- Joined
- Feb 5, 2015
- Messages
- 662
(I posted an evaluation of an imported 6-jaw chuck a few days ago. I wasn't happy with the product, it didn't suit my purposes but the Chinese are capable of making very high quality metalworking equipment, as we know. I'd like to share the following experience as an example of this point. This was written several years ago, by the way.)
Recently, I decided to take a chance on another, different imported workholder, from a different supplier. I've dealt with the distributor before and would characterize them as follows: the products are fairly typical with a slight advantage in cost, depending on what you are buying, naturally. Many - but not all - distributors have a good return policy (my cynicism is showing: they HAVE to have a good return policy or we'd never buy anything from them, LOL, since the return rate is high).
This is the workholding device that I required for various needs, in the form of a packaged set:
The set consists of an ER-40 (similar to CAT-40) collet chuck, locking nut and spanner wrench plus a set of ER40 collets from 1/8 inch to 1 inch diameter in 1/16 inch increments. I received the shipment a couple of nights ago and finally got around to removing the individual parts from their sealed-in-oil plastic baggies.
The Chinese manufacturers have improved greatly over the past two decades in their manner of packaging and preservation, especially for small parts. (Machinery is a slightly different proposition, the use of a gummy protective coating, applied liberally to near-inaccessible areas is a source of frustration when trying to clean and lubricate a new machine.)
These parts cleaned up nicely in a plastic bucket of mineral spirits with a soft cloth, and they look great ! Except for the spanner wrench, the external surface of the locking nut and the large driving ring on the ER-40 chuck every surface is ground (all of these parts appear to be forgings, black-oxide coated, by the way). The fifteen collets are likewise ground on all surfaces except for the rear and the thin slots on each collet that allow the devices to collapse over a wide range. The number of slots varies, depending on collet size, from sixteen down to twelve on the very smallest collets for obvious reasons.
It is because of the many slots (and the fact that the collets are slotted from both the face and the rear) that these devices have come to be greatly appreciated for their ability to accommodate a wider range of diameters than any other collet device except for the steel-blade-in-rubber-bushing types like those found in tapping heads. Here is a close-up photo showing front and rear of typical collets:
Most popular workholding collets (e.g. 5C and R8) have three, sometimes four slots and the useful range is the nominal diameter minus perhaps .001 or .002. The reason is simple: larger workpieces "spring" the collet tapers outward and they no longer fit the spindle taper properly. Smaller workpieces achieve only line contact along the center of each individual "finger" of the collet rather than the desired area contact. The accuracy is slightly degraded by this condition but more importantly, the gripping power may be reduced and there is risk of damage to the collet if the drawbar is torqued excessively to provide more grip.
After looking around for a while, I formed the impression that ER collets seem to be more widely used when spindle horsepower is greater than around 2 HP. The various "ER" series of collets with their many more "fingers" to hold the workpiece, can maintain a tighter grip over a wider range of workpiece diameter I suppose. I also note that there are special "CAT-40 tightening fixtures" so that the collets are tightly secured. I didn't buy one of these, choosing a simpler method as shown at the end of this post.
Frankly, I'm no expert regarding these collets but I have read that they are capable of accommodating a diametral variation of 1/16 inch (imperial sets), which suggests that 1/16 increments should be able to grip any diameter between the minimum and maximum collet diameters, provided that the workpiece isn't tapered. Here is an intentionally exaggerated sketch of the two types of collet construction:
The diameter of the workpiece and the diameters of the collet "fingers" are identical in both sketches. Notice how much greater the gripping area of the second collet has become by doubling the number of slots and narrowing the fingers. And because the individual fingers are more flexible, they can accommodate greater deflection and therefore greater variation in workpiece diameters. (The actual ER-40 collets have even more slots thus greater flexibility and gripping power.)
After cleaning and wiping down the set, I installed the ER-40 chuck in my horizontal milling machine, snugged the drawbar and arbitrarily selected a 3/8 collet for my first "qualification" measurement. Placing a four-inch length of ground 3/8 shafting in the collet, I tightened the locking nut and measured the minimum/maximum runout. I wasn't at all pleased to see runout of .0030 inches when the booklet that came with the set stated that the factory specification is 0.0008 inches. I tried a 1/2 inch collet and saw similar results. My initial thoughts were, "Here we go again !"
