ER40 vs. 5C collets accuracy in a lathe?

[frugalguido]

I just received one of those D1-4 ER40 chuck and sent it back. The measured TIR on it was, the worst at .0013" and the best at .0009" . The advertised TIR is .0008". I wish some one would produce a quality chuck that is a reasonable price, maybe someone like Bison.

 

[darkzero]

I think some people machine their own setup with a separate back plate, and leave the register a bit sloppy on purpose so they can use it as a "bump true" adjustable chuck for initial setup and for dialing it in with fussy work. That makes good sense to me...

Yup, that's what I did with my ER40 chuck for the lathe & the cheap 3-jaw that came with my lathe. I call it "tap-tru".

I don't trust anything that is direct mount cam lock, I always prefer a seperate backplate. Well except for my 4-jaw. It doesn't really matter cause it's independent, direct mount in this case reduces weight & overhang.

 

[mikey]

I just received one of those D1-4 ER40 chuck and sent it back. The measured TIR on it was, the worst at .0013" and the best at .0009" . The advertised TIR is .0008". I wish some one would produce a quality chuck that is a reasonable price, maybe someone like Bison.

If you are referring to the D1-4 ER-40 chuck from HHIP, I have no clue why their concentricity is so out of whack. I do wonder how folks are measuring concentricity and whether or not they are using quality collets and an accurate pin to measure run out. The nut that comes with the HHIP chuck is typical Chinese quality and the nut, all by itself, will introduce some run out. Chinese collets may run out with double or triple the run out of a quality collet. So, is it the user, the nut, the collet, the pin or is it actually the chuck?

My experience with the HHIP chuck was that it was an amazingly accurate chuck for about $70.00.

 

[frugalguido]

If you are referring to the D1-4 ER-40 chuck from HHIP, I have no clue why their concentricity is so out of whack. I do wonder how folks are measuring concentricity and whether or not they are using quality collets and an accurate pin to measure run out. The nut that comes with the HHIP chuck is typical Chinese quality and the nut, all by itself, will introduce some run out. Chinese collets may run out with double or triple the run out of a quality collet. So, is it the user, the nut, the collet, the pin or is it actually the chuck?

My experience with the HHIP chuck was that it was an amazingly accurate chuck for about $70.00.

I purchased the chuck after reading your favorable review, since I also have a Super 11 that I purchased new. I did question HHIP about the listed concentricity of .0002" before ordering, they said it should be listed as .0008" and they changed the listing to reflect this. I measured at the inside outer edge at ER register with no collet with a tenth indicator, just wouldn't meet their spec, ( which seems to me to be a little fluid and changeable ). In fact I did find the chuck listed in other parts of the world with a TIR of .002" with the same part number. Maybe they changed their supplier from when you purchased yours or your just one lucky guy. I also did try the measurment with a swiss manufactured collet, no change in the measurment as a test. It's to bad, cause I could really use a quality ER40 chuck for the Super 11. I did try to get them too send me a replacement,(maybe just got a bad one) but they said that their out of stock now, even though they show stock on the Ebay listing, so who knows.

 

[mksj]

I ordered the same ER-40 with a direct mount D1-4 from HHIP about a year ago, mentioned a TIR of 0.0002" at the time, but this seems to change all the time. Mounted it an did a slew of measurements including indicating the chuck collet seat. It was out over 0.002", so off by a factor of ten and had even more skew with a ground rod. I used a Maritool bearing nut and high quality collets, still out 0.002-0.003. I think it is hit or miss with these ER-40 chucks. I spoke to HHIP, they say they are just an importer, they have nothing to do with the inspection or quality of what comes in, not will they check anything. Other ER-40 direct D1-4 mount chucks all seem to be the same Chinese manufacturer. Tormach has one that mounts to a back plate and looks a bit different, but they have no information or specifications on there website. Pretty lousy pickens. I will probably go with a back plate version and mount it to a D1-4 back plate so I can tweak the TIR adjustment. Have looked at other ER-40 options for the lathe like mounting a collet block in my chuck, but not impressed with what is available. My 5C does the job 99% of the time, there are a few occasions where I would like more holding power from an ER collet. I mostly use the ER system for holding end mills either in the lathe tailstock or an R-8 holder on the mill.

Many individuals claim the advantage of the ER collets is greater range/compressibility, relative to 5C collets this is true, but I haven often had problems with ER collets collets not compressing down uniformly when at close to the edge of its specified range (+/- 0.5mm). So although it is claimed that a 1 mm increment set will work with both metric and English sizes, I opted for a 1/32nd increment ER-40 set along with a 10 and 12mm high precision collets for my edge finders.

The 5C racks are slanted and work fine in a horizontal or ventricle format. I have often have seen machine shops with wall racks for all the 5C collets, but also some nifty sliding draw racks.

 

[mikey]

Thanks, guys. It appears then that we are dealing with the typical Chinese variations in quality. I think I must have just lucked out and gotten a good one. My apologies if my review mislead anyone. I was being honest but it appears that things have changed.

I actually have one of the Tormach ER-40 chucks. Tormach told me it is hardened. It, too, is Chinese and meant to mount to a rotary table. It can be adapted to a back plate, the accuracy of which depends on the end user. I have the back plate for it but ran into the HHIP chuck and never mounted it.

 

[whitmore]

Cool definately ! Usefull in home shop Maybe.
My current project involves making 1/2 doz identical short pieces made from 30mm bar. A lever action collet closer ...
Ron

The clear bore of 5C collets is one inch (there are larger grips, to 1-1/8 inch, but those
are really step collets, the work bottoms in them). So, 30mm won't fit through. You can get
short grip on large items like that, with big-faced premade or blank collets, though.

 

[ronboult]

BSS1 raised an interesting idea in post #19 when he suggested getting collet blocks for use in the mill vice.
I suddenly remembered that I actually have square and hexagonal through ER32 collet blocks that I use on my mill.
Why not mount the square ER32 collet block in the 4 jaw, adjust for minimal runout and use to mount longer stock through the spindle on my lathe.
Anybody need a surplus MT3 ER32 collet chuck.
Ron

 

[frugalguido]

I just wish a quality manufacture would make one. There would seem there is a market, it's not a complicated item, not like a scroll chuck . I did take some basic measurements of the chuck before I sent it back, so I might make one and send it out for finish grinding. I don't think the HHIP one is even hardened, because there was some dings on it.

 

[Bob Korves]

A collet chuck is not inherently an accurate plug and play setup, because whatever accuracy it has can be compromised by multiple issues outside of the collet chuck itself. The spindle, the headstock bearings, the spindle register, the alignment of the headstock, etc. The good news is if you get the spindle turning without radial and thrust bearing looseness and in line with the lathe bed, regardless of spindle runout, a collet chuck can be accurate to however tight a tolerance you want to chase it, regardless of the accuracy of each component individually, IF it has a separate spindle mount. If anyone thinks that a stack up of tooling, none of which has been carefully calibrated individually, should make the final tool in the stack accurate, they are kidding themselves or do not understand how tolerance stacking works.