ER40 collet chuck
- Thread starter Brian H
- Start date
I'm surprised we don't see lathes with electronic lead screw automation as a standard feature. There are some but $20K+. These kinds of mechanical setups with different gear trains, levers, etc. are not really needed. Reminds me I need to finish my Rocketronics ELS install. I've been distracted with projects 1 through 41.
We are in polar opposite camps John. I'd much rather a solid 127 gear transposition, a variety of gears to cover the vast majority of threads, and the ability to cut any gear set I might need to get the exact pitch I want that might not come standard. Nothing about this problem reduces the pure love I feel for my good old fashioned purely mechanical lathe.
My South Bend has gear levers for all the common threads. It's the older single lever version. I while back (8 years maybe?) I needed to cut some pipe threads for my vacuum tank (for degassing epoxy) and interestingly there was no lever position for that pipe size. Now with the mechanical taper attachment it was easy enough to set the taper for the pipe thread. And then key the pitch into the ELS and cut a pipe thread. No messing with gears. No questions to South Bend group on what gears to buy and then wait for them to arrive. No running to Princess Auto to buy a pipe threading kit.
Five minutes later I had the pipe threaded and went on with the rest of project #42. It really doesn't matter what type of ELS is installed. I have not yet added power cross feed to the South Bend even though it would make threading then as easy as it is on the Gingery. And depending on how one even connects the hardware the lathe can (as in my case) remain 100% mechanical.
But once you've had the carriage and tool stop withing 0.01" of the chuck every single time while turning to a diameter or boring a blind hole, the idea of hanging over the half nut leaver or power feed engagement knob you never go back. I don't know of anyone who buys a motor cycle with an electric start and a kick start pedal and rarely uses (if ever) the electric start because the kick start is so much more satisfying.
Or someone who has an gas stove with Piezo ignition but prefers to light the burner each time with a match because the Piezo ignition method is new technology and an old fashioned stove is just so much more satisfying to light with a match.
Or someone who has an gas stove with Piezo ignition but prefers to light the burner each time with a match because the Piezo ignition method is new technology and an old fashioned stove is just so much more satisfying to light with a match.
We can have lots and lots of fun with this. And I have no doubt that the vast majority will be in your camp. But I am not.
For me, it's sorta like dancing with a robot, or driving an auto transmission, or washing dishes in the dishwasher, or cooking in the microwave, or a gas fire on the patio, or a gas fireplace instead of wood, or a blow up mistress, or or or
After a career of electronics, and automation, I prefer manual. Remember, it's not about saving time for me. It's about enjoying every second of what time I have left. How much I get done simply doesn't matter.
I'd never be in a hurry to fix something at this point in my life. If it took a little longer, it would just be that much more enjoyable. In fact, it's about finding something else to fix so I can spend more time doing it. In many ways, it's not about wishing the storm would end, it's about enjoying the dance in the rain.
Penmanship penmanship! And cursive writing! Everyone should know how to trim a feather into a good quill! Just yanking your chain Susquatch. After a long career of programming I stopped. Could not stand it. Did other work. After about 15 years I finally feel like I could do some programming again. I can see not wanting to even think about automation after a career in it!
Brian H
Super User
Threading update:
I measured a selection of threads that are divisible by 8TPI and here are the results of what I measured
8TPI- 1.035"
16TPI- 1.022 "
24TPI- .966"
32TPI- 1.054"
40TPI- .994"
48TPI- .902"
56TPI- 1.024"
I'm not sure what can be extrapolated by these measurements, all I know is none of them are exactly 1". Is this something I could reach out to King for suggestions for correction? I don't even know if or what could be changed.
I know I am very disappointed by this since one of the reasons I purchased a new lathe was the ability to do single point threading...
I measured a selection of threads that are divisible by 8TPI and here are the results of what I measured
8TPI- 1.035"
16TPI- 1.022 "
24TPI- .966"
32TPI- 1.054"
40TPI- .994"
48TPI- .902"
56TPI- 1.024"
I'm not sure what can be extrapolated by these measurements, all I know is none of them are exactly 1". Is this something I could reach out to King for suggestions for correction? I don't even know if or what could be changed.
I know I am very disappointed by this since one of the reasons I purchased a new lathe was the ability to do single point threading...
May I ask for a clarification?
