MG MGA Pros and Cons on 3.9 rear axle conversion BBS discussion at MG-Cars.net

MG MGA - Pros and Cons on 3.9 rear axle conversion

Rich You either have short winters over there or you work slow! I swapped over the diffs (including the splined drive gears) in an afternoon. For those unfamiliar with the process, Barney has a very good illustrated step-by-step article. But I guess changing the thrust washers is a bit longer an operation. Perhaps worth changing the diff front oil seal during the changeover while you still have easy access. I did it after installation, following the discovery of an oil leak, and found it bit of a game lying on my back under the car. Probably better on the workbench provided you can hold the unit to get enough torque on the end nut (about 135/140ft lbs). I made up a tool to hold the UV flange steady so that I could torque the nut - see image. Steve

Steve Gyles

When my 4.3 differential self-destructed (see pictures) I replaced it with an early MGB 3.9, but used MGA axle gears so the splines weren't an issue. The only negative effect I've noticed is that occasionally when I turn onto a road with an uphill grade I need to take it in second instead of third to keep the engine from lugging a little. As far as I can tell, the speedometer isn't any less accurate now than it was before. It seemed like before it was reading a little fast, now it's reading a little slow. Since it's the least important gauge in the car, re-calibrating it hasn't been a big priority. But, it does make the occasional freeway run (I try to avoid freeways when driving it - no fun there) a lot more soothing.

David Breneman

Here's another shot of the destroyed differential. Yeah, that's a chunk of the bearing wedged in there.

David Breneman

Strength of the spline is not a problem. I have never known the straight 10-tooth spline nor any of the finer pitch involute splines to shear (until someone says otherwise). It is the strength of the shaft that is different. The larger straight splines are cut deeper into the shaft. That makes the root diameter smaller, and the shaft with the smaller root diameter will break with less torque. The half shaft fracture usually occurs immediately at the side of the differential sun gear, at the point of interface of the ends of the splines in the sun gear. I have a very good picture of this, see attachment. Any of these shafts and splines have a very large safety factor for the expected load. The shear failure is ultimately the result of a fatigue crack that may develop after many years of service. It is not the engine torque or power that determines the peak load on the drive train, it is the grip of the tires on the pavement. A stock 1500 engine is capable of spinning the tires in 1st gear if you get the flywheel wound up enough. A 150 HP race engine with a competition clutch will spin the tires with exactly the same axle torque, given the same tires and same vehicle weight. What does put more load on the drive axles is using sticky racing tires. If the tires have 50% more grip, and you apply enough input torque to spin the tires, you will put 50% more torque through the half shafts. In most cases this will still not break the halfshafts. I had used sticky race tires for autocross nearly every weekend during the driving season for several years (in the last decade). For a few years I also used a competition pressure plate. That could produce a fair amount of tire spin on the 1-2 up shift, and a little spinning of the tires on a 2-3 up shift with a quick shift under power near engine red line, even with the stock 1500 engine. It is the inertia of the flywheel that contains and releases this energy during a high speed power shift, producing enough torque to spin the tires a bit in 3rd gear (with a strong enough clutch). Autocrossing every weekend may not produce as many track miles as a vintage racer does in a year, but it definitely does produce more jackrabbit starts and likely just as many power shifts. I autocrossed regularly for several years on race tires (maybe 2000 very harsh laps) without ever breaking one of my 10-spline half shafts. Then a few years later one of the half shafts failed during a casual drive home after a casual road rally on street tires with no power shifting going on. This just happened to be time for that fatigue crack to show its nature (see picture). This happened at 340,000 miles on my MGA original rear axle. YMMV. I now keep a spare halfshaft on hand, as I have good reason to expect the other one to fail in due time, having served through the same long term duty cycles. Cars used regularly for competition work are likely to fail halfshafts more often. I do not believe the involute spline shafts are immune to this problem, but they may last a little longer before they fail. The solution for this is to have the halfshafts checked for cracks occasionally as periodic maintenance, using magnaflux or ultrasound or dye penetrant, and replace any part that may have a crack before it fails in service. As a follow up, I would like to know if anyone has had a halfshaft failure after having it checked for cracks.

Barney Gaylord

This thread was discussed between 18/09/2007 and 25/09/2007

MG MGA index

This thread is from the archive. The Live MG MGA BBS is active now.