What's new
What's new

FP2 non keyed X axis nut, why??

thanvg

Hot Rolled
Joined
Mar 3, 2015
Location
Greece
Hi Deckel fans,

today, my FP2 almost scared me. I am saying almost because I have taken the saddle apart so it didn't take long to figure out what was the cause of what I was seeing.
So, I was roughing out a slot, using a 16 mm roughing end mill at 12 mm depth of cut, so I was pushing kind of hard, when I noticed irregularities in the sound of the cut. Seemed as if the feed was pausing for a while and starting back again. Indeed, checked the DRO and, while the handwheel was rotating at constant rate, DRO seemed to freeze for a while then resume for a while and so on.

So, removed the belows on the left side of X and saw the nut rotating with the screw, as expected. (couldn'd have been the shear pin because then the handwheel wouldn't have been rotating in such a constant rate). Tightened the ring nut and all returned to normal (except for the poor end mill that broke during the initial, non-consistent feed).

This got me wondering again, why on earth would these guys leave the X nut without a key? Y and Z are properly fixed (ok, Y is rotating but the screw is fixed), as one would have imagined. Also, on my ALG100 X nut was keyed, as I think is the case with the FP1, right?

Does anybody have any explanation on this, strange, design decision? Can't be a safety thing, there is the shear pin for that. I would be very interested to hear if there was some kind of rationale there...

BR,
Thanos
 
Hi Thanos,

removed the belows on the left side of X and saw the nut rotating with the screw, as expected

This is NOT what is expected. Normally the lead-screw nut is fixed and the screw rotates. What happened was that the ring nut that secures the lead-screw nut was too loose and the lead-screw nut started to rotate with the lead-screw. This should not happen (which obviously you understand, because that's what you wrote later). So I guess you meant "expected from the bad behaviour" as described below.

Anyway, the point is that the ring nut which retains the lead-screw nut needs to be TIGHT because otherwise you will get end play in X. So there is no need to key it. If the ring nut is loose you will get end-play, so the ring nut has to be tight.

Thanos, you need to make a custom wrench to tighten this up properly. You can see pictures of at least three different wrenches (mine, Ross's and Bill's) in my long FP2 teardown thread.

Here is mine: http://www.practicalmachinist.com/v...own-reassembly-277534/index5.html#post2228664

Here is Bill's: http://www.practicalmachinist.com/v...own-reassembly-277534/index3.html#post2219224.

Here is Ross's: http://www.practicalmachinist.com/v...-x-axis-issue-320074-post2762734/#post2762734

Cheers,
Bruce
 
Also, on my ALG100 X nut was keyed, as I think is the case with the FP1, right?

Wrong. ;)

The FP1 x-axis nut is not keyed and I think there's no key simply because there's no need for one (assuming the ring is properly tightened, that is).
No need to look further.
 
Been running an FP2 for 20 years+..never an issue with the nut rotating..(made a replacement for the nut some 15 years back). Key is not needed, but good mechanical judgement and the proper use of well fitting tools is a must.

System / design is fine...just need to apply proper assembly/maintenance routine....
Cheers Ross
 
Hi guys, thanks for the replies,

Thanos, you need to make a custom wrench to tighten this up properly. You can see pictures of at least three different wrenches (mine, Ross's and Bill's) in my long FP2 teardown thread.

Hi Bruce, thanks for the resources, I'll get around to build my own one of these days...yesterday, I used the 'other' method of tightening the nut, you know which one...:)

I think he meant as expected in light of the observed bad behavior, not as expected of proper behavior.

Yeap...should have rephrased probably.

Wrong. ;)

The FP1 x-axis nut is not keyed and I think there's no key simply because there's no need for one (assuming the ring is properly tightened, that is).
No need to look further.

No key, right TNB...sorry for that, it was late in the night. I think I got confused with the feed sleeves that are keyed to the screw and some spacers that were keyed to the saddled at the ALG, but not sure.....

Been running an FP2 for 20 years+..never an issue with the nut rotating..(made a replacement for the nut some 15 years back). Key is not needed, but good mechanical judgement and the proper use of well fitting tools is a must.

System / design is fine...just need to apply proper assembly/maintenance routine....
Cheers Ross

Hi Ross, not having a nut rotating in 20 years is a good enough testimony that this design can be made to work. However, allow me to argue that a maintenance free design is always a better design. Deckel designers, we all know, put a lot of effort designing these machines, and I am sure that they did not take the route:

"Fritz, why did we leave the X nut without a key? Don't worry Klaus, we'll tell them to tighten the ring nut often and it'll be alright"...