I removed the chuck and cleaned the tapers of the spindle and chuck again, then I measured the spindle runout of the machine itself (although I knew it was true). No problem, less than 0.0003 inches (I was using a .0005 dial test indicator for these measurements since one was already parked on the mill table in a magnetic holder. (I'm usually reluctant to install the 0.0001 indicator, both from laziness and because I use it only when I must, due to the limited range of .005 travel.)
Repeating the runout measurements, I obtained substantially the same numbers so I got out the camera and started documenting the runout numbers, reluctantly conceding that this set had to be returned. After taking several photos, on a whim I removed the chuck and rotated it in the spindle 180 degrees.
BINGO ! The following are the minimum/maximum runout measurements (0.0005 inch graduations):
Maximum runout measurement (0.0005 inch graduations):
I was pleased with the 0.0006 inch runout, measured 1.6 inches away from the collet face, which corresponded nicely with the factory specification of 0.0008 at the same distance. Now I don't know if the collet chuck actually has to be oriented in this exact manner to obtain these results or if I simply dislodged a particle of dust when I turned the chuck around, LOL. Doesn't make a bit of difference to me … if it has to be mounted a certain way, I have no objection to doing that. After all, that is routine practice to obtain best performance from a 3-jaw chuck in a lathe. What could be simpler than punching a couple of marks on spindle and on chuck ?
So, that's it - after checking a couple of different collet diameters, I pronounce this experience completely satisfactory. I don't know if I got a particularly good set for my $175 or if the distributor received a particularly good lot - but I am sufficiently happy with the tooling that I promptly ordered both morse taper and R8 collet chucks from the same vendor (to accomodate the set of collets). This will allow the same set of collets to be used in other machinery.
The two tools on the left are new ones. The small Morse Taper/ER-40 fits the vertical head for my horizontal mill and the other is R8/ER-40 in the event I may want to use the ER collets in the other vertical mill. A further use for this versatile holding system follows from the fact that the Morse Taper in the horizontal mill vertical head is the same as the one in my smallest lathe’s tailstock.
The ER-40 collets in the lathe tailstock are handy for counterboring (using end mills) as well as holding drills, reamers, broaches and other tailstock tooling. Jacobs chucks are not particularly consistent in their runout over their adjustment range. That is not a problem with the ER-40 collet system, once the tailstock is properly aligned.
(An additional benefit is the ability to hold tooling that is too large to fit in a 1/2 or 5/8 drill chuck.) Here's a photo of a one inch ball end mill with 3/4 inch shank held in the tailstock for making a finishing pass on a ball-socket joint:
One reason for tooling the horizontal mill with ER-40 collets was the high cost and limited availability of suitable arbors. For best utilization of the machine, the one-size-fits-all arbor may not work well and probably explains why so many used arbors are bent (from using arbors that are too long for heavy work). Some of the tooling that came with my mill included a shell mill arbor and two full-length arbors (7/8 and one inch diameters).
Here's a photo of the shell mill with carbide inserts making a finishing cut on some HRS without coolant. The electronic flash did a good job of "freezing" spindle rotation and flying chips:
This photo shows the full-length arbor with slitting saws and various spacers installed:
I wanted the ability to install intermediate length arbors so that the end support could be "choked-up" on any cutter (or combination of cutters) for maximum rigidity. That led me to consider simple ways of making arbors of any particular length and diameter. Here's a sketch with a possible arbor configuration below a standard stub arbor.
Although spindle interfaces are standard, the end configuration isn't. The end can be threaded for a retaining nut, turned to suit a bronze end-support bearing, center-drilled for a hardened end-support center, a combination of several options and so forth. Unless you're lucky, any arbor - used or new - is going to require some work.
The lower part in the sketch illustrates a fairly simple arbor that can be made to any arbitrary length that suits a particular set of cutters. The material can be ground drill rod (or a semi-hard alloy of reasonable cost), I think three feet of ground one inch shafting costs about $25.
The following thread is the natural evolution of the above idea:
http://www.hobby-machinist.com/threads/a-very-simple-arbor-for-horizontal-mills.32778/
But the ER-40 system has a lot more going for it than this single reason; that's why I decided to go ahead and buy collet chucks for several other machines. One cautionary note: when these collet systems are used in high-torque machines (horizontal mills), some serious torque is required to prevent slippage. My primitive solution to this problem was brute force, welding 12 inches of black pipe to the wrench supplied with the collets.