From what you wrote, I assume you set a given tpi using the gears and levers on your lathe. Then you cut threads at 8, 16, 24, 32, 40, & 48 tpi. And lastly you measure the span of those threads to see how many inches they were, and got the numbers listed.
Is this what you did?
The variation seems to fall both above and below the correct length.
If the lead screw was wrong, I would expect everything to either high or low, not both.
It also starts high then goes down and down then high, down down again. However the relationship between series is not consistent.
It suggests to me that something is setting the ratio off by a factor.
Perhaps the 127 gear is engaged when it shouldn't be?
Perhaps the feed is engaged when it shouldn't be?
Do you have the 40 tooth gear on the top shaft, driving the 86 tooth idler gear, the 86 tooth idler gear driving also another 40 tooth gear on the bottom shaft? This is from the page with figures 15,16,17 on it. Hope its the same manual you have. I noticed the manual pictures don't show the gear box selectors quite in line with the proper selector positions on the plaque, maybe a problem?
If you were closer, but that is as it is.
If you were closer, but that is as it is.
Last edited:
Brian H
Super User
@Susquatch , the process I went through was to set the levers to the thread pitch, engage the threading lever, rotated the chuck by hand to a set of wirness marks (to take out any backlash) zero the indicator, then rotate the number of threads and measure the travel of the carriage.
I also noticed the inconsistencies with the measurements.
@Bandit , I will confirm this after work tonight but yes, I put the gears back as they were before I tried metric threads
I also noticed the inconsistencies with the measurements.
@Bandit , I will confirm this after work tonight but yes, I put the gears back as they were before I tried metric threads
historicalarms
Ultra Member
I have wondered this before and your post
I have wondered this before while reading other threading pages that if it isnt simply a gear engagement setting that produces the small differences in cut threads. Even a slight difference in gear to gear meshing should make the change in tool speed per rev. and if your gear changes arent exacting , could that not account for the small discrepancies.
Possibly another way to check, with the 40,86,40 gears on,(for imperial threading.) Put the gear box thread selectors in B - 1 position for 8 threads, engage threading half shells, take slack out of gear train. Set indicator to measure carrage movement, mark the lathe chuck/spindle and a reference mark on the head stock. Turn chuck/spindle 8 turns, carrage/saddle should move 1 inch, and or while turning spindle when saddle has moved 1 inch, spindle should have turned 8 turns.
There is an answer, just have to find it, or as Susquatch says" the solution is simple, once you know what it is".
B-1 on the threading gear box should do above, if not "WHY"?
There is an answer, just have to find it, or as Susquatch says" the solution is simple, once you know what it is".
B-1 on the threading gear box should do above, if not "WHY"?
I think we are all wondering similar things but sounds like you have them checked off. 40-86-40 gear arrangement. Obeying the levers shown on manual. Are you engaging clamshells on the same thread tick mark, or leaving the clamshells engaged through tests? Once engaged, are you giving it a few spins in same direction tp take out gear lash before registering your zero position?
Assuming your measurements are accurate I tried to see if the deviation was favoring one handle position or the other, but if there is a trend I don't see it.
On my banjo I can kind of peek inside the gear cluster with a mirror & shining flashlight. Can you visually see if they are meshing properly? Is the cone stacked from big to small (no gears out of position... which I find hard to imagine it would engage?) Do you feel any resistance engaging on either side of the hole position?
Assuming your measurements are accurate I tried to see if the deviation was favoring one handle position or the other, but if there is a trend I don't see it.
On my banjo I can kind of peek inside the gear cluster with a mirror & shining flashlight. Can you visually see if they are meshing properly? Is the cone stacked from big to small (no gears out of position... which I find hard to imagine it would engage?) Do you feel any resistance engaging on either side of the hole position?
I see. Thanks for that, it's very helpful.
Which side of the carriage was the indicator measuring? Any chance the indicator was helping or hindering chuck movement?
I just want to understand your test methodology. Example for 8 TPI you rotated chuck exactly 8 times & measured 1.035" carriage displacement? Then on the high pitch end you rotated 56 times & got 1.024"? Or did you rotate something less & multiply, like 26 rotations measured 0.512 and 2*0.512=1.024 equivalent? Your indicator was positioned exactly the same & parallel to movement each time? (a little bit of plunger angle an make for incorrect measurements)
Is this how you measured the 1” carriage travel when you did your “pitch validation” as you posted in #68 above?
