Improvising a power steering rack seal removal tool

FliesLikeABrick

Wastestream salvage addict
H-M Lifetime Diamond Member
Joined
Oct 30, 2019
Messages
225
Hi all,

I am working on another mid-90s Ford Ranger at home. This one has a power steering rack leaking out the driver's side (and thinning/washing out the grease in the rack+pinion), meaning a failing seal that separates the hydraulic side of the rack from the mechanical side of the rack.

Unfortunately this seal is buried right in the middle of the housing (approximately 12 inches from each end) , so it is not easy to grab with a pick, pliers, chisel, or anything else. Further, it is installed against the shoulder formed by a reduction in diameter. This shoulder is large enough that only the inner lip of the seal is "visible" from the rear, nothing substantial to grab for removal.

I don't have any extra-extra-long picks, so I made one. Using a ~20" long piece of 1/2" round aluminum stock, and a small piece of spring steel from an old windshield wiper, I made this:
PXL_20211207_225429309.jpg

and tried the spring steel in two positions; the one shown above as well as 90 degrees - more like a knife where it could bite into the rubber.

This did not work, because the meatier part of the seal was protected by the reduced diameter/shoulder behind it. All this improvised pick was doing was tearing up the inner lip of the seal. No matter how many times I tried, I could not get the pokey bit of the pick to get purchase between the seal and the shoulder, it was just too tight. I also was concerned that the seal might be metal-sleeved, in which case it would have more holding force than I could overcome with this solution.

I was thinking about how professional rebuilders might remove this seal, and was picturing some kind of tool that could expand inside the seal and grab it. Hydraulic, pneumatic, or screw-driven mechanisms would be too involved to throw together in the home shop for this one-time use, and may not even be workable for me since I didn't have the exact dimensions of what I was facing.

Unfortunately I don't have pictures of the full fabrication process since I was making this up as I went, but here was the idea, with some pictures below:
- The head of the tool should have 3 teeth that can cam outwards and bite the metal
- They would need to have an outer diameter under 31mm when folded down
- and an outer diameter ideally around (or over) the outside diameter of the seal when biting and extending outwards
- They need to move easily enough that tickling the rubber with them lets them start to get purchase so they want to bite in further

I did not lay out the shape of the teeth with the min/max OD in mind, and kinda winged it.

Manufacture consisted of (shown below, where I had pictures)
1) Find a ~1" piece of steel in the scrap bin
1.5) Find some metal for the teeth - found a little 3x3 piece of 1/8" plate
2) Cut a groove for the teeth, roughly 1/8" wide
3) Drill holes for roll pins as pivots for the teeth
4) Drill and tap the back of this "head" for 3/8-16 so that it can be attached to a handle
5) If this all looks promising, add a T-handle at the end for power and control

First I cut the groove on the lathe. In the picture below the slot is concealed just inside the collet.
The jaw layout wasn't perfect, I did this vertically in the drill press using a 6-sided collet block, and didn't put enough care into maintaining the same position in the vise. They are roughly 120 degrees apart...
PXL_20211208_151537928.jpg

With 3/4" long, 3/32 roll pins installed, I put in the first set of jaws (they ended up being a bit too short). To make the jaws I cut the 1/8" plate into pieces on the bandsaw; drilled a hole for the pivot off-center at the end; and rounded off that corner with a file so that it could fold down without hitting the back of the groove.

PXL_20211208_195409383.MP.jpg

Fortunately I did think to under-drill the holes for the roll pins and/or not over-drive them so that they could be removed - since I did need to remove the jaws for tweaking and putting in a longer set later on.

PXL_20211208_195416652.MP.jpg

With the longer jaws, this was looking promising. The jaws do engage once I tickle the ID of the seal with them, and they do want to bite in more. I realized that turning the end with a pipe wrench won't give me the control I want (modulated twisting; and pulling at the same time), so I added a T-handle to the back.

PXL_20211208_212824009.jpg

I drilled a 5/8 hole in a piece of angle iron to act as a handle, and welded it to the end of the steel shaft. With a few attempts, and finally remembering to try pulling as I rotate (after getting purchase) -- the seal did come out.


PXL_20211208_221158674.jpg


PXL_20211208_221931739.jpg

Hopefully I can get the right seal kit and clean up the inside of the rack well enough for a successful rebuild from here. I doubt I'll need this tool again, but it can't hurt to unscrew the business end and keep it around; the round stock and angle iron can get reused for something if needed.

If anyone knows what the right tool is for the job, or what professional rebuilders use for removing such seals installed against a shoulder (effectively a blind hole in terms of getting useful purchase on the seal for removal) -- I would love to hear it.

Thanks for reading
 
Last edited:
I do blind seals all the time and what I use is a slide hammer with expandable arbors for different size races and seals. OTC is the brand.
 
The vast majority of shops probably don't try to repair these- they just replace the whole rack. Sadly, it's reality nowadays
Even if the factory provided a special tool for this (which I'm guessing they don't) there's the liability issue and the time involved
-M
 
I do blind seals all the time and what I use is a slide hammer with expandable arbors for different size races and seals. OTC is the brand.
Yeah, that crossed my mind and I almost grabbed my slide hammer to try something like this -- how does the arbor get expanded to grab the seal (which kind of mechanism is used)?
 
Last edited:
It's expanded by the threaded end of the puller. Like a collet in reverse. Mike
 
I like your tool. Must have been very satisfying when it worked.
After the 2-3 evening tinkering sessions of not being able to meaningfully reach/grab this seal -- yes I was pretty happy to get it out without doing something drastic. I spent less time making the tool than I did fussing with other solutions beforehand, maybe that works out to meaning it was time well-spent
 
Last edited:
Follow-up to the seal removal -- installing a new seal turned out to be a bit challenging as well. At first I thought I could press or tap it in from the top with a large socket and extension. However, the play between the extension and socket allowed the seal to keep tilting to the side instead of driving in straight. Additionally, it is a snug fit -- tapping the extension with a hammer caused the inner portion of the seal to fall out.

I ended up turning a small aluminum seal installation tool that I could draw inward with 3/8 threaded rod that I had on hand. The ID was slightly under the rack's shaft size but large enough to hold the seal. The OD was approximately .005" under the seal's OD, so that the tool wouldn't interfere with the housing's ID as it draws in.

PXL_20211212_193029416.MP.jpg

One thing I might have done differently if I did it again is to make the tool a bit longer and leave a shoulder at the back with an even larger diameter with a clearance fit inside the main bore of the housing. This could provide extra insurance that it is set up straight. This crossed my mind a few times while turning, after I selected this short offcut from my bin. However I rationalized that the business end of the threaded rod is over 12" long, so as long as I keep that near-centered in the bore as I draw the seal in, it will pull the seal in straight and hopefully self-center it.

PXL_20211212_193202775.jpg


Good fit, I'm glad the smaller diameter is under-sized because it looks like my insert had a chip and the surface finish on the aluminum had a slight texture. I wouldn't want it to be a tighter fit and risk damaging the inner lip of the seal.

PXL_20211212_193211702.jpg


I used a ratchet to draw the seal in, and felt a clear change in resistance once the seal was fully seated.

PXL_20211212_194155942.MP.jpg


I cleaned the bore thoroughly before install. There are a few little scuffs from my attempt to remove the original seal, but they should all be on non-critical faces.

PXL_20211212_203005021.jpg
 
Last edited:
Back
Top