HELP! Newbie Having issues with ER32 tool pullout.

Hi all,
First post here :)
I am a newbie to cnc Machining. We recently bought a machine to machine Aluminium moulds to make composite parts (Mostly small wings). Using 5mm diameter Ball nosed endmills at about 17k rpm we get great surface finishes. The problem is that during finishing passes (removing 0.2 to 0.4mm of material) the tool slowly pulls out of the ER32 collet and the cuts become deeper by the time it reaches the end of the part.
I assumed that it would be impossible for the tool to pull out with such little loads? Are we doing something wrong or do we just need Hydraulic chucks?
Does anyone have similar experiences?
Thanks a lot.

This suggests to me exactly what you are thinking that the collet in your chuck is not gripping the tool tight enough. Have you measured the length of the tool outside the chuck and confirmed that it is actually pulling out.

Steve.

Hi Gian - Make sure everything is clean and fits correctly and the collet is the right size for the bit. Then tighten it up really tight. Also make sure it is engaged correctly. If it is not in far enough it will loosen. Peter

Yes, The tool is pulling out, measuring it using tool length sensor and the change in tool length is consistent with the error in the actual part. My main issue is understanding the reason the tool is pulling out. The loads should be pretty low since its only a finishing pass. It seems impossible that I have to spend 600 bucks for a hydraulic chuck for such a light operation. :(
I guess my next step is to try some high precision collets that only fit one size. The ones I have now have a range of 1mm
Edit: we are already using high precision collets :(

...try adding a plug/spacer in the bottom part of the collet ..(what ever size it is) so you compress the collet evenly at both ends of the taper.

Attachment 449608

I would switch to a smaller ER series like ER 16 or 11. At 17000 rpm you have a lot of mass spinning around a comparatively tiny tool. The near solid ER32 collet is a lot of mass dealing with a lot of centrifugal force at that rpm. Try a nice clean and dry ER 16 and see if that doesn't help. Look in the books for the proper number and tighten with a torque wrench. BTW - proper ER32 is 100lbs.

Quote:

---

Originally Posted by GianFerr

Hi all,
First post here :)
I am a newbie to cnc Machining. We recently bought a machine to machine Aluminium moulds to make composite parts (Mostly small wings). Using 5mm diameter Ball nosed endmills at about 17k rpm we get great surface finishes. The problem is that during finishing passes (removing 0.2 to 0.4mm of material) the tool slowly pulls out of the ER32 collet and the cuts become deeper by the time it reaches the end of the part.
I assumed that it would be impossible for the tool to pull out with such little loads? Are we doing something wrong or do we just need Hydraulic chucks?
Does anyone have similar experiences?
Thanks a lot.

---

ER32 for a 5mm cutter is crazy, you should be using no more than a Quality ER16 and torqued to the correct torque required then you won't have any problems, or better still a shrink fit tool holder no collet at all

While the ER32 if a quality tool holder should work, if it is torqued to the correct spec's, the size of the ER tool holder the speed and centrifugal force is most likely releasing the collet grip on the tool

.4mm is a loaded cut for a bullnose cutter .1 would be considered a light cut

ER 16 slotted 42 ft./lbs
ER 16 hex 42 ft./lbs

ER 32 slotted 100 ft./lbs

hy gian :) go with bearing nuts so to avoid useles torque on the collet ( that leads to micro-damage and less clamping force ), and be sure that the tool shank that goes inside the collet is blank ( thus not coated ), cilindrical ( no flats ), and mount it dry ( not only clean, but also dry), and use a H7 collet ( designed for a h7 clamping range )

