Power Steering Pump Upgrade
by David Lane
dlane@peabody.jhu.edu
Sunday, 11 November 2001
The following is a discussion of issues and solutions to 1st gen Power
Steering problems. I hope you find it interesting-- even if you don't
have PS on your car.
Those who bought 1st gens new found them to be exceptionally
reliable--much like today's Miata. However, compared to the Miata,
the 1st gen RX-7 has more "GT" in its DNA, making it more suitable for
long commutes and long distance traveling. Thus, low mileage examples
are rare, and those who buy the cars used must deal with age and "use"
related problems.
(Sounds like something my doctor would say, as he explains why the most
recent proof that I am no longer a teenager is going to hurt for awhile,
and then--maybe after a year--will go away.)
One of the less known, but still irritating problems with 1st gens is the
power steering. Not many of the cars came equipped that way. I know it
was an option on GSL-SEs-- possibly bundled with leather seats. There are
two major parts involved, the steering box, and the pump. Seals in both
wear out, and the connecting hoses also eventually leak at their ends.
The result is slowly dripping ATF--usually first noticed by a few drops
of red stuff on the ground under the car, or hanging off the bottom.
Donnie Peters, at Sun Auto (a mostly RX-7 "used parts emporium" and
service facility near York, PA), put a new seal kit in my PS box.
It has held nicely. The situation with the pump is different. I have
had two seal kits put into it over the past two years, and neither has
lasted very long. I am told by my local mechanic that the end play in
the pump's rotating shaft is controlled by a simple circlip ("snap ring"
in the shop manual). He says he has seen those things fail in this
kind of pump--releasing the shaft (with the pulley on the end) to go
floating forward. Donnie says the particular pumps in 1st gens don't
have a record of failing that way, but both agree that it is a bad design.
I don't even want to think about it.
Once worn, it is logical to guess that a little extra play in the shaft,
combined with wear, will create a situation where the red stuff can
again seep past the oil seal.
Pump replacement is the logical choice, but last time I checked into it,
the cost was more than a set of aftermarket wheels. I don't know if
what Mazda is selling is new, or factory rebuilt (like the last starter
I bought from a dealer), but I didn't want to spend that kind of money
and still end up with the same problem down the road.
The other obvious solution is to swap out the power steering for manual
steering. Donnie has done this operation--a labor intensive process
involving a different steering column. It is at least twice as costly
as replacing the pump, and you lose your power steering.
The PS system on 1st generation cars is a good one-- sensitive to
vehicle speed, and with a quicker steering ratio than the manual system.
I wanted to keep it.
Mazda went to a more robust pump on the 2nd gen cars--a better engineered
pump with no reliability issues. Donnie proposed adapting one of those
pumps for the GSL-SE. Some of the 2-gen PS systems were "engine" speed
sensitive as opposed to "vehicle" speed sensitive. Most, if not all
of the later cars ('89-'91) were back to "vehicle" speed sensitive,
so that is the pump to use.
Transplanting the pump was not exactly plug and play. The hole through
which the pump protrudes in the mounting frame needed to be larger.
The pump itself is best modified to remove no-longer-used mounting
brackets--to bend the filler neck so the dip stick will be accessible
(if the receiving car has A/C) and to redirect the hose barb to work with
the parts of the GSL-SE power steering system. A proper connecting hose
must also be made to work with existing fittings.
The worst part of the project was finding a way to get the GSL-SE
two-wire pressure switch to mate to the 2-gen pump. This switch is
the only electrical link between the pump and the rest of the system.
The sealing technology is different, and later 2-gen pumps used a one
wire switch that (by experiment) did not work with the GSL-SE electronics.
However, early 2-gens did use a 2-wire switch, which (of course) was not
physically compatible with the later 2-gen pump. In the end, Donnie
had to mate parts from two second gen pumps (one early, and one late)
to get it to work. Also, they had to put the proper GSL-SE connecting
plug on the end of the 2-gen pressure switch wire.
It is amazing to do this kind of thing at Donnie's place. He has been
dismantling wrecked RX-7s for many years, and has a barn full of used
parts. I can't imagine a Mazda dealer saying, "Let's go in the back
and grab another pump from the earlier cars and see if we can't make
something up." By the time the job was done to Donnie's satisfaction,
we had gone through four PS pumps.
Results are good. The system feels just like it did when new, and for
the first time in what seems like years, I am not trying to figure out
where the red stuff is coming from. The only difference in practice is
that the later pump uses Dextron II fluid, while the earlier one used
"ATF Type F."
I told Donnie others might be interested in this upgrade. He says he can
produce a kit using appropriate parts from two used pumps. There will
also be a connecting hose, modified mounting frame, and pressure switch.
Price for the kit would be $450, and Donnie guarantees all of his
used parts. I figure a good mechanic can install the kit in under
two hours. Shade tree types like me would probably take half a day of
head scratching, goofing off, cursing, and cleaning up spilled red stuff.
For what it's worth, Donnie has worked up fixes for other RX- 7 problems,
like leaking front mounted oil coolers, bad 2-gen logicons, and switch
sets. Contact info is at:
http://www.sunautorx7.com/
Addendum: 12 November 2001
I just got a note from Donnie, who read my post about upgrading the
PS pump. He said:
"For the record, I agree with your local mechanic that the
original pump clip often DOES become too small to hold the
shaft in, and the shaft moves forward with catastrophic
results."
That's an important bit of infomation. I'm glad he wrote to me with
the correction.
Best wishes,
David Lane
dlane@peabody.jhu.edu
'85 GSL-SE (Cartech Turbo) http://www.wankel.net/DavidLane/
|