After all, in Y and Z, nuts are secured from rotating (not quite true for Y, but key in nut and pin in screw do what I am talking about). Their own aproach on the other two axes kind of proves that some sort of securing against rotation is a common approach and, I would assume (though without having taken apart any other milling machine) that this is the norm in machine tools, but I might be wrong.

So, I'll return to my question. Does anyone have an idea why did they choose to go down the particular path with the X axis nut in particular? Did it facilitate production so they let it this way as being 'good enough'? Not very possible since they don't seem to have being trying to save on effort and production cost by compromising parts.

Asking out of pure academic interest....

BR,
Thanos
 
a maintenance free design is always a better design. Deckel designers, we all know, put a lot of effort designing these machines, and I am sure that they did not take the route:

"Fritz, why did we leave the X nut without a key? Don't worry Klaus, we'll tell them to tighten the ring nut often and it'll be alright"...

To a certain extend, even with a loosened ring, a keyed nut will result in axial play, so the ring HAS to be tightened anyway.
But then, the key becomes useless. ;)

As I see things, the screws of a milling machine are subjects to very light axial loads and the related rotationnal effect on the nut is probably pretty low.

Imagine the effort you'd have to exert on the handwheel to load the x-axis screw of your FP2 as much as you load a simple M12 stud when with your 19 wrench when you clamp a part on the table of the machine... Ouchhhh !
We never get even close under normal working conditions.

For that reason, I think that Fritz and Klaus simply choosed the best combination of function/ease of manufacturing/ease of installation/removal for their design.
For the x-axis a key did not seem mandatory and for the other axis, I'm not sure that the design they choosed was mainly dictated by the fear of seeing the nut spinning in its housing.
 
Hi guys, thanks for the replies,
Hi Ross, not having a nut rotating in 20 years is a good enough testimony that this design can be made to work. However, allow me to argue that a maintenance free design is always a better design. Deckel designers, we all know, put a lot of effort designing these machines, and I am sure that they did not take the route:
BR,
Thanos

Hi Deckeleers,

Thanos has put forward a very valid point but as usual on this forum, where some members will go to any lengths to explain away the bad points of the Deckel designs, black has indeed been attempted to be changed to white!

Instead of trying to fob off members that dare to criticize the Deckel designers that in their opinion walked on water, why don`t you tell us another machine that does not in some way positively resist the rotation of leadscrew nuts and I don`t agree that a screwed ring is a positive restriction to rotation as per the Deckel design.

Another point that you should consider is that if the X axis nut just rotates a few degrees from normal then in the case of the FP1 the nut will not get lubricated!

Alan
 
Hi to all,

my feeling is as Alan described, I think that all similar nuts in machines are keyed, even 2 out of 3 on the FP2 indeed are (more are than are not, on the same machine), which is what got me thinking and asking for a robust explanation.

On top of that, we are talking about a precision instrument in the non-DRO days. So, if the user was not able to afford the, as I can imagine, insanely expensive optical measuring system, the case of the nut coming even a little bit loose would, possibly, result in inexplicably wasted parts.

TNB you are right in that even a keyed nut would need some kind of axial stop, e.g. a ring nut. However, we have to note that the ring nut's life becomes much easier if it only has to do one thing (limiting axial movement of the nut) in the case that the nut is keyed.
(though, at this point, I have to admit that, unless I got it wrong, I think that the motion that loads the ring nut, that is moving the table towards the the operator, tends to screw the ring nut instead of unscrewing it. Well, it still manages to get it self loose somehow, doesn't it?)

BR,
thanos
 
Last edited:
Well, it still manages to get it self loose somehow, doesn't it?

So in the end, the question is : do you now intend to key the nut of your FP2, or will you consider the design good enough nevertheless ? (wich would be a way to answer your own question) ;)
 
Thanos, I think the basic design is fine. As I said (and as TNB repeated) having a key is useless because if the ring nut gets loose or free (which is the only case where a key is needed!) then you have axial play in the nut which destroys accuracy and makes climb cutting dangerous. So put your time and effort into making a good wrench for tightening the ring nut and you will not have this problem again. Cheers, Bruce
 
Hey guys,

thanks for the replies, although they still answer another question than the one I asked. There wasn't ever a debate about what I'll do next, of course I'll make sure the ring nut is properly tighthened, with a proper tool and include it in the regular maintenance schedule.