Note that I added only one more set of collets for the various collet chucks ordered. Two sets of collets have been adequate for the four chucks so far.
Recently, I decided to take a chance on another, different imported workholder, from a different supplier. I've dealt with the distributor before and would characterize them as follows: the products are fairly typical with a slight advantage in cost, depending on what you are buying, naturally. Many - but not all - distributors have a good return policy (my cynicism is showing: they HAVE to have a good return policy or we'd never buy anything from them, LOL, since the return rate is high).
This is the workholding device that I required for various needs, in the form of a packaged set:
The set consists of an ER-40 (similar to CAT-40) collet chuck, locking nut and spanner wrench plus a set of ER40 collets from 1/8 inch to 1 inch diameter in 1/16 inch increments. I received the shipment a couple of nights ago and finally got around to removing the individual parts from their sealed-in-oil plastic baggies.
The Chinese manufacturers have improved greatly over the past two decades in their manner of packaging and preservation, especially for small parts. (Machinery is a slightly different proposition, the use of a gummy protective coating, applied liberally to near-inaccessible areas is a source of frustration when trying to clean and lubricate a new machine.)
These parts cleaned up nicely in a plastic bucket of mineral spirits with a soft cloth, and they look great ! Except for the spanner wrench, the external surface of the locking nut and the large driving ring on the ER-40 chuck every surface is ground (all of these parts appear to be forgings, black-oxide coated, by the way). The fifteen collets are likewise ground on all surfaces except for the rear and the thin slots on each collet that allow the devices to collapse over a wide range. The number of slots varies, depending on collet size, from sixteen down to twelve on the very smallest collets for obvious reasons.
It is because of the many slots (and the fact that the collets are slotted from both the face and the rear) that these devices have come to be greatly appreciated for their ability to accommodate a wider range of diameters than any other collet device except for the steel-blade-in-rubber-bushing types like those found in tapping heads. Here is a close-up photo showing front and rear of typical collets:
Most popular workholding collets (e.g. 5C and R8) have three, sometimes four slots and the useful range is the nominal diameter minus perhaps .001 or .002. The reason is simple: larger workpieces "spring" the collet tapers outward and they no longer fit the spindle taper properly. Smaller workpieces achieve only line contact along the center of each individual "finger" of the collet rather than the desired area contact. The accuracy is slightly degraded by this condition but more importantly, the gripping power may be reduced and there is risk of damage to the collet if the drawbar is torqued excessively to provide more grip.
After looking around for a while, I formed the impression that ER collets seem to be more widely used when spindle horsepower is greater than around 2 HP. The various "ER" series of collets with their many more "fingers" to hold the workpiece, can maintain a tighter grip over a wider range of workpiece diameter I suppose. I also note that there are special "CAT-40 tightening fixtures" so that the collets are tightly secured. I didn't buy one of these, choosing a simpler method as shown at the end of this post.
Frankly, I'm no expert regarding these collets but I have read that they are capable of accommodating a diametral variation of 1/16 inch (imperial sets), which suggests that 1/16 increments should be able to grip any diameter between the minimum and maximum collet diameters, provided that the workpiece isn't tapered. Here is an intentionally exaggerated sketch of the two types of collet construction:
The diameter of the workpiece and the diameters of the collet "fingers" are identical in both sketches. Notice how much greater the gripping area of the second collet has become by doubling the number of slots and narrowing the fingers. And because the individual fingers are more flexible, they can accommodate greater deflection and therefore greater variation in workpiece diameters. (The actual ER-40 collets have even more slots thus greater flexibility and gripping power.)
After cleaning and wiping down the set, I installed the ER-40 chuck in my horizontal milling machine, snugged the drawbar and arbitrarily selected a 3/8 collet for my first "qualification" measurement. Placing a four-inch length of ground 3/8 shafting in the collet, I tightened the locking nut and measured the minimum/maximum runout. I wasn't at all pleased to see runout of .0030 inches when the booklet that came with the set stated that the factory specification is 0.0008 inches. I tried a 1/2 inch collet and saw similar results. My initial thoughts were, "Here we go again !"
