Subject: Alusil is prob NOT Nikasil
From: Paul
Date: 2/9/2000, 11:36 PM

Bill Freeman wrote:

Paul,

The two systems are very different.  The two "aluminum
wearing surface" processes that I am aware of and that
are in wide use are (1) the Reynolds 510 "process" and
(2) Nikasil.   They have almost nothing in common.

I believe (based on the name and some older info on
what MB was using for their motors) that the alusil is
the Reynolds-developed system which was used in the
very last McLaren G7 cars.  I spoke with the mechs
and they said that cyl wall wear was WAY down over
the cast iron liners.  This system was used on the infamous
Vega linerless aluminum block.  Basically the system uses
a very high silicon content aluminum alloy and the mold
liners for the cyl bores are chilled at pouring.  This causes
the alloy to form a structure with very high silicon granules
surrounded by a matrix of nearly pure aluminum at the cyl bores.
 The rest of the block has equilibrium conditions with the silicon
and aluminum blended and it acts like "normal" aluminum,
i.e. not good for sliding wear.  The cyl bores are machined and
then etched with acid to remove a small amount of the alum
matrix around the silicon wearing particles.  These channels
hold in the oil, which (IIRC) does not wet the silicon nodules.
This system works VERY well, with one SERIOUS problem,
as discovered in the Vega.  If overheated, the alloy becomes
uniform and the cyl wall assumes all the wear resistance of
a Coke can.  Near-instant total destruction.  The only POSSIBLE
fix is to install a ferrous liner, which was commonly done
with the Vegas, and apparently worked fine.

Now - NIKASIL.   This is (I think, please correct me if you
have better info) a heavy nickel plating with silicon carbide
granules embedded into it.  This is somewhat like a diamond
saw, which uses plating (nickel, I think) to bond the diamonds
on.   The surface is ground and the silicon carbide particles
do the wear resistance job and the surrounding nickle matrix
is wetted by the oil.   This is used on Porsche 911 and Stihl
chain saws, plus some motorcycles.  The Porsche 928 (water
cooled) used the Reynolds process (alumisil?), NOT nikasil.

Hope this clears up the confusion.  A quick glance at the BMW
complaints sounds exactly like the old Vega problems.  The
solution for Chevy was to get rid of the marginal radiator and
add an electric fuel pump with a shutoff for high coolant
temp.  The reset was hidden under the hood, so idiots were
(once again) protected from themselves.  Apparently, if you
shut down promptly on overheat, no damage occurred. But,
if you insisted on driving to the next station, your cyls were
toast.

Bill

I would think the same thing could happen to Nickasil. If you heated
it enough underneath the thin coating it would merely scrape off.
>From the looks of it the coating is only thousands of an inch thick.
For durability I still say there is no beating a trough hardened
piece of steel just like  Cont. and Lyc. are using on their
cylinders. When overhaul time comes just re-lap it. 

Also a bonded on mild steel with porous  chrome plate could
be repaired by taking off the chrome, grinding it flat again
and re-chroming. 

Unfortunately this can not be done so easily
with the rotor housings as they would get too big. One might
be able  metal spray or D-gun on steel to build them up
and re-machine them in an NC mill with a good finish and then
re-chrome them. This may become necessary in the future if we
cannot get new rotor housings from Mazda.

Paul Lamar


The Aircraft Rotary Engine Newsletter.              Powered by Linux.

http://home.earthlink.net/~rotaryeng/            http://www.linux.org