The relationship between the bushing and the input shaft does not stay constant over time. Over time the i.d. of the brass bushing will open up and therefore create "slop" at the input shaft. If you grab hold of the Muncie input shaft you can physically "move it" up, down and side to side. With a TKO you cannot do that. The tolerances are tighter. By using a needle roller pilot bearing with the Tremec, the i.d. will not "wallow" out over time, it will stay constant. Tremec will not honor any warranty if you use the brass bushing. They require use of a needle roller pilot. Lakewood actually claims the needle roller reduces parasitic drag and does not absorb the heat the brass bushing would. In all honesty, depending on the quality, a needle roller should last about 100,000-150,000 (not as long as brass) but the the i.d. will stay constant over that time where the brass bushing would not.

Richard
Tech Support
Keisler Engineering

 

Richard I was actually at the head of the math class and I am still having problems deciding what you just said?
Do we divide by 2 to get the runout?? Is it possible to have two positive reading at both 3 and 9??
If 12 oclock is Zero and 6 is .014 how much is the runout???

 

Thanks,that explains it. I was leaning towards the bronze bushing but, since you put the warranty up for grabs, I'll be using the roller bearing.

 

No, you divide by 2 to determine which offset pins would correct your measurement. In your example, your TOTAL runout is .014, You would need .007 offset pins to correct it. You would clock the offset at the 6:00 position to bring the centerline of the register hole perpendicular to the crank. In other words, the centerline of the register hole (in your example) is higher than the centerline of the crank. Yes it is possible to have a positive at 3:00 and 9:00 and depending on the exact measurements, yes it could be corrected.
This is a very hard concept to grasp and as I already stated, I am no math wizard. I only hope I am clarifying and not confusing this issue. I have made some drawings (examples) of runouts and as soon as I can figure out how to scan it from work here and post them I believe it will help. I will work on that on Monday.

Richard
Tech Support
Keisler Engineering

 

If you say the max dial indicator reading is .005" then the offcenter limit would be 0.025". This is an incredibly small amount. When I worked in a machine shop the tolerance on the parts I was was making was +-.003"
I know these were not high precision parts.

Now getting back to the engine, crank centerline, bellhousing alignment etc. There are many parts involved here from the crank to the crank main bearing to the bellhousing itself. 1 thou here and there added up it does not take much to get out of tolerance as , I wonder how many are dialed in the first time they check the bellhousing alignment with the dial indicatior.

 

[QUOTE=norvalwilhelm] Is it possible to have two positive reading at both 3 and 9??
QUOTE]

Only if the register hole is oblong.

 

I agree. This seems like an incredibly tight tolerance. I believe you mean 0.0025" offset limit. I'll be trying to achieve this tighter tolerance and I'm sure it'll lead to a better quality job, but in all honesty I have to wonder if it's justified. Looking at the finish of the ID of the register hole on the Lakewood bellhousing, I doubt it's finished to within 0.0025" The surface is rough. It looks like it was punched out with a dull tool, definitely not machined.

 

I'm glad to see i'm not the only one who's been pondering the tight tolerences regarding runout measurement.

I have just finished swapping from auto to 5sp manual.
When I came to dial in the old Borg Warner cast bellhousing I'm using, the tolerences were outwith Keisler specs', but the inner surface of the gearbox hole (register?) was painted.
I removed this and cleaned the surface with fine emery cloth, ensuring I didn't remove any "meat". Then rechecked the fit onto the transmission to ensure no slack has resulted. It still seemed nice and tight, and after refitting the bellhousing to the motor, my tolerences were within Keisler spec'.

I have seen no mention of checking the bellhousing register to transmission tolerences.
0.005" is a fantastically tight tolerance for mating two great lumps of engine and transmission together.
Transmissions/gearboxes (over here anyway) tend to have an certain degree of lateral float in the input shaft to take account of minimal offset.

Do the production lines of brand new cars that use TKO's measure each motor to bellhousing to transmission runout? I don't think so, mate them up and bolt them down!

Anyway, I've done my best and wrote the figures down to show it.
Once I've finished fabricating the clutch linkage, I'll get it turning and see how it goes.

Cheers,
Douglas
Glasgow, Scotland.

PS. This is my forray into forums, after nine years of owning my Corvette, and two years of owning a computer!

 

If your using a stock bellhousing inspect the two dowel pin holes for fine line cracks to the outside. If you try to use offset dowels to align one of these you will only be chasing your own tail.

 

And a fine foray it was.

