ER40 versus 5C Collets - Considerations and Trade-offs

[mksj]

ER collets for the most part should have stock/end mills through the full length of the collet to prevent it from distorting when clamped. It is hard to hold short stock in an ER collet, and also if passing bolts through the collet ER collets they tend to be too long for smaller hardware. With a 5C collet I have held stock as short as 0.1" in depth, and also use it frequently for holding threaded items. I use 5C collets primarily, and reserve my ER collets on the lathe for metric/odd size stock. Due to the short clamping range of 5C collets, a 1/64th increment set is often required to hold a full range of stock. This is adds the expense vs. an ER collet set.

I use a speed handle for my 5C chuck which makes for quicker collet changes, there is also speed chucks as well as a 5C drawbar which would be quicker than using an ER collet chuck. I primarily use my ER collets for end mills on the mill and tailstock of the lathe.

5C one can screw in a stop system to the back of the collet, but they are not repeatable for accurate work vs. a through the spindle stop system that can be used with either type of collet system. My ER lathe chuck does have a rear screw in stop system.

5C collet holding small part, 0.2" depth is being held by the collet.

 

[Larry$]

I got an 5C "Set-Tru" type chuck when I bought my lathe from PM in 2016. There are no markings indicating country of origin. It is nicely made and runs as true as I can measure. I bought a set of 5C collets with the chuck, in 1/32s & have added some by 1/64ths and some Hex & squares, & an internal stop. I've never had an issue with parts slipping. I've held small parts by only 1/4" in the collet. I polish small parts right up to the collet with no fear of hurting my fingers. I've got 5C's - Spin indexer & collet blocks.
I do have an ER40 chuck & R8 to ER40, but will only use them when I can't use the 5Cs, like holding metric end mills or shafts.

 

[mcdanlj]

Oh, another example! You can get expanding 5C collets in a variety of sizes. They are generaly machineable, so you can turn them to a desired size for a fixture where you need to hold a part by the bore.

 

[pontiac428]

...But ER collets can hold *tools!* The whole world of manufacturing runs on expanding collets, they're great!

I do use 5c for collet blocks and for my Sheckel grinder, but I've invested minimally in them. ER on the other hand I have in inch and metric and have invested in good holders for.

Workholding in the lathe, as I said at the beginning of the thread, I'm all about flex collets. The Jacobs is a great system. If your work is short, plug the back just like with ER- no big deal.

 

[Parlo]

I'm not sure why this myth still exists either. If you read the document I posted at the beginning of this thread you will see this statement about 5C clamping range/length:

View attachment 478595

Thanks for the info it explains the workings well. The only wording I am not sure about is.
"compress against the part being held, first at the nose".
I can't see how this is universally true.

If a collet at rest is naturally sprung larger than a nominal rod, then the contact in the bore is at the rear first when the rod is inserted at rest. The tapers will engage at the tip first until the bore is parallel and the tapers are aligned. Clamping force then closes square to the spindle axis.

If a collet at rest is naturally sprung smaller than a nominal rod the rod will open the jaws until they are parallel, the tapers will engage along their length and clamp in a direction square to the spindle axis.

Perhaps the quote helps to propogate the myth?

 

[Larry$]

If a collet at rest is naturally sprung smaller than a nominal rod

If sprug smaller than the work, you will hate the damned thing trying to get the work in!

 

[davidpbest]

Thanks for the info it explains the workings well. The only wording I am not sure about is.
"compress against the part being held, first at the nose".
I can't see how this is universally true.

If a collet at rest is naturally sprung larger than a nominal rod, then the contact in the bore is at the rear first when the rod is inserted at rest. The tapers will engage at the tip first until the bore is parallel and the tapers are aligned. Clamping force then closes square to the spindle axis.

If a collet at rest is naturally sprung smaller than a nominal rod the rod will open the jaws until they are parallel, the tapers will engage along their length and clamp in a direction square to the spindle axis.

Perhaps the quote helps to propogate the myth?

Both conditions you bring up represent a damaged or distorted 5C collet.

 

[Parlo]

Both conditions you bring up represent a damaged or distorted 5C collet.

Yes deliberately to try and discover any situation where a collet would only grip at the tip on a nominal rod.

