PCFree, how did you know what shim to size to use. Those spacers are internal when the hub is installed on the car. How did you measure?
There's a Ford SST used to torque the bearings to spec prior to pressing the hub in - once torqued you use a in-lb to measure how much resistance there is. You match the resistance you measured to the chart in the FSM and you can tell if you're within spec, or need to go up/down in spacer side.
Not sure where my old reply or edits went, but here goes:
The outer seal is not installed backwards, and it did seat most of the way once the hub was pressed in. I've found a much better method of installing the hub that seats everything nicely, and the next time I have to do one I'll document things & take new pictures. No-one has offered up any pictures or write-ups of their own, so these will stay for the time being - it at least shows the basic process & how everything goes together, which is a source of much confusion with the Capri.
Now - the spacer is MATCHED TO THE KNUCKLE, folks. The hubs are all the same, the new bearings should be close to each other - there is a very fine tolerance on bearings, it should not change the spacer size much or all all.
Why? Here's my educated guess/theory/whatever. There is a ridge inside the hub, where the bearing races seat opposing each other. The knuckle is an iron casting or forging - and the most difficult to machine. The thickness of this ridge inside the bore determines how far apart the bearings sit. This seems to vary a bit from knuckle to knuckle - and
directly corresponds to which size preload spacer is needed (presumably because of the fixturing/steps/difficulty in machining an odd piece like this)
The ridge is visible in this picture, below the race:
![](http://www.werbatfik.com/img/bearings/bearings14.jpg)