A few months back Brad posted questions regarding perceived high vibration in his car,  If I recall the details, it has a 347 stroker and a five-speed conversion and he bought the car with these mods already in place.  In the background I have been working with Brad to see if we could find the potential cause/issue.  There are some unknowns that complicate matters so it is difficult to nail this exactly.  For example, the stroker motor was built by a shop that had some significant quality issues before they went belly up so the actual balance quality of the rotating and reciprocating parts cannot be ensured.  There is a balance sheet that the original purchaser received but there is no way to know if the numbers are accurate or "pencil whipped".  The balance quality of the flywheel and pressure plate is also not known.

However, all that said, in order to troubleshoot a vibration issue you need data.  Seat of the pants and assumptions will lead to tail-chasing and the replacement of parts that are often not related to the problem.  I found an older vibration tool in my pile of equipment that is capable of collecting high quality vibration spectra,  The unit provides accurate vibration frequencies and amplitudes.  I sent this off to Brad along with some recommended settings so we could collect appropriate readings.  The instrument was obsoleted a number of years ago but with persistence I was finally able to snag a copy of the supporting software so I could more fully analyze the data.  Brad collected data at several locations including the shifter, dash, and on the engine block itself.  The vibration that Brad felt while driving and and sitting still in neutral increases with engine speed reaching a peak around 4250 RPM and it actually falls a little beyond that.  This tends to eliminate simple unbalance.  Unbalance forces increase with the square of the speed - double the speed, four times the force.

To see what was changing with the speed, readings were collected at 2000, 3000, and 4250 RPM.  The readings were collected in units of velocity (inches per second) which is a common measurement unit used to evaluate overall rotating equipment vibration performance.  In cases where there are no natural frequencies, the vibration at the shaft speed will increase in a linear fashion in velocity.  in a reciprocating engine some of the vibration will increase at a slightly greater rate.  For example, engine firing forces will increase even when under no-load conditions because of increased bearing friction as speed increases.  Very large increases in vibration with increased speed, on the other hand, are not normal.

The vibration readings taken on the engine block in the horizontal direction show a fairly uniform increase consistent with speed.  The levels are not what I would consider to be precision but I am used to working with large industrial machines more than car engines.  I hope to take some readings on a neighbors small block when time permits to compare with Brad's engine levels.

The readings on the shifter, also in the horizontal direction, paint a different picture.  The first reading shown below was taken at 2000 RPM.  I have labeled the running speed (engine rpm) and 2X running speed.  2X running speed can have a number of sources including a bent shaft or axial flywheel runout but the most common source would be reciprocating forces.  For every revolution of the shaft there are two reciprocating forces - the piston moving up to the top of the bore and then returning to the bottom.  The reciprocating forces in some cases will cancel when the assembly is properly balanced and in some engines such as in-line 4's they never will.  This is why you will see "balance shafts" in some of these engines - especially those running at high speeds.  As can be seen in the following plot, the 2X vibration level at 2000 rpm is actually fairly low.  To the hand, which is a fairly good vibration instrument, the vibration would feel low or normal.



At 3000 rpm, the vibration increased slightly as expected.  At 4250 rpm, however, the vibration is dramatically higher.  With the older software I was not able to plot the two readings on the same vertical scale but I was able to get them close.  Visually, it is easy to see that the vibration at 1X (engine speed) is larger but the vibration at 2X is huge.  A basic calculation of the overall vibration puts it above 0.9 inches per second.  A very smooth 3600 rpm pump in a refinery might run at .02 or ,03, a common alarm for industrial equipment is .3, so .9 represents a very high amplitude which is easily felt with hands.  By the way, there are some peaks that show up a 1/2 times, 1-1/2 times rpm, etc.  These are usually fractional subharmonics that are common on reciprocating machines and at this point are not cause for concern.



The very large increase cannot be explained by normal vibration forces.  There is a strong indication that there is a resonance issue where there is a natural frequency of the engine/transmission assembly on its mounts near 4250 rpm.  It takes two conditions to produce resonance - a force and a natural frequency.  The engine cannot be eliminated completely because of the unknowns listed above.  Since we are not going to tear apart the engine then the only other option at this point is to attempt to alter the natural frequency.

So, Brad, when you pull the car out of winter storage, here are a few tests.  This is troubleshooting and without much more sophisticated tools the only option is to make calculated changes and observe.

1. Remove the rear transmission mount (not the cross member) and stack some rags in place of the mount to replicate the height of the mount.  Start the engine and test (by feel) the vibration at 2000, 3000, and 4250 rpm and report the results.  Obviously, driving tests are not recommended with no mount.

2. Make a wooden block and put it in place of the mount.  You can probably just use longer bolts on the mount to transmission and lag screws from the cross member to the mount.  Repeat the tests conducted in 1.  As long as you don't dump the clutch you can also perform a quick road test.  Wood will have enough "give" so that you are not producing loads on the transmission case that a solid mount would.  With the wood mount the the noise level inside the car will be higher but focus on the shifter.  A calibrated finger or hand vibration level is sufficient to compare as-is, to no mount, to wooden mount.

3.  If not too difficult remove the shifter handle and repeat the rpm tests on the stub shaft on top of the transmission in the horizontal direction with the current mount.  Though not high on my list, the combined mass and stiffness of the shifter system may have an rpm sensitivity and that the mount system is fine.  You won't know unless you test.

In test 1, the natural frequency is reduced which should move any problem vibration to a lower rpm..  in test 2 the natural frequency is increased and the problem rpm should also rise.

If there is no change with either of these tests it likely indicates that there is something else causing the increased vibration such as the exhaust system becoming a tuning fork.  Test and report the results before moving to other testing or components.

The apparent system sensitivity to 2X rpm also indicates that the driveline could also contribute to the problem.  In 4th gear at 1:1, a driveline misalignment due to unequal u-joint angles will excite torsional oscillations of the driveshaft that are not properly cancelled.  This torsional vibration will also be a 2X engine speed (again when in 4th).   If I recall, I believe that you have messed with driveline angles and that the vibration did improve slightly.

Last, if a problem with the rear mounts is identified, simply changing the mount to poly or rubber or some combination may not be the solution.  Many of the aftermarket cross members that are used for conversions do not properly support the transmission.  If you read the Tremec installation instructions they clearly state that a cantilevered mount should not be used.  In some cases you may get away with it and in other cases the loading into the cross member will be in torsion (twisting.bending moment) rather than a nice straight load.  Between two  older cars with unknown histories and modifications you may find that the same setup on one car may work well and on another will shake like crazy.  For my TKO conversion I moved the entire cross member back to center it under the mount.  Probably retentive but Murphy and I don't get along well so I try to prevent issues in advance.