'It is on the face of it a bizarre specification, proportionally the same as defining a linear measurement to be 1/4 of an inch, plus or minus 1 inch.'

BW - isn't that the exact specification for panel gaps on Defender and Disco 1 & 2?

Keith

It should be straight but given that the factory can't build anything straight and haven't been able to for the last 70 odd years, 6mm is probably pretty good  ---
2006 Defender 110 SW 300Tdi • 2011 Ford Ranger XLT crewcab • 2015 Mitsubishi Shogun SG2 • 2015 Defender 110 Station Wagon Utility TDCi

With a tolerance of up to -55' of camber, that means that on a standard 235/85 (805mm or 31.7" diameter) tire you can have as much as 12.9mm of camber, just over 1/2inch between top and bottom of the tire.

But LandRover also specify 1degree as cross camber. So if axle is symmetrical, then should only allow up to -30' per side, although it could also be -5' one side and -55' the other side. But that would probably not look right (unless it is going for Nascar).

Curious to know if the axle flange, where the stub axle is bolted to, is machined before or after it is welded to the axle. 1988 90 Hard Top, 19J Diesel Turbo, Shire Blue - Restoration ongoing
2012 90 CSW, 2.2TDCI, Santorini Black

I missed the cross-camber figure. You could certainly have an axle that met the spec in the table but looked extremely odd! They are very strange figures, for sure.

It would be interesting to know how the axles are manufactured and the sequence of operations (also who actually makes them, since I imagine the axle cases or even assembled axles were bought in to the production line as finished units).

The same issues that the rear hub flanges display logically should also apply to the swivel housing mounting flanges on the front axle (the ones that famously break off the axle tube), I wonder what the tolerances on those are.

Machining all three faces of the axle (the two flanges and the diff mounting face) would be easy enough with bespoke tooling but not easy otherwise, since all three are critical regarding their relative positions to each other, and the radial positioning of the pinion housing mounting bolt holes is also significant.

I dare say based on those values that the end flanges are pre-machined and then welded up in a jig but could be wrong. Not sure how you'd manage that with the diff mating face though given the thickness of it. It's a strange design really.
Having spoken to a few land rover oracles locally, the most common response was 'I've never had any reason to check the camber, they're probably all like that'. You'd think you would notice 12mm of camber though!

I got my uni assignments out of the way last night so I'll see how I get on in the workshop tonight and report back. 1986 127 V8 Build Thread
Series 1 80"
Series 1 88" 4x2 V8 Build Thread
Instagram - @byjoshsansom

I would not be surprised if there is no further machining at all after the welding process.
I was in the belief that the older axles had the end flanges friction welded, but is more of an assumption than a confirmation.

What I can see on my 2012 front axle is that the end flanges seem to have been welded with a more conventional process. I see the bead starts and ends towards the front of the axle (3 or 9 o’clock position depending from which side you look from). I think this gives more control on camber than toe if parts are jig welded. Toe can be easier corrected by steering components.

I cannot see how the rear flanges are built on the rear on disc brake axles. But my 1988 drum brakes seem good quality, possibly friction welded flanges.

But have no idea who makes the axles, or how they are really fabricated. 1988 90 Hard Top, 19J Diesel Turbo, Shire Blue - Restoration ongoing
2012 90 CSW, 2.2TDCI, Santorini Black

Earlier axles are certainly friction welded in my experience, I think the issue with the mating flanges snapping off is limited to the more conventionally welded axles?

The rear axle in question (post 2002 110/130) has what look to be cast mating flanges which are then I presume MIG welded onto the the two C sections of the axle casing. I've not closely inspected a 90/Disco rear axle but I recall Salisbury axles having a similar arrangement to the late 110/130 puma axles.

Maybe someone can shed some light on it for us! 1986 127 V8 Build Thread
Series 1 80"
Series 1 88" 4x2 V8 Build Thread
Instagram - @byjoshsansom

Never owned or worked on a 110, so my experience stops on a 90.

But I would definitely see 12mm between top and bottom of the wheel, especially on a Defender with its vertical sides as reference.

