Driveline Angle Analysis and Results

[RRoller123]

I have finally gotten the Spider back on the road after a long hiatus to upgrade the engine to 9.8 CR, and at the same time, went through nearly every other system and refurbished it. 7 long months. Great to be running again!

I can now report that the relatively small changes I made to the Drive Line angle yielded amazing results. The rear axle and drive is so smooth now as to be nearly unnoticeable. Very much improved. Bottom line was three things: Reduced engine roll to 0.1 degree. The reduction of the delta between the drive angle of the Guibo and the transmission output, and the overall matching of the plate angles of the engine/transmission assembly and the rear pumpkin, as a parallel system.

I have been both too busy and too lazy to make some sketches, but I plan to do it today and post them here to explain what I found and what adjustments I made. Posting this ahead of the drawings will hopefully shame me into finally getting it done.

[18Fiatsandcounting]

As further incentive to continue your noble work (i.e., the drawings), I am interested. I have been working on my two spiders back and forth over the past year, and I've finished the drivetrain on the '71 (except for the transmission which works fine). I have not started the drivetrain on the '69. However, when I get down on the ground and look under the car from the rear towards the front, it appears that the angle of the driveshafts between the two cars is different. It's hard to describe, but the '69 looks more "crooked".

So, please post a sketch and I'll compare it to what I have. Thanks!

-Bryan

[RRoller123]

OK, so here is the first sketch, which illustrates the drive line components and the various available degrees of freedom of movement. Will go through this sketch first, to define the terms and support locations, and then post the second sketch, defining the drive line angles and the goals of any adjustment that might be made, and then a third sketch will show the results for my particular car.

ENGINE, TRANS AND DIFF are obvious.

A = ADJUSTMENT POINTS FOR THE VARIOUS DRIVE LINE COMPONENTS.

A0 = Engine mounts. Very little freedom of movement here of course. Rigid to the crossmember. However, there is some opportunity for ROLL adjustment, which I will discuss later.

A1 = TRANSMISSION MOUNT BRACKET AT REAR OF TRANSMISSION.

A2 = PILLOW BLOCK MOUNTING POINT.

Note that A1 and A2 can be adjusted with hardened steel shims (washers). The support brackets attach to studs that are mounted directly on the body. Now one might worry that this weakens the mounting system, but I do not believe so, because the contact area of large washers between the brackets and the body is not decreased by use of the hardened washers. i.e. The bracket contact area is about the same with or without the washer shims.

G = GUIBO

DS1 = DRIVE SHAFT FRONT SECTION, #1

P.B. = PILLOW BLOCK.

U1 = U JOINT #1, THE FRONT ONE.

DS2 = DRIVE SHAFT REAR SECTION, #2

U2 = U JOINT #2 THE REAR ONE.

NOTES:

So, in essence this is a THREE U JOINT system, becase the Guibo acts as a U Joint, or as a CVJ. I am not sure which. But this is important, will discuss in the next sketch.

These are SIDE views, so the top one shows an exaggerated view where there is an angle passing through the GUIBO, and the DS1 front drive shaft section is tilted down, relative to the Engine and Transmission.

The middle one shows the opposite, with the DS1 section tilted UP relative to the Engine and Transmission.

The Bottom sketch shows the DS1 section exactly in line with the Engine and Transmission.

Also important to note is that the only degree of freedom that depends upon the car being laden is the Differential Up/Down angular movement. All the other attachment and adjustment points are hard mounted to the body and have nothing to do with the front suspension.

[RRoller123]

OK, so why is any of this important?

This is why:

MISMATCHED DRIVE ANGLES THROUGH U JOINTS INDUCE VIBRATION.

Watch both of these:

https://www.youtube.com/watch?v=LCMZz6YhbOQ

https://www.youtube.com/watch?v=gmV4qwLfOMY

The bottom line is that if there is any mismatch between the initial input angle and the output angle, then vibration is induced into the system.

[RRoller123]

So in our cars, we have:

ALPHA 1, which is the angle of the engine and transmission assembly, i.e the crankshaft line. (This is the angle variance from horizontal, as are all of these angles). It can be measured when the head is off by measuring the top face of the block, or when the head is on, by measuring the angle of the front Pulley and calculating offset by 90 degrees..

ALPHA 2 is the drive angle of the front driveshaft section.

ALPHA 3 is the drive angle of the rear drive shaft section.

BETA 1 is the initial Tilt of the system (and relates directly to ALPHA 1) I use it here because it makes it easier when looking at the rear of the system.

BETA 2 is the tilt angle of the U Joint Mounting Plate on the Differential.

