Measuring ring gear lash

Topics Last Day Last Week Tree View    Getting Started Formatting Troubleshooting Program Credits    New Messages Keyword Search Contact Moderators Edit Profile Administration
Model T Ford Forum: Forum 2008: Measuring ring gear lash
Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Sunday, January 11, 2009 - 07:28 pm:

This is the first time that I have ever done this. I purchased a cheap dial gauge and set it up as shown in the picture. The question I have is, the gauge is on about a 40 degree angle. Since the gauge is not square with the movement must I take the angle into account. With this setup I measure about 0.019" lash, but I suspect that the actual lash is more by some ratio. Am I thinking correctly?


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Sunday, January 11, 2009 - 07:30 pm:

Forgot the picture


Top of pagePrevious messageNext messageBottom of page Link to this message  By Les Schubert on Sunday, January 11, 2009 - 08:06 pm:

Mike
I would be inclined to measure the lash at the pinion although I don't guess it really matters. Before I did that I would spin the drive shaft with a wrench or something and listen and feel how it turns. If it "clicks" then it is probably loose. It it binds as the teeth engage then it is probably tight. If it turns freely and quietly then it is probably OK. .019 sounds like a little on the high side to me.


Top of pagePrevious messageNext messageBottom of page Link to this message  By james dimit on Sunday, January 11, 2009 - 08:08 pm:

I wondered about that too, I found I could get a lot straighter shot on the pinion as shown in this picture.

I figured it was easier to move the pinion without moving the ring, than the other way around. Jim


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Sunday, January 11, 2009 - 08:11 pm:

Everything is free and turns fine. I'll try the pinion idea. I was just reading about dial indicators. Apparently I have a probe dial indicator, which is used for up and down measurements and they talk about 'cosine error', bet that is the angle compensation. I need a lever arm dial indicator to use it this way. I'll get it a try after supper, thanks.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Stan Howe on Sunday, January 11, 2009 - 08:24 pm:

Don't make it too hard, Mike. Run a piece of solder through the gears. If it is about the thickness of a playing card, it has a little too much clearance. It is is the thickness of a black hair it's about right. If it's the thickness of a blonde hair it's too tight.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Walker, NW AR. on Sunday, January 11, 2009 - 09:04 pm:

And if it's a RCH, it's WAY too tight! :-)


Top of pagePrevious messageNext messageBottom of page Link to this message  By Norman T. Kling on Sunday, January 11, 2009 - 09:11 pm:

Are the gears new or used? It is also important to get a good mesh pattern. If they are tight in one spot and not all over, they will wear out soon in fact they could even chip teeth that way. Use some prussian blue to check the pattern after you get the clearance set. Used gears sometimes get worn so that when you set them closer together they don't mesh very well. You want to get the lash as close as you can and still get good mesh.
Norm


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Sunday, January 11, 2009 - 09:14 pm:

Well, I tried again, using James setup.



I noticed that to get consistant measurements you have to measure at the same place. So I did it over and over. Made a chart and this is what came out

12 o'clock 0.014"
2 o'clock 0.014"
4 o'clock 0.013"
6 o'clock 0.014"
8 o'clock 0.015"
10 o'clock 0.017"

I tried a couple of teeth near 10 o'clock and they are all 0.017" 0.018". Tried the solder method and it agrees pretty well with the indicator, except maybe 0.001" less. Could be the difference between my indicator and vernier. Most likely my skills at using the tools. Anyway, I think the lash is too much and the run out is not perfect but OK. If I put a 0.010" shim under the ring gear, how much would that reduce the lash?


Top of pagePrevious messageNext messageBottom of page Link to this message  By Richard Gould on Sunday, January 11, 2009 - 10:36 pm:

I like that amount of lash you have. I'd button it up. I am impressed you have such a little amount of run out, too. I think you're right on the money. If you add a .010" shim you'll have between .005 and .010" lash. That's getting a little close for me.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Richard Gould on Sunday, January 11, 2009 - 10:39 pm:

I like that amount of lash you have. I'd button it up. I am impressed you have such a little amount of run out, too. I think you're right on the money. If you add a .010" shim you'll have between .005 and .010" lash. That's getting a little close for me.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Jerry VanOoteghem on Monday, January 12, 2009 - 03:44 pm:

Looks good Mike.

