QT Bellhousing Concentricity Check

[cv2065]

Wanted to get some other eyeballs on my QT indexing results. I followed the Holley video this time around to index the QT bellhousing. Torqued the bell evenly at 25 ft/lbs. Dial indicator was setup correctly and flat in the middle of the surface around the bellhousing opening. Went a few times around once setup and kept coming back to zero, so was good to go.

Holley video said to find the highest point on the dial indicator as the crank turns and then mark the new zero point right as it drops, which I did (highest point was .007 and it dropped to .006), and that gave me a new zero point right at the 270 degree mark. Readings after that were as follows:

0: 0
90: -.0058
180: -.0055
270: -.0015

https://hosting.photobucket.com/images/c310/cv2065/IMG_9427.JPG?width=960&height=720&fit=bounds (https://hosting.photobucket.com/images/c310/cv2065/IMG_9427.JPG?width=960&height=720&fit=bounds)

A couple of questions.

1. Do these readings look acceptable? The 90 and 270 readings are within .005 of each other.

2. When finding the new zero at the highest reading, the reading stayed at .007 for a few turns of the crank between the 180 and 270 degree marks. The Holley video said to mark as it was coming down from the high mark, where in my case I hit .007 around the 220 degree mark, then dropped from .007 to .006 right at the 270 degree mark, which is where I put my new zero point. I saw other videos that set the new point right when the highest value was first encountered, so wanted to verify which was correct?

3. I believe the proper torque is 35 ft/lbs for the bellhousing but I used 25 ft/lbs. Would this matter on the readings as long as all bolts were torqued equally? That's what the Holley video stated but wanted to verify that as well.

Thanks!

[tonywy]

Hello, I would get the offset dowels and try to get that more even all around, closest to zero, as this setting is critical. You have the engine out on a stand now is the time to get it right.

[mrglaeser]

First I would not expect to have negative numbers opposed from each other unless the opening is very out of round. -.0015 at 270 and -.0058 at 90. I would have expected one to be positive and one negative. Did the dial go clock wise from the zero for one reading and counter clock wise for the other? If so then you would have total travel of the dial indicator of .0073 between 90 and 270 degrees and .0055 between 0 and 180. These numbers are divided by two to get .00365 and .00275. For a TKO/TKX this measurement needs to be with in .005. So you would be good on those readings.

I do wonder if you found the correct place to 0 out the gauge. Maybe run it a couple of times and watched exactly where it starts to drop and back up 5-10 degrees? Also don't just check at 90 degree increments I believe you need to find the biggest total difference between readings taken 180 degrees from one another.

https://www.shiftsst.com/blog/how-to-dial-in-a-bellhousing/241#:~:text=If%20your%20number%20is%20greater%20th an%20.005%E2%80%9D%2C,high%20point%20was%20.015%E2 %80%9D%20and%20your%20low

[cv2065]

Thanks for the feedback. The dial was always increasing from zero until it fell from .007 to .006 around the 270 mark. Maybe I need to zero out when I hit .007 initially? I thought the guy in the Holley video was getting all negatives as well, but I'll have to verify. If I subtract the .0015 from the .0058, I get .0043? Here's the video I'm referencing.

https://www.youtube.com/watch?v=riay1R38wQo

I'll try again with the instructions you provided, as they are a little simpler and report back.

[herb427sc]

I think with the numbers you have, you need the offset dowels.
Up .002 and left .002.
Reset to ZERO and you are less than .001 all the way around.
I think there is an offset dowel .007, that is what you need.

[jrohrig]

Just spent way too much time on my qt bellhousing 6060. I watched the same video as you, many times I might add. Stumbled across this video and I'm glad I did. Made the process much easier for me: https://m.youtube.com/watch?v=fu9bg-fJdnc&t=125s&pp=ygUgc2lsdmVyIHRyYW5zbWlzc2lvbnMgYmVsbGhvdXNpbmc %3D

I have a tkx transmission and needed to add the silver adapter ring to get everything to fit. Just don't measure your run off with it on. The fine folks at sliversports transmission assured me of this and and I was finally able to calibrate my dial indicator

[OB6]

I used this video too and agree it made it so simple. Fortunately everything was in spec with no needed adjustments.


