Date: Fri, 24 Sep 1999 07:33:45 -0700
Reply-To: Derek Drew <drew@INTERPORT.NET>
Sender: Vanagon Mailing List <vanagon@gerry.vanagon.com>
From: Derek Drew <drew@INTERPORT.NET>
Subject: German Trans Responds On Viscous Matters
Content-Type: text/plain; charset="us-ascii"
This posting is a series of emails between Forrest at German Transaxle and
myself on the issue of how to properly rebuild Viscous Couplings.
This series begins with a response to me from Forrest to a posting I made
about two months ago suggesting that there were questions about the the
rebuilding techniques being employed by rebuilders based in North America.
There is then a response from me to Forrest, and a response back.
Reproduced at the very end is a copy of the original posting that I made
that prompted Forrest to write to me initially.
All of this material except the material at the end is newly posted to the
Vanagon List.
----------------------------------------------------------------------------
-------
Derek,
Chris at Weddle Engineering snagged your vc article from the web, and
faxed it to me. Here's a copy of my reply to him:
> Chris,
> Got your fax about the viscous couplings. Don’t know who their expert is,
but when we’ve removed the little access screws from brand-new ones, we
have not seen any air in them, certainly not 14% of the volume of the unit.
Perhaps domestic or Japanese vc’s are made differently from the VW ones; I
dunno. Also, we can’t figure out how the ‘microscopic burrs’, if they in
fact exist, would wear out, as the plates do not touch one another. The
plates always turn >with< one another except for the few seconds or minutes
a year that there is slippage between the front and rear wheels; then there
is slippage between the plates, but the plates only rub against the
silicone fluid, which is pretty benign stuff, not against one another.
>
> As far as the ‘microscopic burrs’ wearing off and turning the fluid
black, the guys in the shop burst our laughing when I told them about that
theory. When the fluid is black, the whole shop stinks when the unit is
disassembled. It’s quite obvious to anyone with a nose that the black-fluid
units have grossly overheated and burned the fluid.
>
> When we decided to rebuild these units we sent a sample of new fluid to
Dow-Corning. They matched the VW fluid the best they could to their own,
then, when it seemed like our units were a bit tight, sent us some thinning
fluid and instructions/formulas how to adjust viscosity. Right now we’ve
got about 80 units in the field, and since we lightened up a bit I don’t
think we’ve had any complaints at all. As to the price we charge for the
fluid, a drum of the stuff cost us close to $3,000, and Dow made it sound
like they may or may not ever sell to us again, as they normally sell it by
the train-car direct to manufacturers; they acted like they were doing us a
big favor selling such a small amount, so we’re reluctant to re-sell it at
all.
>
> Cheers...
>
> Forrest
----------------------------------------------------------------------------
----------
Derek's Response:
I am not an expert. I am simply reporting what the licensor and
manufacturer told me. I do have some clarifications to offer in this email,
however.
1. Both GKN and Styer Daimler Punch say that the proper % air is critical
and there is a very small margin of error here. An engineering article from
the fellow Wolfgang when he began designing this syncro viscous coupling in
1979-1980 seems to confirm this.
2. Both GKN and SDP say you will not see this air on disassembly because
the air becomes mixed into the fluid and lodged in it or in the little
holes in the plates that turn so that the air is not visible on
disassembly. They say that you would never see air at disassembly unless
you obtained a brand new unit where the air had not yet mixed into the
fluid (like a brand new unit which had not been mixed up after disassembly.
So they are essentially arguing that you have been tricked into thinking
the unit needs no air because you were not privy to the manufacturing
process and subsequent mixing of the air into the fluid. They say that the
likely effect of not having enough air in the unit would be that the unit
would be much too stiff. If their comments are accurate, then it is
speculation, but possible, that a party would move to a thinner fluid
formulation thinking that less viscocity was needed whereas what was
actually needed was a certain percent of air. The percent of air seems to
remain a trade secret with them but 14% is an aggressive percent as regards
these units generally (this number is not specific to Volkswagen; it is a
generic VC number). We do not know of any VC number that has ever been
aired by an authoritative source that applies to the Vanagon coupling.
3. Both GKN and SDP officials did an equivalent of a laugh on the phone
with me when I suggested that there were VC rebuilders were thinking there
was no air in the units and rebuilding them without adding a specified
amount of air.
4. The statement that there are microsocopic burrs that wear off comes from
senior engineers at SDP itself. They do not explain why or how these burrs
wear off, but merely state that they do. It is theoretically possible that
SDP says this simply to sell more VCs but, to add an editorial comment,
somehow I doubt this as I don't think they really care. And anyway, my
contact there is an engineer, not in sales.
5. Noone at any time told me that it was the microsocopic burrs suspended
in the fluid that turns it black. What they did say is that: the burrs
become suspended in the fluid and that this is part of the process by which
the fluid becomes bad. Separately that black fluid indicates a bad VC. It
was an error introduced by me in writing to suggest that these two events
were related--that it is the burrs that turn the fluid black. No person
ever told me that the burrs turn the fluid black.
6. The forgoing information suggests that there is work to do in the area
of % air and attempting to confirm/disconfirm whether the burrs wear off
and are important to proper VC action. The official at GKN and SDP both
mentioned the burrs independent of any prompting by me.
7. If indeed there is a proper percent air, then it is possible that the
original strength Dow Corning fluid that you received was indeed the proper
viscocity and that instead of trying to thin it, the better course would
have been to add some air.
