Advantages of Morse collets is that they take up less headroom than an ER chuck, and are possibly slightly more rigid due to less “stick-out”. Having said that, I rarely use mine compared to my ER chuck which is quicker to change cutters.

MT3/ER32 is the way to go for the reasons stated above. Noel.

I have both Morse collets and ER collets, the ER collets gets used far more than the Morse collets because they can hold a range of sizes. I also use the ER collets for work-holding in my lathe. Only when extra headroom is needed I use one of the (few) Morse collets I have.

Thor

I would say use an ER32 collet for most things. A useful addition though is to get a Ø16 MT collet and a Ø16 parallel shank ER16 collet holder (goes up to Ø10 for smaller tooling). This can be used to get into smaller areas that the ER32 will struggle with and also gives the option of having the parallel shank sticking out different distances to suit what you are doing. If you get two ER16 collet holders you can quickly switch between a spotting drill and a jobbing drill for example or a pilot drill and a final size drill. You can also use it for extra headroom compared to the larger ER collet holder.

Martin C

Morse collets are only split at one end and the piece being held must be a close fit with the collet.

ER collets are split at both ends and the piece being held is clamped over the full length of the collet. The piece being held does not have to be a close fit.   eg. ER32 collets have a a clamping range of 1mm.  ER collets protrude from the end of the spindle and reduce head room.

The fundamental difference is that a morse taper is self holding with a long shallow taper inevitably generating high frictional losses as the collet comes up tight. The mechanical advantage of the shallow taper helps offset the frictional loss but ultimately a morse taper collet will not hold as tightly as purpose designed tool holding collets at any given drawbar torque.

If you do jam a MT collet up really, really tight they can go well and truly into self holding mode becoming a total pain to remove. As MT spindles are generally associated with the lighter breed of milling machine going postal on the drawbar for removal is probably not a great idea.

A screw thread type removal device is vastly preferable to bonking the drawbar on an MT taper machine. I’d be very unhappy about using a machine not so equipped. The sharp rap technique is getting perilously close to the demarkation between “not ideal but if works given sensible care” and “to be avoided as its likely to damage things in the event of error.”

An oft overlooked issue with ER collets is that, being double ended with a floating nut, they require high tightening torques in larger sizes to achieve proper grip and the specified concentricity.  This table from Shars is pretty good.

https://www.shars.com/media/manuals/clamping-torque.pdf?srsltid=AfmBOopd_5Tbe-Un9xdGU-zRQAbkWr8zamsT7g_ENOvWaKPwHOVVb1Xf

Rego-fix basically invented and productionised the ER collet so they should know but their torque table is very much TMI

https://us.rego-fix.com/system/files/docs/2017-09/RFAG_TD_Anzugsdrehmoment_EN_0.pdf

At the core the differences are fundamentally down to the length of the gripping portion which is shorter for the smaller collets in the range. Oversimplifying to the point off absurdity grip is fundamentally down to force pre unit area over which the collet touches the tool shank. Smaller tool doesn’t cut so hard so it needs less holding force and has less grip area anyway so less torque to get to loading force.

With ER25 and 35 collets you will be struggling to get the required torque on larger sizes with the usual bent metal spanner thingy in the common kits. It’s surprising that the bearing type nuts actually need more torque in some sizes than standard ones as they are generally sold as needing less torque overall. I’ve no idea why less torque is specified for mini nuts.

Much, if not all, of the gripping on smaller tools is done under the nut portion of the ER collet system which has implications for concentricity. For obvious technical reasons the nut has to be made with sufficient clearance to float slightly as it pushes the tail end of the collet firmly into the taper in the holder body so things run concentrically. If the collet is not really accurately made it can take considerable force to generate concentricity. Theoretically it’s all self correcting at sensible torque levels but if the collet is not sufficiently accurate it may never run really true. Expensive collets are expensive for reason!

There is no doubt that for general purpose moderate duties on smaller cutters the R8 taper is the best of the commonly found bunch. Being single ended with a relatively steep closing angle frictional forces are lower yet mechanical advantage is sufficient to get more than enough holding power at 25 to 30 ft lb torque. There is a reason why about a zillion other designs of collets for more specialised machines follow this general layout.

Clive

Simply put, there’s absolutely no single work or tool holding method invented yet that works for everything Sonic. The real answer for what might be the best choice boils down to what are you wanting or needing to do with those collets, the machine size, type and maybe even brand of mill your using. With a larger industrial level mill, I might be more tempted to use MT collets just for there increase in rigidity. Ignoring the actual design differences between your MT 3 spindle and my R8, there still fairly similar in holding power, maximum tool shank size etc. My first mill could only use ER 25 sized collets for holding items like end mills, so I got real familiar with using them. Instead of choosing R8 collets after buying my last mill, I instead decided on ER 40’s even though the maximum tool shank size that collet series can use is slightly more than 1″ in diameter. Which in reality is above what any R8 taper milling machine can ever properly utilize. But I also wanted dual usage from them as a work holding collet on my lathe.