As Les suggested, proper mesh is equally, or even more important, than measured backlash. A smooth, quiet mesh lets you now that everything should be aligned correctly and the measurements should, (and seem to), confirm what you're feeling when rolling the gears in mesh.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Jerry VanOoteghem on Monday, January 12, 2009 - 03:45 pm:

Oh, to answer your initial question, yes, the angle makes a difference. You seem to be beyond that point however.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Monday, January 12, 2009 - 05:30 pm:

The gears are new. I never done a pattern with perisan blue, but I have seen the suggested pictures. I'm going to ask some friends about this and hopefully get some local insite. Thanks.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Hal Davis on Monday, January 12, 2009 - 06:51 pm:

Several years ago when I was doing my Model A (Only diff I've ever built, so I'M NO EXERT!), I went through all the backlash measurements and did the prussian blue indicating, moved everything around where the pattern was centered like it was in the book......Dang thing roared on the bench turning it by hand! Bear in mind, I was on a budget and was using the old gears. I said the heck with all the backlash and the centering up of the pattern and put the prussian blue pattern back on the worn spots where it had been riding for 70 years....Guess what? Quiet as a mouse! Still is, several thousand miles later.

For new gears though, I'd follow Les and Jerry's advice.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Monday, January 12, 2009 - 09:56 pm:

Well..... here is what I did, Since I didn't have a ring gear shim, but did have a thrust washer shim (0.010"), I installed it and tried to measure the new lash. Turns out, it tightened things to the point that it was hard to rotate the differential. So I think that a ring gear shim would do the same thing and I'm going to leave out the shim and live with the lash as it is. I also stoped at NAPA on the way home and purchased some persian blue. I dabbed some on three teeth with my finger and rotated the ring gear thru the pinion gear by turning the drive shaft. Here is the pattern.



Looks to me that there is contact all along the tooth. I see bright from the inner radius to the outer radius. There seems to be a dark straight line near the top and bottom of the tooth, is that were the contact stops. I think that the contact is good and I'm going to live with the lash as it is. So next is to torque the ring gear to 25 ftlb and safety wire it. After that I have to adjust the thickness of the right side thrust washer, I've noticed with the housing bolted together the axles are stiff, so I'll grind off a few thousands and try it again. Is there anything else I should be looking at?


Top of pagePrevious messageNext messageBottom of page Link to this message  By Les Schubert on Monday, January 12, 2009 - 10:45 pm:

Mike
Another option to gain some clearance in regards to the thrust washer would be to stack another gasket between the housings. I believe this was the method that Ford recommended and I there might have been various thicknesses of gaskets available, at least I know there were for the A and V8 axles. You could easily cut one from some thick paper I would start with 2 standard gaskets if you had them.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Monday, January 12, 2009 - 10:58 pm:

Are you thinking that the gasket would be an alternative to grinding the thrust washer?


Top of pagePrevious messageNext messageBottom of page Link to this message  By Tony Bowker on Tuesday, January 13, 2009 - 12:52 am:

In answer to your original question about the effect of the acute angle of the dial guage on the reading. It depends on the angle, if we guess say 30 degrees, then the .019 will REDUCE to .016 in linear motion at the ring gear. The computation is the cosine of the angle times the observed travel. If the angle is 45 degrees then the linear change is .013.
Mind you the better way is to keep the ring gear stationary and observe the travel on the pinion gear as suggested by other writers. Once the angle is less that 15 degrees it almost dissapears form the calculation.
In the original MTFCA Ruckstell manual, not the current issue, Bruce recommended using paper to measure the clearence between the pinion and ring gears. You run the paper between the gears and rotate the gears and look at the paper. There should be 'some' marks. No marks indicates too great a clearence and broken paper too small a clearence. Much to unscientific for the latest issue of the manual which recommends the backlash measurement using the dial gauge on the pinion.
I like the old method, but for customers I use the dial gauge.....


