterday that worked at ACEC in Gent Belgium that made Deckel FP1 2 and 3 there
He said their castings were much better as the german ones
This college talks about it too on his website
Over ons

Peter

Peter from Holland said:

Talking about Deckel mysteries
I talked to a guy yesterday that worked at ACEC in Gent Belgium that made Deckel FP1 2 and 3 there
He said their castings were much better as the german ones
This college talks about it too on his website
Over ons

Peter

Click to expand...

Google translate from Dutch...


How are we actually started?
============================
The company machines Crispyn Founded in 1982, though Crispyn much longer involved with metalworking machines.
3 generations ago, my grandfather Adolphe Crispyn employed by ACEC firm in Ghent.
By ACEC they made when the legendary Deckel FP1 and FP3 models.

These were conventional milling machines to a quality not to emulate today. My grandfather was at least partly responsible for the casting of the Deckel parts.Deckel was a German manufacturer of milling machines which still exists under the name DMG Mori which now manufacture quality CNC machines."Parts were immediately remelted at least casting mistake that Germans were there incredibly strict on" I can still hear my grandfather tell.

Ghent had developed a special type of cast iron and Germany could not match the quality.
There is until today still exist Belgian Deckels which are known as the better Deckels.

The son of Adolphe, my father Wilfried had as a child bitten by mechanics.
On the kitchen table at home there were dozens of gears and shafts of the mechanical work of a clock.
"But now he will never hear more together" Adolphe's father said Wilfried.
Nothing was further from the truth, Wilfried dismantled and restored the toughest mechanical parts.

When Wilfried graduating as a metalworker he worked as a technician in various companies including many years with a machine dealer.
There was Wilfried risen to the specialist mechanical repairs on brands like Behringer, Eisele, Alzmetall, Peddinghaus name it.
Several times when Wilfried has been to producers and he has as his specialization evolved to what it is today.

In 1982 it was time to start their own business.
First single with the official car spot repairs and do something later Wilfried rented a shed.
With this pilot Wilfried was also well equipped to do complete overhauls of machines.
With a lathe, a milling machine and saw Wilfried could do all lot of work.