I can see your method to be a bit problematic:
a) typically, one does not use a 1” capacity DI to measure a one inch distance; you should use a 1.5” or 2” capacity one
b) the measuring instrument has both hysteresis & error - if not calibrated / compensated for there can easily be a 10 thou (or more) error in either direction, for 20 thou total in your measurement depending on the quality of the instrument
Another thing: check for axial end-play of your lead screw. If it can move axially (possibly with some resistance) then it might move a certain amount when you did one pitch and more / less / none at all, when you did another.
Keep the lathe turned off.
Pull the lead screw cover (spiral spring) back from both ends.
Install your DI as you did in the picture above (for this test it will be good enough).
Engage the half nut.
Turn the carriage wheel CCW - once the backlash is taken up, the carriage should not move.
Observe any axial movement of the lead screw.
Zero your DI.
Turn the carriage wheel CW; the carriage should not move other than the backlash in the lead screw nut.
Observe any additional movement caused by end play in the lead screw support thrust bearing(s) on the DI reading.
Report your findings here.
There should be no (or only very minimal) end play in the lead screw. If there is, it should be addressed.
Did you count the # of teeth on all your drive gears to confirm that what you are supposed to have, is actually present?
I can see your method to be a bit problematic:
a) typically, one does not use a 1” capacity DI to measure a one inch distance; you should use a 1.5” or 2” capacity one
b) the measuring instrument has both hysteresis & error - if not calibrated / compensated for there can easily be a 10 thou (or more) error in either direction, for 20 thou total in your measurement depending on the quality of the instrument
Another thing: check for axial end-play of your lead screw. If it can move axially (possibly with some resistance) then it might move a certain amount when you did one pitch and more / less / none at all, when you did another.
Keep the lathe turned off.
Pull the lead screw cover (spiral spring) back from both ends.
Install your DI as you did in the picture above (for this test it will be good enough).
Engage the half nut.
Turn the carriage wheel CCW - once the backlash is taken up, the carriage should not move.
Observe any axial movement of the lead screw.
Zero your DI.
Turn the carriage wheel CW; the carriage should not move other than the backlash in the lead screw nut.
Observe any additional movement caused by end play in the lead screw support thrust bearing(s) on the DI reading.
Report your findings here.
There should be no (or only very minimal) end play in the lead screw. If there is, it should be addressed.
Did you count the # of teeth on all your drive gears to confirm that what you are supposed to have, is actually present?
Rudy @RobinHood above has some good points to check.
Brian - You've got me wondering about my own lathe. Now I need to do the same test.
Yes, you should be able to go back to King - this should and will work.
A magnetic base for your dial indicator would make it much easier to measure carriage movement and eliminate some sources of error. Did you validate your dial indicator is working correctly?
One last point - just the basics - there are two handles which engage the carriage - the left one in the pic above is the power feed, the one the right engages the clam shells on the lead screw. The right hand one is the one to use.
Brian - You've got me wondering about my own lathe. Now I need to do the same test.
Yes, you should be able to go back to King - this should and will work.
A magnetic base for your dial indicator would make it much easier to measure carriage movement and eliminate some sources of error. Did you validate your dial indicator is working correctly?
One last point - just the basics - there are two handles which engage the carriage - the left one in the pic above is the power feed, the one the right engages the clam shells on the lead screw. The right hand one is the one to use.
haha, my post was plugged in the internet pipes last night when it went down & I went to bed. It echo's what @RobinHood said
Another test. When you engage your threading clamshells & just confirm its 'threading' by hand rotating chuck, put an indicator on the carriage. Push & pull along the bed direction. Are you seeing any needle movement? I guess this could detect leadscrew/clamshell thread lash. Or if your leadscrew has axial play for example: shear pin in hole, shaft in bearing/bushing, loose clutch if you have one there...
Another test. When you engage your threading clamshells & just confirm its 'threading' by hand rotating chuck, put an indicator on the carriage. Push & pull along the bed direction. Are you seeing any needle movement? I guess this could detect leadscrew/clamshell thread lash. Or if your leadscrew has axial play for example: shear pin in hole, shaft in bearing/bushing, loose clutch if you have one there...