Stumbled across this video and I'm glad I did. Made the process much easier for me: https://m.youtube.com/watch?v=fu9bg-fJdnc&t=125s&pp=ygUgc2lsdmVyIHRyYW5zbWlzc2lvbnMgYmVsbGhvdXNpbmc %3D

[cv2065]

Thanks guys. Watched the video and yes much simpler. I’m going to do this again tomorrow, but based on my current readings, 180 degrees from my high point of .007 where I zeroed out was -.0055. Divide that by 2 and I’m at -.00275 and within spec? Not sure how the negative factors in?

[ggunter]

Looking at your readings you would have to tap your bellhousing a couple thou from the four o'clock position towards the eleven o'clock position. You are aloud .005 total runout to be within spec. But always try to get as close to zero as you can.

[gbranham]

I watched the video referenced above, and it makes sense, but I don't understand how you can get a good reading of runout if the dial indicator isn't dead nuts in the center of the bore. Theirs doesn't appear to be, so how can you effectively measure runout if the gauge isn't dead center in the bore? Or maybe the better question is...how do you ensure your gauge is dead center before you start measuring?

Greg

[OB6]

It doesn't need to be centered. Here's a better explanation from Silversport:

201624

If I recall correctly, you need to zero-out the dial at your starting point.


I watched the video referenced above, and it makes sense, but I don't understand how you can get a good reading of runout if the dial indicator isn't dead nuts in the center of the bore. Theirs doesn't appear to be, so how can you effectively measure runout if the gauge isn't dead center in the bore? Or maybe the better question is...how do you ensure your gauge is dead center before you start measuring?

Greg

[gbranham]

Still doesn't make sense, but maybe it will when I do this with my bellhousing.

[OB6]

Still doesn't make sense, but maybe it will when I do this with my bellhousing.

It's not terribly intuitive. I look at it this way...if you've got X amount of runout, the dial will read X whether it's at the center or closer to the edge of the bellhousing opening. And the dial will rotate on a perfect circle because it's fixed to the flywheel. This still may not be helpful.

[gbranham]

Yeah, that's helpful, thanks. I thought a lot about this last night, and I think I understand it. Regardless of where the dial indicator is placed, it's still going to scribe a perfect circle as the crank is rotated, so runout is still measured as the distance from that circle to the edge of the bellhousing bore.

Greg

[cv2065]

Just to follow up, I took the bell off and noticed that the mounting area had been sprayed with the block and there were some highs and lows with the paint. I took a 220 grit sanding block and lighting sanded any high areas smooth all the way around the mounting surface to get the best fitment/reading. I also noticed that the dial indicator bracket was not holding the indicator properly, so took it back and got a new one. Popped the bell back on and went through the motions again and verified the readings and new zero twice. This is what I came up with:

High Point and New Zero: (+.0032)
90: -.0039
180: -.007
270: -.0032

https://hosting.photobucket.com/images/c310/cv2065/IMG_9439.JPG?width=960&height=720&fit=bounds (https://hosting.photobucket.com/images/c310/cv2065/IMG_9439.JPG?width=960&height=720&fit=bounds)

Starting at 12 o'clock, my high spot on the circle was +.0032, which became my new zero. At 180 degrees, I recorded a -.007. Dividing the .007 by 2 and I'm at .0035, well within the .005 tolerance. Or, would that reading be .0032 (new zero) minus .007 and then divide by 2? That would give me .0019. Either way, I think I'm good unless someone sees something troubling, I'll move on. Thanks for the input.

[herb427sc]

Since your "0" point is on the left side, you need to go left .0035 and up .0035.
If you did that the bell housing would be less than .001 all the way around.

[JohnK]

(Deleted my prior post, as I realized I screwed up the math)

Did you zero the gauge at 12 o'clock and then take the other readings, or are they all taken relative to each other? If they're relative to each other (didn't zero the gauge at 12 0'clock) then your offset is:

Vertical (12 o'clock to 6 o'clock): -0.0032 - (-0.0039) = 0.0007 / 2 = 0.00035

Horizontal (3 o'clock to 9 o'clock): 0.0032 - (-0.007) = 0.0102 / 2 = 0.0051

If I'm doing the math right (hopefully others will double-check me as I already screwed it up once) you are nearly perfect vertically and are just about at the allowable max horizontally. A set of 0.007 offset dowels would shift the bell housing to one side and then you'd be within 0.002.