What are the adjustment goals here?:

FIRST, there should be as close to ZERO angular change as possible passing through the GUIBO.

SECOND: There should be as close to ZERO Delta as possible between BETA 1 and BETA 2.

THIRD: There should be no more than about 1.5-3.0 Degrees between BETA 1 and BETA 2. I was able to get this far lower.

BRINELL HARDENING: To avoid Brinell hardening of the U Joint rollers, U Joints should operate at an angle of at least 1 or 1.5 degrees or so, minimum. This is easily achieved with the Spider setup. I had measured 5.1 Degrees for DS2 , and 4.0 for DS1, a Delta of 1.1 degree, which is excellent. Plus the DS2 section is normally moving up and down anyway.

The key is to get the Delta through the GUIBO as close to ZERO as possible.

[RRoller123]

So first, knock off the easy ones:

ENGINE YAW: (side to side position of the block):

This is not really adjustable, but it can be accidentally set in great error by reverse mounting the support bar for the transmission. We discussed this in an earlier post. If you set the support bar orientation correctly, then this comes out right, and as designed will be set as well as you can get it anyway.

Engine ROLL: (In relation to the Cross Member)

This can be adjusted, but only very slightly, by influencing where the engine mounts sit on the Cross Member and the engine support arms. I found that just by slowly lowering the short block onto the mounts and fiddling about for only a minute or so, that I could get the Block ROLL down to 0.1 Degree, which is basically nothing. It is difficult to measure the Cross Member angle, but I did it by using a 4 foot steel level set across from bottom left to bottom right, underneath the Cross Member, attached to a piece of hardwood that was cut to fit just exactly between the shock mount points on either side, then compared that measured angle to the angle of the top face of the short block with the head off. The bottom line is that with minimal effort, ROLL can be reduced to basically nothing, e.g. 0.1 degree, which is what I have.

ENGINE PITCH: (This is the tilt up or down from horizontal, as measured by the front of the block tilting up or down along the drive line)

OK so this is the big one. I did not measure these angles with the OLD mounts, but they would be WAY out. We had sag and yaw of over an inch on the worn engine mounts themselves, which would greatly effect the whole system.

I set the car on 4 stands at the farthest corners possible, and filled up the gas tank fully, and put my usual stuff in the trunk to simulate the best laden condition i could get. The Diff travels up and down, so only the average position is really of importance anyway. The Car sat within 0.6 degrees of Horizontal.

With NEW engine mounts, I got a reading for ALPHA 1 of 3.9 degrees, and ALPHA 2 of 4.4 degrees, so there was a 0.5 degree delta through the GUIBO. We want this to be ZERO.

The mounting studs on the chassis are quite long, and allow for well over a half a degree of adjustment by using shims. For example, from zero to three washers changed the angle by about 0.7 degrees.

So in my case:

ALPHA 1 was measured and set to 4.1 degrees, with A0 hard mounted to the chassis (i.e. NO shims/washers)

ALPHA 2 was set to 4.0 degrees, by using just a single shim under A1. (Shim is an enormous hardened washer) So I was able to get the delta through the Guibo down to 0.1 degree, basically nothing. (It is only the Delta that matters here, I would have to adjust all these numbers for the 0.6 degree from horizontal car set up, but it is not important what the raw numbers are, only the delta through the 3 sections).

BETA 1 and BETA 2:

I was able to adjust BETA 1 and BETA 2 to a measured Delta of only 0.1 degree, (85.9 degrees front, and 85.8 degrees at the Diff plate where the rear U Joint mounts). Excellent, and basically nothing.

So how does it run?

I have put about 75 miles on the car and it is WAY noticeably smoother than before. It is so smooth that there is literally no noticeable vibration, or anything coming out of the rear end. I was actually quite amazed! I still have a strong vibration in the front end that I need to track down, probably tire wear and alignment, but the results are amazing regarding how smooth the rear end now is.

This whole study is all in relation to only the drive line angles measured from the side, it doesn't include them measured in terms of the YAW of each component, but given how little adjustability there is for YAW, I don't intend to pay any further attention to it. Time to move on to something else!

[18Fiatsandcounting]

Time to move on to something else!

Excellent work Roller, and I'll have to study this more before I fully digest it. The YouTube videos were good, although they did have a cold war era feel to them of, "engineering is our partner in democracy".

But, I'm still surprised you didn't pick up on my treatise on coolant heat flow of a couple weeks back. I mean, what's a couple differential equations amongst friends?

-Bryan

[SteinOnkel]

Yeah, this is a lot to take in.