 

This is just plain wrong. The spec. is plus or minus .005" for a total of .010". In otherwords a total 360* rotation needle movement of a maximum .010". Not a total of .005" divided by two, .0025". A piece of bond paper is about .0035" thick and you want to get within .0025"??? This specification has been around for years but has never been an issue because. Muncies and T-10s and other transmissions like them use a ball bearing front bearing that will tolerate a slight misalignment without bushing wear and shifting problems. The Tremec uses a tapered roller bearing (like a pinion bearing) with an end play spec of .000" to .004". What this means is when the transmission heats up there is a slight (intentionally built in) preload on the bearing. This means no side play on the input shaft which is why they require the spec. of plus or minus .005" to be there. If you are off this spec you could have pilot bearing failure and shifting problems or worse. Look at motorheads copy of Lakewoods instructions #3 One half the indicated reading .005"

Lets look at the Lakewood specs below. The bore diameter is 4.684" and the retainer flange on the Tremec and Muncie is 4.680. What this means is that the hole in the housing is .004" bigger than the transmission. This will leave a .002" space all the way around and will also mean that when installed your Tremec will be .002" out of spec even if you get your Lakewood PERFECTLY aligned. In order to get the spec. of .005" total that Richard suggested you would have to be aligned to within .001" total (because of the Lakewood larger bore diameter)

Brand: Lakewood Suspension
Product Line: Lakewood Safety Bellhousings
SFI Approved: Yes
Block Plate Included: Yes
Bellhousing Bore Diameter (in): 4.684 in.
Bellhousing Depth (in): 6.450 in.
Clutch Fork Included: No
Adjustable Fork Pivot Recommended: Yes
Bellhousing Material: Steel
Bellhousing Finish: Red enamel

Lakewood housings are made for racing! They do what theyr'e supposed to do. Contain an exploding clutch. Most Racers use the weld on ring and dowel kits http://store.summitracing.com/partde...part=LAK-15981 for their alignment and then forget about it, the other alignment spec not mentioned here is the parallel within .002" spec. The flat transmission mounting area on the bellhousing should be within .002" of parallel with the back of the engine. If a piece of bond paper is .0035" thick, how thick and even is the paint on the Lakewood housing??? If you consider the fact that your Lakewood housing MUST be aligned and you MUST use offset dowels priced at ($17.00 for the cheapies) to ($47.00 for the RobbMc/Lakewood E-Z adjust) and the time necessary to do this, wouldn't it just be easier to install a housing that is made to be in spec. from the get go? If you want an SFI approved blow proof housing that is made to be in spec, and has a correct bore diameter, spend a little more money and get a McLeod and be done with it. These housings were made for the guy who not only wants to be race legal, but also to have parts that meet factory requirements and just bolts on. Just make sure your engine has never been align bored. If it has, you will have to align your bell housing regardless of the make.
JJ

 

Maybe this will help



JJ

 

The spec is: .005 TIR (total indicator reading). This means +/- .0025 indicator reading.
.010 TIR will ruin the bearings in a Tremec in no time at all.

 

I am with JJGoodwrech on this.
If a cop stops a drunk and gets him to walk the white line. In the his mind the cop says if you deviates from that line by 6 inches I am going to consider him drunk.
The "drunk" wanders 4 inches to the left then 4 inches to the right. back and forth each step.. Has he deviated from the center line by 4 inches or a total of 8 and the cop then can charge him?

When a spec says a diviation of .005 it means plus or minus of .005 not 1/2 the value or .0025 If you want a spec of .0025 that specify that amount not make the person do the math and figure it out for himself.

 

Gentlemen
The center of the input shaft can not be more than .005" from the center of the crank centerline. Norvalwilhelms drunk is busted. Norval gets a gold star. Tremecs number is 1 (800) 401-9866 (U.S. and Canada only) Give them a call.
JJ

 

I agree with the first part, the two centerlines can never be more then .005 apart, that is plus or minus or a total reading of .010 is at the limit.

As for my drunk he never deviated from the line by 6 inches, he always stayed within 4 inches plus or minus.

Have you called?? If not I will tomorrow.

 

I have been repairing transmissions for 32 years and am very familiar with the spec. and how to measure it. I'm sure about it. You can call if you wish.
JJ

 

JJGoodwrench we are not disagreeing. I believe the dial reading can be 010 maximum for a .005 runout. I thought you were having second thoughts.

 

The drunkard analogy reminds me of an old W.C. Fields line: "I'm deeply indebted to my last ex-wife... she drove me to drinking."

 

Oh sorry. Yeah, I misunderstood. No second thoughts.
JJ