My question was on machine design/production and I think that I posted enough arguments to support asking the particular question, and I hoped for some inside information or a fruitfull discussion. I may be missing something that you guys, with more experience, might know. I mean arguments supporting the reason why Deckel engineers chose this particular design on this particular leadscrew and comparison to the other two keyed screws on the same machine and other keyed screws on manual and CNC machines showing the differences that justify the particular choice.

Anyhow, there is no reason to beat this horse any more, I'll answer myself as in 'yeah...right...this is a strange enough design, haven't seen it much on other machines....wondering why they did it, but I don't really know...'

BR,
Thanos
 
Hi Deckeleers,

Thanos, You let them grind you down! That design should be keyed as well as axially located with the ring! Why didn`t Deckel do this? Nobody has submitted a convincing argument about this omission, perhaps The Deckel designers just forgot to include it or it was a cost saving! Your comment is more valid than those others that we`ve seen on this thread.

Now why did the leadscrew nut on your FP2 let go? is there something amiss with your machine, is the outer diameter of the nut too small? A key would have prevented any torque applied to the nut from loosening the retaining ring, that`s a fact!

Alan
 
Hi to all,

Alan, you can't force an answer out of anybody. As I said, I was hoping for some productive discussion on this, let say, peculiarity of Deckel, but the forum chose to focus elsewhere.

It is, indeed, something trivial, but I thought this forum never feared (productively) discussing triviallities....

Let's leave it here I say

BR,
Thanos
 
Thanos please do take what I will write without the slightest offense, but the fact that you're not convinced by the answers that were given to you does not mean that your question was left unanswered.

Deckels are not perfect machines and they have their flaws. There's no argue about that, but most are well known by now.

To my known (not infinite, I admit :rolleyes5:), the particular problem you encountered is discussed here for the very, very first time.

We're talking about a design that evolved to the form we still know now back in thirties... may be the forties ? At least 80 years ago ! A design that was applied to at least two models (FP1 and FP2) that prooved successfull more than any other comparable machines in the world, and remained unchanged through various evolutions...

So if a key was really mandatory for the machine to work properly, don't you think Deckel engineers would have realized it over the years, or the problem would have already been discussed more than once ?

I don't want to be a smart ass and don't claim there was never a loosened ring beside yours, but I think the statistics are against you nevertheless (and we don't have no evidence that it's the rotational effort of the screw that loosened your ring to begin with. Could very well be the result of a sloppy tightening by the previous owner after a nut change).

Note that I never wrote that the keyless design was better than it would have been with a key.
I wrote that it was *good enough* for the intended task and that was probably the only answer to your question.

I sincerely think there is no other reason other than the related benefits of not over complicating a design that works : ease of manufacturing and cost savings. Isn't it enough ?

I would not have answered this post but like you, I would like the forum to remain a place for productive and friendly discussion.
That can be only be if one accepts to consider other's arguments and don't try to make a fight out of every exchange.

So again, do not get offended because nobody ever tried to "grind you down".
 
Hi TNB,

Thanks for taking the time to reply, no worries, no offence taken, still treating this as the friendly forum we know.
So, let me try to explain what I was after, again by example:

'dad, why isn't this nut keyed?' 'Never mind son, it works this way' 'Yes, but all my friends' mills have keyed nuts, what different with this one??' 'Er....it's good enough'.

See what I mean? Treat me like an ignorant newbie and indulge my curiosity! I am realy confused as to why they chose this path, since I kind of feel that this is not the norm in machine tools. Indeed you did provide an answer, with the good-enough/cheap/easy-to-produce argument, but could this be the case with those guys? I mean, they went a long way to build these machines that it does not make sense to me that they compromised their design for cost savings. And also, if Deckel had proven that this is the way to go, wouldn't someone else have copied this design? That was the main question that was left unanswered, why them and not the others, why X and not Y or Z...?

As I said earlier, that's all trivialities, but if you are a gearhead this kind of trivialities are the essense of the whole thing...

BR,
Thanos
 
I don't think I've heard of another incident where the nut turning caused a problem but yes it could be keyed and it would never happen. Now that it is brought to light and giving it some thought I don't think it really should be a problem?
My first thought is how tight mine fit in the saddle, had to use a plastic tipped mallet to drive it out and in, on top of that is the spanner nut to insure it won't move. I think with proper lubrication there will never be enough friction from the fit of the lead screw to nut to move the assembly. If there is that much friction then there is something other than a missing key creating a problem. It seems there would need to be chips packed in the assembly or the screw would have to have an awful finish on the threads to cause enough friction to break the nut loose from its mounting.
I'm just thinking out loud here and offering my 2 cents worth on the subject but I'm happy with that design aspect of the Deckel FP line.
Dan

PS Another after thought; If there was a key there and the nut turned imagine the problem of fixing it after words.
 