I would think that a ton of your adjustments are negated by the absolutely pillowy soft transmission mount. Even with a brand new mount from AR, I can move the transmission 1/2"in any direction with three fingers. Factor in the car actually moving and going over bumps and this thing is all over the place.

I wonder if there are stiffer transmission mounts available.

[aj81spider]

Hi Pete,

That's a great write-up. Thanks. My main question is how you measured the angles. I'd be hard pressed to measure 0.1 degree with a protractor at my desk - so any tips you have on the measurements would be appreciated.

[RRoller123]

Used an HF digital angle gauge. Also really great for setting table saw blase angles, etc!!

https://www.harborfreight.com/digital-a ... 63615.html

It is not important that it be all that accurate, but it is important that it be precise and repeatable. It has a magnetic side, so to check that, put it on one of the drive shaft sections, take it off, walk around the room, checking a bunch of other stuff, shake it up, drop it, whatever, and then go back and recheck the same drive shaft section. I did that a few times and always got the same reading. So it is very repeatable. I will assume that from a precision standpoint, it is fine with the measurement of angles that are of very small differentials, as the same method on multiple locations always yielded the same result. I.E. it is not the actual angular number (accuracy) that is important, it is the Delta that is important, and I think the gauge works very well for that purpose.

[RRoller123]

And as always, it has been proven that the Bumblebee can't possibly fly. And yet it does.

The single most important thing is to maintain a zero delta angle through the Guibo. I.E The drive line goes straight through the Guibo with no angular deviation. This ensures that the Guibo acts as a CVJ, not a U Joint, and takes this 3 U Joint system down to a very easily manageable 2 U Joint system. That should be obvious in reducing vibration. Combine that with matching the input and output plate angles (at the Front Crank Pulley and at the Rear Diff), and the vibrations are reduced to a minimum, regardless of the normal movement of the system. There will of course be movement around the datum, it is the starting point of the datum that determines the degree of vibration induced. The further the system moves from the optimum point, the worse the induced vibration is. I have no idea if this is linear or not, i think it is, but not sure.

Many hot rodders will purposely set the rear Diff plate angle a little low, a deliberate mismatch from the engine drive line angle, so that when the car squats during a run up, the rear axle sets in and the angles match up during the run. Obviously no need for that level of granularity here. LOL.

[RRoller123]

So it's on to the next thing. Since I have no ability to resolve the Covid mess (although my close friends do and are), I am taking Spider study topic suggestions. We covered the various Nitride Hardening techniques, I found that really interesting and there was a lot of new data discovered that disputed the common thinking on it.

I am really interested in the AFM, how it works, how to fiddle with it, its effect on the engine's performance, etc. I think that will be next, unless someone has a better suggestion. This study will require an AFR Gauge, properly installed, so I will work on getting one of those installed and then report back. Any suggestions for a good AFR gauge are appreciated. I think dash mounted for permanent use is the way to go. I have never used one of these myself and know almost nothing about them. Perfect starting point!

[RRoller123]

Maybe a better topic for the next study would be the dialing in of the cams, for me at least, since I am in the process of doing this. We have the unique ability to have 9 full degrees of freedom in cam adjustment, where as a single cam only has 3 degrees. LSA is fixed in a single cam system. There is a surprisingly small amount of info available on the forum on the effects of varying each cam individually from its factory setting. Even in the general literature, you have to dig pretty deeply to get any general info on what results generally to expect if you change each one independently. So this might be a good topic for the next study, and leave the AFM to later.

[aj81spider]

Thanks Pete,

That's a great tool - I bought one. At a minimum it will save lots of time fiddling with squares to get my table saw blade vertical!

[18Fiatsandcounting]

Maybe a better topic for the next study would be the dialing in of the cams

I'm game, and this might be better as a new thread, but here goes. I "sorta" did a similar study with my '69 spider back in the early 1980s, but since I didn't have adjustable cams, I just moved the cam pulleys one tooth at a time (either advance or retard), and drove it around to see how it felt with that setting. A lot of work actually, and my only assessment was qualitative in terms of how the car drove. No dyno tests or anything like that. The bottom line is that in some cases the low end torque improved, while in other cases, the upper rpm range seemed more powerful. However, in terms of balance across the whole rpm range, the best setting seemed to be.... Wait for it... The factory setting. Perhaps not surprising.

I tried 9 different combinations:
Exhaust retarded by one tooth: intake retarded one tooth, then intake on factory, then intake advanced one tooth.
Exhaust at factory: intake retarded one tooth, then intake on factory, then intake advanced one tooth.
Exhaust advanced by one tooth: intake retarded one tooth, then intake on factory, then intake advanced one tooth.

-Bryan