I removed the chuck and cleaned the tapers of the spindle and chuck again, then I measured the spindle runout of the machine itself (although I knew it was true). No problem, less than 0.0003 inches (I was using a .0005 dial test indicator for these measurements since one was already parked on the mill table in a magnetic holder. (I'm usually reluctant to install the 0.0001 indicator, both from laziness and because I use it only when I must, due to the limited range of .005 travel.)
Repeating the runout measurements, I obtained substantially the same numbers so I got out the camera and started documenting the runout numbers, reluctantly conceding that this set had to be returned. After taking several photos, on a whim I removed the chuck and rotated it in the spindle 180 degrees.
BINGO ! The following are the minimum/maximum runout measurements (0.0005 inch graduations):
Maximum runout measurement (0.0005 inch graduations):
I was pleased with the 0.0006 inch runout, measured 1.6 inches away from the collet face, which corresponded nicely with the factory specification of 0.0008 at the same distance. Now I don't know if the collet chuck actually has to be oriented in this exact manner to obtain these results or if I simply dislodged a particle of dust when I turned the chuck around, LOL. Doesn't make a bit of difference to me … if it has to be mounted a certain way, I have no objection to doing that. After all, that is routine practice to obtain best performance from a 3-jaw chuck in a lathe. What could be simpler than punching a couple of marks on spindle and on chuck ?
So, that's it - after checking a couple of different collet diameters, I pronounce this experience completely satisfactory. I don't know if I got a particularly good set for my $175 or if the distributor received a particularly good lot - but I am sufficiently happy with the tooling that I promptly ordered both morse taper and R8 collet chucks from the same vendor (to accomodate the set of collets). This will allow the same set of collets to be used in other machinery.
The two tools on the left are new ones. The small Morse Taper/ER-40 fits the vertical head for my horizontal mill and the other is R8/ER-40 in the event I may want to use the ER collets in the other vertical mill. A further use for this versatile holding system follows from the fact that the Morse Taper in the horizontal mill vertical head is the same as the one in my smallest lathe’s tailstock.
The ER-40 collets in the lathe tailstock are handy for counterboring (using end mills) as well as holding drills, reamers, broaches and other tailstock tooling. Jacobs chucks are not particularly consistent in their runout over their adjustment range. That is not a problem with the ER-40 collet system, once the tailstock is properly aligned.
(An additional benefit is the ability to hold tooling that is too large to fit in a 1/2 or 5/8 drill chuck.) Here's a photo of a one inch ball end mill with 3/4 inch shank held in the tailstock for making a finishing pass on a ball-socket joint:
One reason for tooling the horizontal mill with ER-40 collets was the high cost and limited availability of suitable arbors. For best utilization of the machine, the one-size-fits-all arbor may not work well and probably explains why so many used arbors are bent (from using arbors that are too long for heavy work). Some of the tooling that came with my mill included a shell mill arbor and two full-length arbors (7/8 and one inch diameters).
Here's a photo of the shell mill with carbide inserts making a finishing cut on some HRS without coolant. The electronic flash did a good job of "freezing" spindle rotation and flying chips:
This photo shows the full-length arbor with slitting saws and various spacers installed:
I wanted the ability to install intermediate length arbors so that the end support could be "choked-up" on any cutter (or combination of cutters) for maximum rigidity. That led me to consider simple ways of making arbors of any particular length and diameter. Here's a sketch with a possible arbor configuration below a standard stub arbor.
Although spindle interfaces are standard, the end configuration isn't. The end can be threaded for a retaining nut, turned to suit a bronze end-support bearing, center-drilled for a hardened end-support center, a combination of several options and so forth. Unless you're lucky, any arbor - used or new - is going to require some work.
The lower part in the sketch illustrates a fairly simple arbor that can be made to any arbitrary length that suits a particular set of cutters. The material can be ground drill rod (or a semi-hard alloy of reasonable cost), I think three feet of ground one inch shafting costs about $25.
The following thread is the natural evolution of the above idea:
http://www.hobby-machinist.com/threads/a-very-simple-arbor-for-horizontal-mills.32778/
But the ER-40 system has a lot more going for it than this single reason; that's why I decided to go ahead and buy collet chucks for several other machines. One cautionary note: when these collet systems are used in high-torque machines (horizontal mills), some serious torque is required to prevent slippage. My primitive solution to this problem was brute force, welding 12 inches of black pipe to the wrench supplied with the collets.
Note that I added only one more set of collets for the various collet chucks ordered. Two sets of collets have been adequate for the four chucks so far.
