When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Well I had hoped not to seek help on this forum before contributing some value, but several days of online research and putting the car on jackstands every evening hasn't resulted in a diagnosis.
Several days ago I swapped a 4.3 Torsen into my '92 with used "one-piece" (no 4-bolt stub) axles and a new driveshaft. Torsen came from a CSP autocross car. Also installed new Xidas, rebuilt calipers, new pads and rotors, rear dust shields removed at the same time.
Drained fluid from diff before installation, appeared nice and clean without metallic sheen or pieces.
Installed diff and axles, set PPF alignment at transmission end, verified that the front joint had plenty of clearance to the PPF. Driving after the swap I noticed a rumble under hard cornering which I feel through the seat, not the steering wheel.
On-road testing:
Investigating, I could induce the rumble swerving down the road at speeds 35+ MPH, volume and frequency increasing with speed.
Noise sounds identical whether loading up the left or right hand side, like the Vrrrrrrp of driving over a rumble strip. Noise and sensation disappear if I disengage the clutch, put the car in neutral, or drive in a straight line.
When taking off from a stop or making slow circles in a parking lot (all permutations of left, right, forward, and reverse) I do not notice anything I would describe as unusual.
Swapped through three sets of tires (admittedly none road force balanced, but have since ruled this out as it disappears with driveline disengaged from the motor)
Jack-stand testing:
Verified all suspension and brake components were properly located and torqued, no rotating components making contact.
Checked for wheel bearing play (hands on tire at 12 and 6), none found.
Verified PPF alignment and driveshaft clearance
Verified plenty of clearance between exhaust and differential.
Driveshaft-Pinion Notes: Isolating donut intact, ~1/8" rotational play, no axial or radial play. Verified that I used the correct fasteners for this connection.
CV Axle notes: No rotational play in axles with car in gear and parking brake applied. Some radial play at differential end, moves ~1/16" inch relative to differential on both sides. No leaks from differential seals, axle boots are intact.
Parking brake off, car in gear, I have around 3/4-1" of rotational play at the tire outer diameter.
Took the car up to 70 mph on the jackstands, definitely vibrates above 60 MPH.
Got under the car with it idling in 3rd gear, could hear an occasional knock or light clunk that did not coincide with every rotation of the wheels. Using a rubber tube as an aural aid I was unable to locate a source for the clunking. Placing the end of the tube on the differential case by the pinion or axles, and on the uprights at the hubs did not find any bad sounds.
Removed the rear wheels, installed a couple lug nuts to hold the rotors in-plane (in hindsight it's probably best to install 4 for balance), had the significant other hold the car at 70 MPH and repeated the rubber tube listening test. No bad sounds found, though in hindsight the wheels should probably be on to increase the drivetrain load and apply some radial force to wheel bearings. Vibration was reduced over prior 70 MPH run but still present.
I believe tires and wheel bearings have been ruled out due to the vibration disappearing when the driveline is disconnected from the motor. I doubt it is the driveshaft as it is a new unit sold by Supermiata (made in USA, balanced from factory).
At this point I suspect either CV axles or differential. I could not locate the intermittent clunk, which I think suggests axles over diff, and CV axles are easier and cheaper to check so I have a remanufactured pair on the way.
Current plan is to swap axles and retest. If still present I will swap the driveshaft over from my other Miata. At that point, if not resolved, I'll probably reinstall the prior VLSD for the season and pull the diff apart to inspect. Very much open to other ideas or tests I should perform.
TL;DR
Vibration while loaded up in a turn at 35+ MPH, disappears when the driveline is disconnected from the engine. Occurs regardless of left or right turn.
Cannot detect any issues when circling parking lot at low speed.
External inspection of components has not identified a culprit - CV axles, wheel bearings, and differential all present fine.
For what it's worth I also swapped a 4.1 Torsen into my '90 a year or two ago and have had no issues, so I do not believe it is user error.
If you've read all this I thank you for your time and offer you a cat.
I have to give this some thought.
Sounds like a bad Torsen though. No NVH when load is disconnected means its drivetrain. Same NVH L & R rules out hubs or axles. Too slow road speed to be driveshaft. That leaves diff. If bearings were bad it would have NVH with our w/o load. Torsen responds to load, thus my guess. Don't know how to isolate or test other than swapping a known good diff.
The way we verify a car has a Torsen is to drive in slow left circle at full lock and listen for tire chattering a bit. Open diffs do not make tire chatter. If diff is bad, you might be able to replicate the NVH at 10mph, full lock circle. LH circle so driver can hear inside wheel better.
Thank you for taking the time to respond. I agree that reproducing the issue both left and right suggests axle(s) is an unlikely culprit, though I hope it is the cause - 4.3 Torsens with rust-free housings don't show up for sale too often in New England! I suppose if it has failed the thing to do would be to put the VLSD back in and replace the torsen unit with an OS Giken when I feel like spending money again.
That said, I do love a good excuse to buy new tools and challenge my knowledge - I guess if I pull the diff apart and the gears themselves are intact a rebuild could be fun.
Quick update - pulled and inspected the axles, did not find any unusual play and they rotate smoothly throughout range of motion. Differential rotates smoothly by hand, did not find anything noteworthy in the fluid after draining.
Swapped the VLSD back in and the vibration is now gone, confirming the issue was in the driveline. Glad to kick this problem down the road and get the car shaken down before New Hampshire Motor Speedway next month.
Miata #2 with a 4.1 Torsen is sitting in the driveway with a no-start condition, once I sort that out I'll be able to swap the driveshaft and axles over to that car so I can rule them out as the source.
Assuming the axles don't cause problems on that car I'll set up a workspace in the project room and open up the 4.3 Torsen.
Quick update - finally got around to opening up the differential. Everything looks great to my untrained eye, side bearings don't exhibit any play.
Bearing preload is within spec (7.3006 & 7.3008), Backlash appears to be close, if not within tolerance (quick setup gave measurements of .0040-.0050)
Would anyone recommend doing anything other than sealing it back up?
I do now have to either inspect the axles more closely, swap them into my winter beater Miata and see if the issue travels with them, or secure another pair and cross my fingers. Hmmmm...
Better dial indicator setup yielding backlash measurements of 0.0060 ± .0005, so I'll be taking a stab at bringing that back into spec (.0036-0.0043 according the the Miata.net Garage Differential R&R write-up).
Since I've got the differential housing removed I've ordered the arm reinforcement plates from Overtime Metalworks and fresh axle seals from Mazda - this car is unlikely to see much more than 200hp but I'm a sucker for scope creep.
03-26-2021, 10:31 AM
Troubleshooting Driveline Vibration While Turning
0
