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.
After a 60 mile run this morning, the steering on my Mk2 got very stiff. When I pulled over, I heard new noises coming from the power steering pump.
For background, the car has XJ6 rack and pinion steering and the original pump driven off the back of the alternator. I replaced the hoses a month or two ago, and refilled the system with Type F ATF, which I saw was recommended on the XKs Unlimited site.
The car was driving beautifully up until the steering got stiff. There are no obvious leaks, and the fluid reservoir is full.
I'm hoping the pump hasn't gone bad, and the use of Type F ATF wasn't a contributing factor.
Any suggestions on diagnostics, or likely cause of the problem?
Sounds like the pump has had it if it's making a lot of noise. Is the noise continuous whether the car is moving or not ? Does it make the noise when the engine is just ticking over when the car is at rest ?
The stiff steering is due to a lack of pump pressure. Is it the original PS pump or a later replacement ?
Unfortunately, the noise is present with the car stationary, engine at idle, and no movement on the wheel. I've only had the car a few months and don't know how old the pump is. The hoses I replaced looked pretty old.
Sounds like I've got a new pump in my future. Any preferred sources that can be recommended?
When was the last time the filter in the reservoir was changed _ assuming that the car has the original reservoir in it.
Also, the noise may be coming from the dog drive that connects the back of the pump to the generator.
This connection is rubber mounted and after long service life the rubber breaks and the drive becomes quite noisy.
It would be unusual for the pump to fail, it's always in oil and always lubricated.
But check those things first before condemning the pump.
I suppose that the internal pressure release valve(s) in the pump may need some new O rings. I would have to check on that, it's been a long time since I looked at a diagram.
I'm betting that the drive dog is shot and the pump isn't even turning _ that would explain the very stiff steering _ it's not even moving any oil.
I don't have a lot of details on the maintenance and restoration history of the car. The original reservoir is there, and the steering rack was installed a few thousand miles ago, so I expect the filter is ok.
Good suggestion regarding the dog drive. That would be consistent with the sudden failure.
No mention of the ATF type, so I'm guessing it's not a likely issue, correct?
Unlikely an oil problem with compatibility. Any "O" rings and hoses that come in contact with the oil will be resistant to anything petroleum based.
Neoprene is pretty tough stuff.
These fancy transmission oils will work just fine here in this case. They are designed for modern transmissions with fragile light weight plastic parts or parts that would otherwise dissolve in the wrong oil.
They also have anti foaming agents in them for that purpose. They are designed for applications under high pressure in a system with small orifices.
The other thing to worry about is how thick the oil will get when it freezes, so your ATF fluid is designed not to get too thick when it does freeze.
It turns out the coupling between the Retroaire alternator and PS pump is badly worn, as shown in the attached picture. The central spine in the coupling is supposed to be continuous. My best guess is that the pump must have been loose at some time in the past, which wore down the spine. Then it was tightened, but with much less metal left to transfer the power. Yesterday the remaining metal let go.
The damaged coupling is a custom fitting welded to the end of the alternator shaft. I'm hoping I can get a good machine shop to weld in new metal to fill the worn gaps in the spine without damaging the alternator.
The mating piece on the PS pump is a little chewed up, but with a little cleaning up looks serviceable.
I'm open to suggestions if any of you have been down this path before.
That looks like an XK modification to the back of their alternator _ according to their stock photo anyway.
If you can't get a jobber to repair the coupler, there has to be a way to replace the coupler. If not for replacement, but for servicing the alternator.
Why would XK's permanently weld the coupler right onto the end of the shaft ? I would ask them and show them your coupler.
How is one supposed to get the alternator apart with that thing welded on the end of the shaft _ has to be a way to get it off.
You will have to remove it for the machine shop to repair it _ one would think...
Also, can you turn the power steering pump easily with your hand _ you should be able to ?
Looking at the stock photo, it may appear that the adapter plate is removed by undoing those recessed allan key hex bolts.
Once that is removed, there may be another single allan key bolt holding the coupler onto the shaft of the alternator.
The XK kit is from Retroaire and I have an inquiry with picture into them, but they're closed until Jan. 5th.
I understand the alternator is adapted from a heavy truck unit, and the adapter is welded to an even larger diameter shaft end. My guess is you can disassemble the alternator in the conventional way.
The PS pump turns easily with no grinding, and just a little end play. No evidence of leaks.
That's good news that the pump turns easily _ as it should.
And the alternator would have to come apart on the drive dog end somehow if one ever needed to replace the sealed bearing on that end.
I guess we'll know for sure after Jan. 5th for sure...
I thought I'd provide an update on how this was resolved.
It turns out that this type of failure is not uncommon on the RetroAire alternators. A friend of mine here in Austin had exactly the same failure on his Mk IX installation, and Rock at RetroAire said that he's seen the same issue on the older design drive couplings.
That same friend offered to machine out the damaged center ridge and TIG weld in new metal. He's got a mill that we used to re-cut the slots to create a new ridge. The picture shows the final product.
I suspect there is slight misalignment in the metal-on-metal contact area in the coupling that wears away the metal over time, so I the slathered the best high pressure grease I have on the contact surfaces before reassembling the PS pump to the alternator. I started up the car this evening, and all seems as it should be, at least for now.
This is an interesting thread to me since I have a similar set up on my Mark 2 altho' as yet all seems well. One can see that if the alignment between pump and alternator is not absolutely perfect there will be sliding on the drive faces leading to fretting and ultimately failure. The hardness of the components will play a part in the life under these conditions. It is essentially impossible to maintain perfect axial and planar alignment so wear and fretting will always occur to a greater or lesser degree.
It sounds like a better engineered solution would be to use an Oldham style coupling (Google it). These industrial drive couplings are designed to accommodate shaft misalignment and have center spiders that are resistant to sliding wear.
The choice of the correct size of coupling is based on the torque requirement and this I do not know ( but it can be roughly calculated)
In any event an industrial coupling will perform better than an ad-hoc thing welded on the back of the alternator shaft.
I'm pretty sure exactly what you describe is the cause of the failure, and the Oldham coupling looks like a great solution. Rock at Retroaire seems like an ernest guy, but basically he's running a cottage business and his products are under continual improvement. I'll pass along this idea to him. Maybe the next generation will use it.
FWIW, nothing is welded to the alternator. The coupling in the pictures is attached (very securely) via a spline shaft fitting.
I'll report back if the fix and heavy grease approach doesn't prove a durable solution.