An attempt to execute the 400 rpm threading for a G0602 plan

It does work, although given a choice a rotary table that is meant to be on its side is superior. An index plate is not needed for this project.With the standard 90:1 table, the 40t gear is 2 turns +1 degree and the 30t is 3 turns even. A checklist and attention to detail should suffice. Now if I make more gears for other projects, I will upgrade to an index and a ground true stub arbor. Gears sound scary. I avoided them in the past, but they turn out to be a lot easier to do than some other operations.
 
I thought that I would be able to mount the clutch on the lathe tonight. But alas, I need to go buy a gear puller to get the spindle pulley off. Off to HF tomorrow for a tool.

Everything spins nicely and the manual shift linkage works well. Neutral is unstable. The clutch likes to be in either forward or reverse. The shift detent will fix that once everything is put together.
IMG_1107[1].JPG
 
I got the pulley off and faced the next dilemma. I needed to turn down the inboard end of the pulley stack to clear the input gear, But no lathe. I got the pulley reasonable centered on the rotary table tooling plate and was able to mill it. The 6" pulley in the 7" tooling plate was an interesting clamping exercise. Sorry, I was to engrossed to take a picture.

I always have an assortment of metric screw in a box, but the longest 8mm screw was just barely long enough to grab a thread on the headstock. I will visit my favorite hardware store tomorrow to get a proper screw. I was able to get the clutch to stay on the headstock long enough to power it up. It was noisy, which I expected, but it spun up, shifted smartly, and turned the lead screw in both directions. The clutch is held in position by a single mounting screw. It tended to rotate on the mounting screw when shifting directions. That may stop with a longer screw that can be tightened more, but I am tempted to drill the body and headstock for a 1/8" dowel pin once everything is assembled and in position. We used to do that kind of thing with linotypes in the olden days of printing..
IMG_1111[1].JPG

The body interferes with putting the belt on the inner groove. I'm not sure why my rendition has that problem, but I took the body apart and cut away part of the body. It is still a little tight, so I may take off a little more before I mount it with the new screw. The picture is a bit dim, but it is in the red tinted area near the pulley.. I guess I could use a good macro lens and better light..
IMG_1113[1].JPG
 
Larry, I am impressed with your progress. Thanks for posting your experiences on this project. Most folks naturally shy away from doing things they have not tried before, like cutting gears. You have shown the community that with a little patience it is not much more difficult than other routine operations. On the rotating clutch body I would take a close look at drag or interference within the assembly. Out of round gears etc, I have never had a problem with rotation. I suspect the gear loading is light and that is the reason. Adding a dowel pin or even a second clamping bolt could be an option. I did not want to bore or tap any extra holes in the headstock, doing such by hand is difficult and ending up with holes that are not square to the surface can be an issue. The pocket that the bolt head fits into is oversize so that a SAE type washer would fit under the socket head screw. This helps spread the clamping load. I turned the washer from some steel rod stock.

A variable speed motor assembly will solve your interference problems. Variable speed was the first change I made after purchasing the lathe. I used a DC treadmill motor and a KB speed control. It has lasted all these years much to my surprise. I have always told myself that as soon as the brushes wear out I will buy a VFD unit and motor to make a more sophisticated drive assembly. The brushes are still working fine, not sure when I will be able to justify the VFD.

Keep posting your progress and any problems you have, you have done a really fine job.

Jim
 
Don't get too lavish with the praise just yet. I got a longer 8mm bolt and clamped the body down on the headstock. It still wanted to rotate a little when shifting at 720rpm. Then it started not wanting to shift in and out at all. I had magically developed a lot of end float on the outer axle. The clutch disk also was also hitting the edge of the recess in the dog gears.

I took it apart and started measuring things. The washer that I turned to adjust the end float had shrunk. More likely there had been a bur somewhere that had been removed when the clutch system had been spun up to speed. I am thinking of drilling and threading for a set screw between the teeth of the output gear above the key. That way I can adjust it to whatever clearance I want. An 8-32 tap will reach all the way to the key. - Edit: A set screw will not work. I need to fill up 0.031 on the inner axle. Clamping the output gear to the outer axle does nothing. I need a shim washer 0.031 thick, with a hole larger than the inner axle and an OD smaller than the clutch boss. Maybe turn the diameters on the lathe (after putting it back together) and then milling it close to thickness. Finally lapping to achieve the clearance I want.

My output gear was defective when I fabricated it. I bored the gear blank to 0.718. It was a metrology error. I have little faith in my hole measuring capability and none in the depth department. My next project will be targeted to that end. The outer axle was measured today at 0.7183. The dog gear was 0.7195 and the output gear was 0.7196. The clearance would be 0.0012, which is in the range of close running fits.