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Tuesday, January 13, 2009 - 10:26 am:

Thanks, your numbers correspond well with what I was seeing. The dial indicator was at about 45 degrees and the resulting measurement was 0.019". When I moved the dail to a better position where it was more perpendicular the measurment was 0.013". I'll try the paper method, just to see what the results are. Looks like you just toke the measurement and mutiplied by the cosine of the angle.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Les Schubert on Tuesday, January 13, 2009 - 11:50 am:

Mike
Yes I am suggesting the gasket approach rather than grinding the washer. If you had a lathe and could machine a consistent amount off the washer then I would go either way.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Jerry VanOoteghem on Tuesday, January 13, 2009 - 02:01 pm:

Mike,

It may tell you more if you put the prussian blue on the pinion gear and then noticed the pattern it imprints on the ring gear teeth. Another way is to run some carbon paper through the gears.

Really, it looks like you've got things pretty well down. There comes a point of diminishing returns.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Ken Todd on Tuesday, January 13, 2009 - 02:06 pm:

If you don't have prussian blue, you can use dark grease.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Tuesday, January 13, 2009 - 03:38 pm:

I don't have a mill, but can gain access to someone who can do the job. My question is how to determine how much to remove from the washer. My approach, was to sand a few thousands off the washer with my stationary belt sander, then try the washer. Once the axles rotate easily, I'd polish the washer face with fine (600) grit sand paper. If I'd use a mill to cut the washer, I would still have to use a trial and error method. How can you calculate a negative clearance. Could I assemble the axle housings and slowly tighten the housing bolts, keeping track of how difficult it is to roatate the axles. Then measure the gap on the axle housing halves? How much thrust clearance should there be on the differential case? A few thousands? Just enough for expansion from heat? I suppose that one solution would be to use the sander idea on one washer, until I have the clearance I want, then have a new washer milled to thickness of the sanded one. This should get me in the ball park.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Jerry VanOoteghem on Tuesday, January 13, 2009 - 04:25 pm:

Mike,

You can do a trial assembly with shim stock, (or feeler gauges), between the two housings until you get a nice clearance. The shim thickness you end up with will be the amount you need to remove from the thrust washers. You next need to determine if equal amounts should be removed from each washer, or just the ring gear side, or just the side opposite the ring gear. Obviously, if you're satisfied with the gear mesh you will leave the ring gear side as-is.

A good clearance would be .005/.010.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Les Schubert on Tuesday, January 13, 2009 - 04:40 pm:

Mike
Jerry's advice is really good. So try some combination of thin washers or shim stock until you get a good fit (It turns freely with no perceptable looseness) then machine the washer to fit (as found minus what you have added).


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Tuesday, January 13, 2009 - 09:15 pm:

Sounds good to me! I'll get it a try. I really appreciate all the comments and help that a guy gets on this forum and you can learn a few things also, Thanks.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Stan Howe on Tuesday, January 13, 2009 - 10:09 pm:

IF, IF you have a stack of shims you can do what I do. Keep adding shims until you get just the clearance you need. Then measure your shim pack. Machine the bronze washer to that thickness and put it together. Other people do it differently. I'm not a fan of the belt sander method of thinning the bronze washer. Pretty hard to get an accurate job.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Tuesday, January 13, 2009 - 10:57 pm:

Yeah, rather crude, I'm going to try the method suggested above. I'm going to shim the housing and then have the washer machined. I was only going to use the belt sander washer as a guide and not use it in the axle. I'd be concerned about loose abrasive in the oil.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Stan Howe on Tuesday, January 13, 2009 - 11:42 pm:

Mike, they make spring steel shims that are the same size as the steel washers that go on either side of the babbitt or bronze thrust washers. You need half a dozen or so of them (.015 x 6 = .090) I have a stack of .015, .010 that I bought plus a couple .020 ones that I made from brass. I stack them up on the internal carrier with it standing on end and test fit the housing. You can see in there and get a feeler gauge in there from the pinion hole. When you get shims enough that you have about .005 clearance, take them our and measure the thickness and machine the bronze washer to that thickness.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Dan McEachern on Wednesday, January 14, 2009 - 10:10 pm:

Mike- you have WAY too much prussian blue on your gear teeth to get a good pattern indication- as others said, the pattern is the most inportant indication of a correctly set up ring and pinion. See that yellow on the ring gear in your original picture- that was where the factory checked the mesh pattern on a test machine to verify correct pattern.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Thursday, January 15, 2009 - 12:48 pm:

I've been looking for some guidance on exactly how to use the perisan blue on the computer, but have not found much. I figure that you just smear some on the gears and run them thru and look at the pattern. You think that I should use a little less? Do you have any other tips on how to use it or interprept the results?


Top of pagePrevious messageNext messageBottom of page Link to this message  By Jerry VanOoteghem on Friday, January 16, 2009 - 12:52 am:

Mike,

Smear some of the prussian blue on a rag and wipe the rag on several gear teeth. You want to "paint" the blue on the teeth such that it's still almost a transparent layer or just a bit thicker. When you run the teeth together you look for areas where the blue has been worn through by the mating gear's contact. Also, looking for traces of blue that has been transfered to the unblued, mating teeth.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Edward R. Levy on Friday, January 16, 2009 - 01:28 am:

I have read the very interesting & to me somewhat complicated posts above & I have a question. When the cars were manufactured what process was used to get the cleranes correct & as a matter of course about how many miles would a rear end run without overhaul. I ask this question only because on any of the modern cars I have owned starting in the late 1950's I never had a differential rebuild or even a noisey situation. Is the modern differential just better by design? Does it have to do with the kind of bearings?
Edward R. Levy


Top of pagePrevious messageNext messageBottom of page Link to this message  By Dan McEachern on Friday, January 16, 2009 - 03:44 am:

Mike- here is a link to a decent discussion of bevel gear contact patterns and some info on how to correct for various patterns:
http://www.qtcgears.com/Q410/PDF/Q410P438.pdf

Dan


Top of pagePrevious messageNext messageBottom of page Link to this message  By Mike Zahorik on Friday, January 16, 2009 - 03:20 pm:

Ed, I'm not sure, but I believe that the Ford factory had jigs and methods that the assembler did not have to check many of these clearances. It was a go, nogo check. Besides someone mentioned on the forum, somewhere that the clearances were very generous. I seem to remember that where we would want 0.005" to 0.007" for lash, Ford allowed to 0.030" This encouraged quci assembly, but caused damage over the long term. I don't think the types of bearing is a first degree term, it's more the clearances and the speed needed to get the product out the door.
Dan thanks for the link. Very Interesting.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Bob Jablonski on Friday, January 16, 2009 - 11:41 pm:

Mike:

Assembly line workers didn't have to worry about "fit" as we do now with worn parts.

Everything was new, & the fit factor was already built-in. Assemble & down the line to finished car.


Top of pagePrevious messageNext messageBottom of page Link to this message  By Bob Jablonski on Friday, January 16, 2009 - 11:47 pm:

Mike:

I used a .010 ring gear shim to take up some of the lash as I had the carrier centered the way I wanted. Ended up with an average of .012. No running noise or detectable clunking starting or stopping.


Add a Message


This is a public posting area. Enter your username and password if you have an account. Otherwise, enter your full name as your username and leave the password blank. Your e-mail address is optional.
Username:  
Password:
E-mail:

Topics Last Day Last Week Tree View    Getting Started Formatting Troubleshooting Program Credits    New Messages Keyword Search Contact Moderators Edit Profile Administration