[cv2065]

The gauge was initially zeroed at 12:00, then my high spot was identified (+.0032) and that became my new zero. Then took readings at the new 90, 180 and 270 marks.

[rich grsc]

And that is why you show negative number---zero at the lowest#, then your numbers positive. Much easier to understand

[cv2065]

Thanks for the replies. Based on the responses, there’s a few approaches here to interpreting the numbers. The video simply shows taking the number at 180 from the new zero and dividing it by 2, which in this case is .0035. Which is what I did. I just recorded the 90 and 270 figures for additional reference.

[cv2065]

And that is why you show negative number---zero at the lowest#, then your numbers positive. Much easier to understand

Thanks Rich. I believe the numbers show the bell is within tolerance. Does that look correct?

[JohnK]

Thanks for the replies. Based on the responses, there’s a few approaches here to interpreting the numbers. The video simply shows taking the number at 180 from the new zero and dividing it by 2, which in this case is .0035. Which is what I did. I just recorded the 90 and 270 figures for additional reference.

No, it has to be in spec both vertically and horizontally. If all the other numbers are taken after you zero'ed, then what you have is:

Vertical (12 o'clock to 6 o'clock): -0.0032 - (-0.0039) = 0.0007 / 2 = 0.00035

Horizontal (3 o'clock to 9 o'clock): 0 - (-0.007) = 0.007 / 2 = 0.0035

So you are within spec in both directions.

[cv2065]

Thanks John. That makes sense. Looks like the HF tools did OK to get identical readings in the vertical and horizontal. The linked video and Holley video only speak to the 180 reading after zeroing out the high spot for whatever reason.

[OB6]

If you only measure vertically and horizontally, how do you account for other axis'? I think one of the points of the method in the video from Silver Sport is to eliminate that problem. I used that method because it's simple, but more importantly because they use it as standard procedure, and because I was told Tremec agrees with it.

https://thefactoryfiveforum.com/attachment.php?attachmentid=196914&d=1710620258

Regardless, glad to see you got it figured out!

[herb427sc]

Refer to # 16 and # 22.

Offset dowels (.007) will get you perfectly centered.

# 24 --- -.005 and +.003 = .008/2 = .004

That is near the top of the tolerance for the bearings in the transmission.
You can get a lot closer with offset dowels.

[cv2065]

If you only measure vertically and horizontally, how do you account for other axis'? I think one of the points of the method in the video from Silver Sport is to eliminate that problem. I used that method because it's simple, but more importantly because they use it as standard procedure, and because I was told Tremec agrees with it.

https://thefactoryfiveforum.com/attachment.php?attachmentid=196914&d=1710620258

Regardless, glad to see you got it figured out!

Did you zero your dial out at +.003?

[herb427sc]

You need 4 points contact for concentricity. top, bottom, left and right or 0, 90, 180 & 270.

The hole in the bellhousing is round. You could be in spec top to bottom, but off side to side.

[OB6]

Did you zero your dial out at +.003?

Yes I did, per the instructions.

[OB6]

You need 4 points contact for concentricity. top, bottom, left and right or 0, 90, 180 & 270.

The hole in the bellhousing is round. You could be in spec top to bottom, but off side to side.

Agree, and that's why the method I used makes sense to me. It doesn't care about 0,90,180,270 etc. It only cares about the max offset (positive and negative) and addressing that because you can only move the bellshousing one direction.

[cv2065]

So one follow up question. If you take the bellhousing off, then reinstall, is one suppose to be able to duplicate their readings from the last time the bellhousing was installed? I've pulled the bellhousing and reinstalled 3 different times and the readings are all different. I've made sure that the dial centers itself back to zero after one full revolution each time, and can duplicate the numbers a second time, so it has to be accurate for that particular measurement session.

This is what I got this time, which looks like I'm at .00425 on the 12 to 6 and out of spec on the 3 to 9 verticals:

https://hosting.photobucket.com/images/c310/cv2065/IMG_9448.JPG?width=960&height=720&fit=bounds (https://hosting.photobucket.com/images/c310/cv2065/IMG_9448.JPG?width=960&height=720&fit=bounds)