Hi Thanos,

To answer your question, when there are keys in the lead-screw nuts, they are there for a reason. For example the Y-axis nut rotates on a fixed lead screw, and is driven by a bevel gear which fits over the lead-screw nut. The key is needed to "lock" the bevel gear to the lead screw nut. The axial motion of the lead screw nut is controlled by a separate thrust bearing. So even if the thrust bearing is working correctly, it won't prevent the bevel gear from rotating relative to the lead-screw nut.

A good design achieves its goal with a minimum of parts. In the case of the X-axis lead-screw nut, a key is not needed, because the ring nut fixes the lead-screw nut BOTH against axial motion AND against rotation. In the case of the Y-axis, the bearings that prevent axial motion do NOT prevent the bevel gear from rotating on the lead-screw nut. Hence the key is needed there.

Cheers,
Bruce
 
Hi guys,

ok, let' s say that a key is not needed in such nuts. I'll make an assumption here, if I am wrong please correct me, if I am right, please explain. My guess is that X nut on a Schaublin 13, an Aciera F4 and, let's say, a Bridgeport, are keyed (never seen any apart so I am really guessing, bar the BP where there are online resources). If my assumption is right then there should be an explanation, not? Why did the other designers not adopt the 'good enough/minimum effort' option and went for something more complicated? Or was it just progress, since the two former are Deckel 'copies' and their designers found some weak points needing improvement?

I checked the manual of the ALG, have a look here:

attachment.php


I think the round thing in the drawing, below the nut, it a cross pin that keys the nut but can't be sure(which also has backlash adjustment provision, nice).

Was this keyless design kept on the squarehead FP2s as well? (ballscrew machines I don't think so, I think ball screw nuts all have some kind of provision for fixing them robustly there)

Hoping that my point on the X nut on the above euromills is valid....
BR,
Thanos

edit: by requesting such a comparison to other existing machines with keyed nuts, I am not after a 'bad Deckels, good Schaulbins' thing. I am after a technical explanation like, for example, 'deckel chose to machine a very tight fit to allow for perfect alignment of the carriage-nut-screw-table assembly, so key was rendered useless, whereas others did not pay that much attention there and have a loose nut that needs to be keyed'. See what I mean? Why is this design good enough for an FP2 and not good enough for the rest of the machines? That's what started the whole post to begin with, that comparison is what got me thinking and asking for ideas on this. Apart from that, the fact that this design is good enough for the FP2 (which I've mentioned several times how much I admire) is nothing that ever needed debate about)

Also, @Danny, I think on mine the nut is not that tight of a fit in the carriage. Maybe is has slipped and rotated a couple of times in its life, which made the fit not perfect, but, you know, this wouldn't have happened if it was keyed, the egg and chicken thing....;)
 
Wow....having trouble moving on from this?
My belief is that you should add a key to your "X" axis lead screw nut as it is obvious that in spite of years of trouble free operation with the Deckel setup,covering literally thousands of machines that,have no key, and no trouble,you seem unable to cope with the simple idea that the ring nut without any key really works fine....Geez...Keep beating this dead horse will ya.

Just because others did it differently is not the standard.that all designs MUST follow...The proof is in the operation. The design works...If the ring nut is tightened using a proper spanner there is no issue...nothing slips , nothing comes loose it just works,
I have years of real machine use to back this up...both on the FP2's and the later designed FP3's.....The nut retention setup is sound...Tighten the ring nut and forget it...

If you are troubled by the "design fault" I suggest you fabricate a real spanner, then tighten the ring nut securely, and put the machine in service...If you have future issues with the nut getting loose
somewhere down the road then your question has merit....If no problem in say...5 years then it would indicate that your question lacks substance....Do the experiment to verify or discount the original engineering setup....
My vote is that your entire issue is the result of an improperly tightened ring nut , nothing more.....

You can't engineer for improper assembly or poor mechanical judgement....Machines are complex and the original builders assume that competent service is in play, that is the basis for making their design decisions.

If you leave the lug nuts loose on your car wheels, and they fall off...its not a design fault...You don,t need to add a an additional retaining ring to catch the wheel just in case it comes off the lugs...
You simply need to have tightened the nuts to begin with....

Cheers Ross
 








 
Back
Top