Interestingly, most Series vehicles that I see have positive camber on the front. Possibly to help with steering effort. No ide where they got the geometry from, but probably the king pin angle / position in the swivel housing. 1988 90 Hard Top, 19J Diesel Turbo, Shire Blue - Restoration ongoing
2012 90 CSW, 2.2TDCI, Santorini Black

So this evening I bolted the diff in place and dropped in a halfshaft. With the halfshaft vertical and the O/S mating flange on a level surface it was evident that the long side was at fault.


Click image to enlarge

I then shimmed the O/S flange to level up the N/S flange and verified that the N/S 'tube' was still straight. Thankfully it was which means I did a good job of keeping the heat down whilst welding up the truss.

Next step was to slice through the truss, a shame but it will hopefully allow me to straighten the casing without inducing tensile stress in the truss. Thankfully I added two drainage slots to the truss which happened to be perfectly in line with the point of max bending moment (immediately adjacent to the pumpkin).


Click image to enlarge

Underside cleaned up ready for welding, marks from the forming process conveniently mark centre which is handy.


Click image to enlarge 1986 127 V8 Build Thread
Series 1 80"
Series 1 88" 4x2 V8 Build Thread
Instagram - @byjoshsansom

An interesting read; https://www.boconline.co.uk/en/images/Fund...113398.pdf

I don't have access to oxy-acetylene unfortunately but the principal is similar. This section in particular got me thinking;


Click image to enlarge

Would I be better off welding longitudinally down the axis of the axle (red) or, transversely (green), around the casing as it were?


Click image to enlarge

I suspect longitudinally will be the most effective and it is the least likely to impact the toe-in/out of the casing. Any thoughts? 1986 127 V8 Build Thread
Series 1 80"
Series 1 88" 4x2 V8 Build Thread
Instagram - @byjoshsansom

Good find there.

Indeed as metal is heated, it expands, but also important reduced its yield strength. So during heating, if it expands such that that area goes beyond its yield strength, then the steel is permanent deformed. When it cools it contracts and pulls the surrounding material.

In my opinion a transverse weld has more chance to be effective to correct castor. But as you say, it might affect toe as well as you need to start from a point, and finish at the end point x seconds later.

The other option that you may consider is to make multiple longitudinal welds, perhaps shorter, and work the welds 'out' Example weld longitudinal at 12 o'clock position. Then another at 1 o'clock, another at 11o'clock, then 2 o'clock, 10 o'clock.. well you get what I mean. Actually might also make more sense to work your way from the outside to the top, so you build heat progressively and concentrate the heat at the top. 1988 90 Hard Top, 19J Diesel Turbo, Shire Blue - Restoration ongoing
2012 90 CSW, 2.2TDCI, Santorini Black

Something like this:


Click image to enlarge 1988 90 Hard Top, 19J Diesel Turbo, Shire Blue - Restoration ongoing
2012 90 CSW, 2.2TDCI, Santorini Black

Yes, whilst I have a good theoretical knowledge of the properties and behaviour of steel as a civil (structural) engineer, my practical experience is lacking somewhat other than that gained through welding. The article gives optimal temp. values for heat straightening based on the material properties and makes reference to watching the heat affected zone. Something I don't think I'd be comfortable attempting without the supervision of someone more skilled in this area.

Agreed re welding transversely, I suppose I could mark start and stop points to ensure the weld is 'centred' as it were, though the heat build up will be greater by the end point than it was at the start point...... I like the longitudinal methodology you suggest, either way I intend to let the weld cool fully (without any forced cooling), re-measure then weld again as required to hopefully prevent the casing being pulled too far. I'm conscious that too much heat could actually give me positive camber! 1986 127 V8 Build Thread
Series 1 80"
Series 1 88" 4x2 V8 Build Thread
Instagram - @byjoshsansom

The problem with welding approach is that you have less control than if it was heating with a torch.
Typically heating with a torch would be done as you say, heat cool and check. If not yet within spec, heat a bit more, cool and re check.
With welding this is a bit different. With a more concentrated heat zone, it can alter the grain size around the welds, typically making them smaller. This will affect the properties around the weld. With torch heat, and slow cooling the opposite will happen. 1988 90 Hard Top, 19J Diesel Turbo, Shire Blue - Restoration ongoing
2012 90 CSW, 2.2TDCI, Santorini Black

Nice video. Seems quite successful. 1988 90 Hard Top, 19J Diesel Turbo, Shire Blue - Restoration ongoing
2012 90 CSW, 2.2TDCI, Santorini Black