When is close enough?

[Royal Viking]

I thought I knew how to center a part in a 4 jaw until now. I have centered a part in my 4 jaw chuck. It took me quite a while to get it centered due to the shape of the object (noted on a different thread). I am using a dial indicator that reads to .001 inch. I have it so that the needle now fluctuates about two needle widths when the chuck is turned. This is about 1/5th of the distance in the dial ticks for .001 inch.

The part that I am making is a taper adapter sleeve to mate the crankshaft of a motorcycle engine to a CVT clutch. The part that I have chucked up is the stub left over from the original crankshaft that failed, it had snapped off the shaft. I need to find the angle for the taper from this piece to make the inner taper for teh sleeve. The failure could be attributed to a flaw in the crankshaft as well as an out of balance at the end of the shaft due to the clutch retaining bolt being tapped off center.

This engine may see 9000 rpm and here I am wondering "how close is close enough"? When you are building a model engine how do you determine the tolerances?

 

[kcmillin]

I cant help you on what the tolerence is, however, If you have, or can get a hold of a dial test indicator that goes to .0005", that would give you even greater accuracy, and frustration.

Kel

 

[Royal Viking]

Thanks Kel, I appreciate any response to my question.

I realized this may not be an easy one to answer. I'm going to try to post some pictures of what I am trying to do as soon as I can get more than 30 minutes at the computer. :

It took me 3 hours to get the part centered as close as I have it now , I understand what you mean by more frustration.

 

[BillTodd]

Compared to the engine vibration, being a couple of Thou" out with the CVT is unlikely to cause you (or you engine) a problem. However, If the CVT is a few inches in diameter a couple of thou" will likely produce a visible wobble and you wouldn't want that would you

Hang your DTI on the engine. If the crank is pressed together (like a two-stroke) it may not be all that square to start with

Bill

 

[ieezitin]

Viking.

People and I am one sometimes get bogged down on precision and tolerances when design and building stuff, I find the best solution is to really think through whats the part your making and its function relative to the large picture. I tend to look for faults or weak points and concentrate on them, you can really spend a lot of time on something that don’t need the effort put into it.

I like to sketch out everything and at the end of the day sit at my desk with a beer just studying the prints, its amazing what a little peace and quiet can do for planning.


Anthony.

 

[Admiral_dk]

Hi Viking

This is potentially a nasty one ....

Years ago I owned a GSX 1100 EZ that "lost" the generator several times .... It's mounted in the same way as your CVT on the left end of the crank. I my case it wasn't the crank that broke, but a very small error in the fit off the two parts that was to blame for my troubles. I couldn't / wouldn't do anything with the crank as such - so I got the best fit by using carborundum (sp?) aka valvegrinding paste between them and grinding them for 30 minutes (both parts are hardened) in order to get a good fit. This didn't ensure that they held together on a 111 HP engine, but it's a good start. If you're making new parts yourself - PLEASE include a woodruff key in the design if it doesn't include one already and avoid most of my troubles from back then.

As to tolerances .... what's really important here besides the ability to withstand the power transmitted, will be vibrations. Since they are a relation between the distance of the offending unbalance to the center of the axel, I would say that the bigger the diameter off your CVT the more precise you need to be.
I would expect the factory tollerance spec to be 0.01mm. ~ 0.004" as this is the outrun most of my Suzuki bikes has as max. on engine axels.

Hope this helps.


Admiral_dk aka Per

 

[Stan]

If I understand your problem, being centered has nothing to do with finding the angle. What is critical is to have the centerline of the stub shaft parallel to the lathe ways. You then move your indicator a fixed distance along the shaft and record the difference. You will then have two sides of a triangle to do the math on.

 

[Royal Viking]

Thanks for all of the comments. :bow:

The CVT is 7" in diameter and is installed in place of the alternator (the car runs off of the battery). I should have investigated the runout for the cranshaft (I have the manual). It is a pressed crankshaft as well.

What got me wondering about this in the first place is the effort I took to get the stub trued. I guess I will get to see if it was worth the effort as time goes on with this project. I really have not done anything like this before. Most of my projects have been crude in comparison.

At this point I have it true enugh to set the compound rest angle to bore hole for the sleeve and to make the arbor to hold the sleeve when I turn the outer taper. I'd love to be able to make a jig to set the angle quickly the next time I do this (will I'll ever have to do this again ???).

I have some time this weekend that I can get back to the shop and have at it.

 

[Admiral_dk]

I've just double checked again and I got those tolerances wrong or should I say, that they depends on how long the shaft is. The max runout on the 16" ~ 40cm. crankshaft is 0.1mm ~ 0.04", where a 3" ~ 7.5cm shaft has a max runout of 0.03mm ~ 0.012" in the same engine.

In your case, with a 7" CVT on the shaft, I'll say that a max rounout of 0.03-0.04mm ~ 0.012-0.016" is in the ballpark. With this figure, you shouldn't have problems with vibrations.

 

[Royal Viking]

One thing I did notice is that the original crankshaft had been drilled and tapped to fit the CVT bolt. The hole was not drilled parallel with the crank. I checked the runout of the bolt and it was .020 off center. I marked that the high side and then marked the low side. When I removed the stub from the chuck I found that the failure of the crank originated opposite the high side (makes sense).

The pen is pointing to the high side. The final failure is the crater.

 

[arnoldb]

RV - unfortunately I cannot add any help regarding your plight.

Please don't think I'm pedantic; I don't want to see anybody making a mistake though, so just to clarify Admiral's measurements:
0.1mm ~ 0.004" and not 0.04". similarly 0.03mm ~ 0.0012" or 1.2 thou.

Admiral, somewhere you're getting your conversions wrong with a factor of 10

Kind regards, Arnold

 

[Admiral_dk]

Ups .... I stand corrected !!!!

Although I've been working for years in mills on PCB design, I'm still getting things wrong when working in inches in all other relations and do most certainly prefer metric anytime - that's with the possible exception ; when designing extra small PCB's I'm so used to mills, that it requires me to think twice when doing those small sizes in metric .... :big: :