17k rpm : is there a balancing machine ? and what is the maximum rpm of your spindle ? you know, so to avoid killing your spindle at max rpm, with an unbalanced toolholder :)

like other said, a toolholder with a smaller collet may help ... if it fails, and you wish to buy a hydra holder, it is no guarantee that it will work; consider shrink-fit ( definetly lighter than a hydra ) or weldon ( cheaper, possibly the screws will fly out, so craft another ones with small play, and adjust your cutting specs acordingly, because they deliver a higher tool run out, and that means that your final mould surface won't be as nice as when machining it with a tool with less run-out, but it may still be ok/salable )

or it may be that your tools are poor quality : if possible, change them often / kindly :)

Another tip I'd add if you go with the ER16 collet route with the bearing nut, is to use a 25mm deep socket attached to a torque wrench (if its a short stick out tool) so you could eliminate the variable of torque during experimental runs to investigate the issue. Cut a long slot and measure opposite ends for depth to see if it's still pulling out.

Quote:

---

Originally Posted by deadlykitten

hy gian :) go with bearing nuts so to avoid useles torque on the collet ( that leads to micro-damage and less clamping force ), and be sure that the tool shank that goes inside the collet is blank ( thus not coated ), cilindrical ( no flats ), and mount it dry ( not only clean, but also dry), and use a H7 collet ( designed for a h7 clamping range )

17k rpm : is there a balancing machine ? and what is the maximum rpm of your spindle ? you know, so to avoid killing your spindle at max rpm, with an unbalanced toolholder :)

like other said, a toolholder with a smaller collet may help ... if it fails, and you wish to buy a hydra holder, it is no guarantee that it will work; consider shrink-fit ( definetly lighter than a hydra ) or weldon ( cheaper, possibly the screws will fly out, so craft another ones with small play, and adjust your cutting specs acordingly, because they deliver a higher tool run out, and that means that your final mould surface won't be as nice as when machining it with a tool with less run-out, but it may still be ok/salable )

or it may be that your tools are poor quality : if possible, change them often / kindly :)

---

You have no clue the ER Bearing Nuts still need the same torque as a regular ER Nut, yes they work better and have a smoother lock up on the collet, so stop the Bs, they still need the same Torque to lock them up

Nothing changes only a smoother lock up action with a ER Bearing Nut, there are 2 types of ER Bearing Nuts, Friction Nuts has hardened inserts which slide and the other has hardened inserts that work like a bearing

The ones that last and work the best are the Hardened friction thrust washers, the bearing type fail after a short while of use, both types provide a higher lock up torque than a standard nut which should not be used for highspeed machining

Quote:

---

Originally Posted by smartcolombyan

Another tip I'd add if you go with the ER16 collet route with the bearing nut, is to use a 25mm deep socket attached to a torque wrench (if its a short stick out tool) so you could eliminate the variable of torque during experimental runs to investigate the issue. Cut a long slot and measure opposite ends for depth to see if it's still pulling out.

---

There is a set torque that you need to use for all collet types and size failing to use the correct torque will end in failure

For High speed machining the ER Nut should have smooth out side which use's a special one way clutch wrench to lock them up this snip is from one of the many manufactures that manufacture highspeed ER tooling

Quote:

---

Originally Posted by GianFerr

I guess my next step is to try some high precision collets that only fit one size. The ones I have now have a range of 1mm

---

What collet size you use? Internal diameter, I mean. I hope you are not using a 6mm one for a 5mm bit.
I hold even 8mm bits with ER16 without problem but each bit is hold by same diameter collet.

Maybe it's not the case but I noticed your quoted text about 1mm range. My 2 cents...

hy mactec, pls, be polite :) i was not refering to the torque required to tighten the nut, nor to that a bearing nut requires more/less torque to deliver same clamping force; in the end, each collet provider has it's own torque tables, that acts like a general guide line

i was only scratching the surface of that a normal nut, in time, leads to collet and chuck damage, because it has a tendency to rotate the collet, while a bearing type won't change collet 'phase'; thus, a bearing type has a higher chance to transform tightening torque into clamping force, since there is less friction

also, a low quality bearing nut won't behave as nice as a normal one crafted with more attention to thread tolerances ( including the thread on the toolholder )

whatever / kindly :)