I also understood that no collet design can do that “everything” as I said. So I’ve still had to buy quite a bit of R8 shank tooling. Keyed and key less drill chucks, R8 end mill holders for use with Weldon shank end mills, boring heads etc. Between those ER collets and anything else I’ve bought with those R8 tapers on them, there’s not much I can think of where the proper R8 collets would make much or maybe any noticeable difference other than possibly gaining me a bit more clearance in the Z axis. And so far that’s yet to ever be an issue. Your own mill and what your usually using it for might well have some very good reasons for choosing those MT 3 collets instead. Only you can really make the decision for what might be the best overall choice.

Depending on the lack of or availability of a screw adjustable ejection design for those Morse Taper collets. An ER collet chuck and collets should drastically reduce the number of times you might be hamming on the end of the drawbar. And that’s also the main reason I was adamant about buying my next mill with a R8 taper even while owning a fair amount of MT 3 tooling from my previous mill. One further item to watch for if you do choose to use those ER collets. Unless your mill does have a proper spindle brake or lock, make 100% sure the collet chuck body your buying also has at least two wrench flats, depending on how much your willing to spend, not all of them do.

It looks wrong. Logic tells me that my favorite cutter shanks will all fit perfectly in MT3 taper collets and save me the 60mm overhang I’d get with an ER32 chuck. Our American chums are forever using R8 collets to hold their Imperial shanks nil problemo, their alignment depending on one taper where I have two. Unfortunately the bee in my bonnet says that parallel shank tooling always goes in a non-retaining taper.

I have an MT3 3/4″ collet with two chucks waiting to be held. An ATC-ER32 and an ATC-ER25. If only I could convince myself that it would work 🙂

Robin

I wouldn’t exactly say there’s zero issues with those R8 or even MT collets since it’s been proven they hold any tool shank a bit less secure than ER collets might. That’s simply due to there design since any R8 or MT collet grips a tool shank by collapsing the collet and gripping a tool shank from only one end of the collet itself. MT collets are probably superior in there total grip over the R8’s though. Yes those R8, MT collets can and still do work or otherwise they would have never have existed for as long as they have. ER’s hold by collapsing the collet at each end. However even they may not do exactly that for tool shanks not inserted far enough into the collet or at the smaller diameters in the larger ER collet series that may be counter bored at the rear and shortening the total length they grip any tool shank. In that case, the torque the cutting tool actually sees or could ever withstand without breaking is very little in comparison to the maximum tool shank sizes each ER collet series can accept.

For almost any model engineering most of us would be doing, without question that extra length an ER collet chuck adds is I think greatly overblown in it’s real world importance as far as cutting tool rigidity. Only a fool would think that’s totally unimportant since anything you can do to increase rigidity can only help, but for what we might be doing, it’s just not nearly as important as some seem to assume. For most of us, gaining the extra Z axis clearance between the bottom of the spindle and the work piece using those R8 or MT collets, and maybe a reduction in possible non concentricity might be there most benefits. Logically most mills each of us might be using including my own BP clone have more built in flexibility in how the heads are mounted or a columns overall rigidity than any real issues about tool extension from there spindle bearings. Yes there using far larger and multiple times more rigid milling machine designs, but if that few inches of extra distance ER collet chucks or even the shrink fit tool holders add was that important, then industry wouldn’t be using them for production.

Any ER collets, the chuck itself and the closing nut needs to be viewed as a system that all have to be made correctly to work as designed. Increasing the tightening torque even when using those ball bearing nuts can’t and certainly won’t ever correct any concentricity errors unless the original nut was already incorrect to begin with. There are various types of nuts designed to reduce the frictional losses, but all that really does is transfer the energy to then increase the collets grip. This Rego Fix video mentions some of the various ER collet nut design types and says literally nothing about any of them either increasing or decreasing concentricity at all. https://www.youtube.com/watch?v=3sLSSJhxnak They also have many other videos about ER collets, chucks and nuts that in my opinion are well worth going through just for a proper education. For something that appears fairly simple, there’s a lot more to the design than it might seem.

For your mill and how you plan to use it, then I’d suggest that only yourself can make a real decision on what the best tool holding collet type might be. But today, I think your going to run into the same issues I have with my R8 spindle. Finding MT collets that are very well made with a guaranteed low run out for less than Hardinge or Schaublin prices.

Yes Simon, that makes perfect sense.  1mm closure for, say a 16mm collet, is a lot less stressful than 1mm for a 3mm collet.

Yes bernard, I’m sure you’re right, and that’s what I might expect, but that’s not what gets generally bandied about.  When 1mm closure is mentioned, I’ve not seen reference to the smaller collet sizes.

It is not just the ER size that determines how much you can close them down, In all the ER sizes the very small ones only have a 0.5mm (not 0.05mm) closing range. That is why I have half sizes even in ER32.