The clutch disk is a problem child. The current one is the third that I turned/milled. It is a deceptively simple object. The clutch boss on the outer axle measured at 0.9433 and the clutch disk bore at 0.949. This is 0.0057, which is much looser than it needs to be. I was targeting a free fit so that it would slide easily under load and got carried away. With such a loose fit, the outer edge of the clutch tilts and sometimes touches the wall of the dog gear recess and jams. I could turn/mill another clutch disk with a tighter bore, but my failure rate in that endeavor is high. I could also turn a new outer axle if I had the material, but I don't. The solution I think might be a simpler way is to mount the clutch disk on the arbor and take off 0.030 of its radius.

Any advice from the experts?
 
Last edited:
Hello Larry, lets concentrate on one thing at a time. The two counter rotating gears and the clutch disc are the heart of the entire assembly. I have many hours of use on my unit and I have never adjusted the position since the initial installation. That has been about a year and one half ago if I remember correctly. I suspect that you may have a gear that has radial runout or the teeth have not been cut uniformly. With the unit assembled and clamped in a vise you should be able easily rotate the input gear and shift between forward, reverse and neutral with no binding. I believe you mentioned earlier that you had to shim the axle, did this perhaps create other problems?

The dimensions of the axle, both diameter and length is very important for smooth shifting. When you add up the length of the outer axle, the end washer, and the circlip you should have a few thousands of end clearance. Probably 0.002 to 0.005 or there about. Next pay attention to the diameter of the bronze bushing and the outer axle portion that the clutch disc rides on. These must be very close, if the bronze bushing is larger in diameter than the outer axle the disc will catch on the lip. You should have a very slight chamfer on the clutch disc bore, this will help the shifting. Next the clearance on the disc inner diameter should be a few thousands, you do not want the disc to cant when shifting. Even with your 0.005" clearance on the clutch disc if the shifting fork is flat and true it should shift smoothly.

The occasional tic-tic of the disc hitting the pin is not a problem. Next pay attention to the shifting fork. This must be very flat, I milled mine from 3/16" flat stock. I only milled the outline initially. After the fork was finished and deburred I milled a very light recess (0.005-0.008) on both sides of the fork. This recess was large enough that the clutch disk would fit inside of it. The purpose of the recess is provide a little more tolerance for the dimensions of the gear stack and the detent position.

Start by making sure all of the gears are uniform have minimal runout. Then focus on the tolerance in the gear stack, clutch and axle assemblies. Then make sure the shifting fork moves without tipping to one side or the other. The pressure on the detent should be very low, I used a portion of the spring from a ball point pen.

Without being able to see the unit my guess would be that the rotation of the unit is a result of gears binding and the shifting issues are in a large part due to the shifting fork not being square when shifting is taking place.

Keep after it and let me know how you are doing. You are solving problems that other are likely to run into, your leading the pack so to speak.

Jim
 
It is back together aned seems to be functioning.

The gears are not perfectly concentric with their bores. It is a byproduct of not turning the mandrels between centers. I've always shied away from working between centers, but it aint all that bad and you get good concentricity, even if you have to swap the part end for end to do a shoulder. It sucks not having a lead screw reverse. I did the inner axle between centers and it is right on. The concentricity error is in the neighborhood of 5mils. I didn't think that was the problem, and it turned out not to be.

The noise got reduced a lot by adjusting the reverse idler position. The teeth were just grazing each other. I moved it 20 mils away from the forward dog gear.

The end play is still there. I tried to make a 30 mil thick washer to tighten it up, but the mill just didn't like working with something that thin. I don't think the end play makes all that much difference, except to look bad.

The shift problem was a multi-problem. The clearance from the clutch to the dog gear recess was too small. I reduced the outer diameter of the clutch by 60mils. The clutch drive key was tight, so I did a little polishing to make it slide easier. I also chamfered the clutch bore and eased the edges of the bushing as much as I could with the dog pins in place. The dog gear bushings are a tad too big. I can feel the clutch climb up on them. Reducing them would require the dog pins to be pressed out, so that will wait until I do version 2.

It shifts right snappy and the body no longer moves around. I did discover that you have to be more careful tightening the gear train when doing frequent shifts in high gear. The shock tends to loosen things up.
 
Well, this a bit embarrassing. Especially for someone that did failure analysis and remediation for new processes. I remounted the shift linkage and it stopped shifting. Closer inspection showed that I hadn't rounded the top of the upper shift linkage enough and it was hitting the headstock. Previously , when I thought that I had clutch problems, I didn't... well I did have some problems, but they had nothing to do with shifting. Earlier, the shifting problems started right after I tightened up the upper shift lever pivot, but I was so engrossed with the shifting process that I didn't inspect the thing I just changed.
 
You must log in or register to reply here.
Back
Top