 

[Jake P]

I haven't read all the post here, but will throw my .02 in non the less.

I started out with an ER40 collet chuck for my 1440GT. I purchased the chuck from PM, the collets all came with my used mill ( a full metric set).

About a year later I added a 5C collet chuck, again from PM as well as the full set of collets from them as well.

My reason for adding the 5C setup was holding short work pieces without having to resort to the technique posted somewhere above using a matching sized diameter piece deeper in the ER collet. And for that purpose it is the clear winner for me. As well as being able to incorporate a depth stop in the 5C. I find my self using the 5C now almost every time over the ER if the work piece diameter is within the tolerance of the 5C collet system. The ER collets are more tolerant of diameter variance than the 5C.

The reason I prefer the 5C if usable is simply convenience. I use an impact wrench to speed up the installation and removal of 5C collets, making the process very fast and easy. But I do not tighten the collet in this manner, for that I always use the chuck wrench. Tightening and removing the ER collets is just more time consuming and not as easy, for me anyway.

As stated, both of my chucks I purchased from PM and both are easily holding and repeatable for me within .0005". Perhaps I just got lucky and got two accurate chucks while others have not had such luck. But I am very pleased with both chucks as well as the PM 5C collets. The only issue I have with the collets from PM is that the sizes are not engraved on the face of the collet and if you are sanding/polishing close to the collet face it is easy to remove the etching.

 

[Joe Pie]

I bought a 5C collet chuck for two reasons. I already had a 5C collet set, albeit only by 1/16ths, and I liked the slim nose on the 5C chuck which allowed me greater flexibility for certain operations.

A down side is the lack of grip range of a 5C collet which will require minimally a 1/32" set of collets and preferably a 1/64" set. The cost of a full set of 5C collets would exceed the cost of a set of ER32 or ER 40 collets. Changing out collets on my collet chuck also requires a full 32 turns of the chuck key, which can be annoying.

I disagree with the idea that 5C collets only grip the work at the nose of the collet. Were that true, the work would be unstable in the chuck. When a 5C collet is closed, it first contacts the the work at three points ,H in the left drawing, at the nose of the collet, as the collet jaws bend in a simple arc in region B in the upper right drawing. The collet also makes contact with the 5C socket at point D with a slight gap at E. The work is only in contact at C in the drawing.
View attachment 441662

As the chuck is further tightened, the simple arc in region B changes to a S shape with bends at F and G as shown in the lower right drawing, closing the gap E. Fully tightened, the collet will be in contact with the socket all along the surface between D and E and the region A will again be parallel to the collet axis, gripping the work along three axial lines along H and stabilizing it.

The limit governing the ability to grip an undersized cylinder will be determined by the three gaps in the collet. The amount of force required to close a collet will increase as the difference in size between the collet and the work increases. In the case of using a lever type collet closer, the available force may very well be insufficient to that required to fully close on the work, resulting in a three point grip on the work rather than three lines.

Only if the workpiece being held is undersized from the nominal ground size of the collet will the collet have a tendency to only grip to the front. In your argument, it appears you may have overlooked some physical facts, so lets go over them. First, a true matching conical taper would only be possible if 2 identical diameters, at exactly the same distance apart are present on both the male and female components. This is an inarguable physical fact. Second..only at one point along a linear axis is a 5C collet truly tangent to the machine spindle nose. Perfectly ground and perfectly positioned, the machine nose and collet tapered surfaces are surface matched. Physical fact. Third...as the collet migrates into the spindle nose to accommodate an undersized workpiece, the relationships between the taper geometries immediately translates to the 6 split line edges of the collet simply because the larger front diameters are now moving backwards into the smaller spindle nose internal diameters. This is also a physical fact. Lets address your "region B " curve. From a material analysis standpoint, multiple things ( times 3 locations ) are happening if this is true. The curved cross sectional profile of the collet body would have to stretch and compress in numerous locations to bend as you suggest. Saw a piece of pipe at a 120 degree section and try to bend it as you say. Watch the edges pucker, inner radiused areas wrinkle and outer radiused surfaces stretch. I completely and respectfully disagree with your conclusion.