The specifications from an SDP engineering, created at the time the Vanagon
viscous coupling was designed, suggest a possible way to make progress on
addressing the issues raised in this email more fully. One could make VCs
with the original fluid you received from Dow Corning and then put in
various amounts of air in prior to testing. One could then see what percent
yields test results most closely to the SDP spec. The SDP spec is as follows:
Testing is being done by keeping the housing stationery and having the hub
rotate with a constant speed of 75 rpm. Temperatured is being measured by
means of a thermo coupling on the outside of the housing. Torque transfer
in Nm vs. temperature (C) at a speed difference of 75 rpm is the key
measurement. The torque level slowly *drops*! as the temperature goes from
10 degrees C to 120 degrees centegrade (from 250Nm to 140Nm) and then
shoots up to an infinite torque sharply at a point between a termperature
of 120C and 135C.
On a theoretical basis, the GKN engineer told me that the range in which
the correct results are obtained is 84% to 86% fluid with the remainder
air. But he cautioned that this spec is not specific to the Volkswagen VC
and was only applicable to all VCs that he has heard of.
If you were to run some tests and announce that you had found that only 0%
air would produce results to the SDP spec, then this would go a long way
toward resolving this issue in the minds of some percentage of your
customer base which is now aware of this issue through exposure to my report.
Please email me back to say whether I can forward a copy of your comments
to me to the Vanagon List. I think it is important that your customer base
can see your comments.
If you would like to switch to the lighter side, take a look at some of my
URLs in my SIG at the bottom of this email, where you will see my van using
its viscous coupling.
------------------------------------------------------------
Forrest's response to that.
Derek,
Thanks for the further clarification. The info that you have is
interesting, and puzzling, as it is at odds with what we have found with
new VW viscous couplings. We bought about a dozen new complete front
diffs a few years ago, brand-new, never been in vehicles (cool units,
though, with locking diffs; be sure to be pointed in the direction you
really want to go before engaging!). Careful disassembly of the
couplings showed no free air at all. The silicone fluid itself is almost
clear, approximately the viscosity of room-temp STP. Plunging a
screwdriver a couple of times into the fluid introduces lots of little
air bubbles which remain suspended within the fluid, bubbles that take
the better part of a weekend to dissipate due to the viscosity. I've
never tried, of course, but I get the feeling that if one used some sort
of high-powered mixer to introduce lots of air into the fluid you would
end up with a cloudy, virtually opaque fluid, sort of like whipped
butter. And, after sitting for a long enough time, the air would
eventually rise and separate from the fluid. This is *not* what we found
in the new units.
One of the problems of being an aftermarket rebuilder is the lack of
factory support. As you probably know, VW/Audi are loathe to share
technical info; all we can do is get any manuals and tech bulletins that
are available, keep on the best terms possible with the local dealer,
and ask questions whenever appropriate. Then, try to do the best reverse
engineering possible when no answers are forthcoming. That's what we
tried to do with the Syncro vc's, relying to a large extent on
Dow-Corning's experience with the silicone fluid, and it seems to be
working; i.e., our vc's don't lock up in normal driving, but do when the
wheels slip in mud or snow. In a perfect world I'd love to be able to do
the sort of engineering temp-vs-torque test you describe, but we just
don't have the time to expend on such a small market. In the time that
we've sold 80 Syncro front diffs we've sold about 16,000 other VW
transaxles. And, for us at least, the whole viscous coupling question
may soon be moot: most of the cores we get back are unrebuildable due to
wear on the shaft which connects to the front diff, VW won't sell this
part (ah, another chance to welcome owners into the 'factory rebuilt'
fold), and it's economically unfeasible to have this piece
manufactured.
Feel free to post my response on your Vanagon list. I
probably won't have access to a computer till sometime next week, as I'm
off to a tranny seminar (yes, there is such a thing!) in Orlando
tomorrow. Magic Kingdom, here I come...
Forrest Friedrich
German Transaxle of America, Inc.
Bend, OR
----------------------------------------------------------------------------
----
Posting from a couple of months ago that found its way to Wheedle, and then
was faxed to Forrest:
1. If you refill a VC so that it is full it will burst its seals. Never
refill so it is full. (The proper percent air is 84% to 86% according to an
estimate from another expert familiar with Viscous Couplings but this
percent CANNOT BE CONFIRMED YET and the percentage is critical.) 290 Grams
of fluid is the SDP spec for the amount for a refill, but we need to find
out of 290 grams is actually the amount to put into the VC or simply the
order quantity.
1a. Someone should check with German Transaxle/Wheedle to see if they leave
the proper amount of air in the unit or not. There is now so much
corroboration (from both SDP and GKN) that there needs to be air in the VC
that, no matter what they told me on the phone, it seems imposssible that
German Transaxle would be filling the VC so that it does not contain air. I
would imagine therefore that it was an error on my part to think that they
were not carefully assessing the proper percent air. But do not have any
repairs done with one of these outfits without at least confirming this
issue with them.
2. The plates inside the Viscous Couplings have microscopic burrs on them
that are critical to the proper torque transfer of these units. This means
that the plates wear out and should generally not be reused!!!!!!
3. Since replacement plates are not available, this means Viscous Couplings
rebuilds are questionable procedures and should be avoided.
4. Any viscous coupling under pressure is likely a bad viscous coupling! If
you open the unit, and a lot of VC juice comes out, the VC is likely bad or
going bad!!!!!
5. Viscous Couplings may be expected to last at least 60,000 to 90,000
miles. After that, they may become worn out and should then be
replaced!!!!!! This makes it sound like they should be replaced as often as
your clutch!!!!!
6. There is *no difference* between the Viscous Coupling fluid in the 14"
and 16" models.
The problem is that there are microscopic burrs on the metal plates. With
use, these burrs get worn down/off so they are suspended in the fluid but
they are important to imparting the proper torque. So the correct method
would be to order new plates at the same time as ordering new fluid and seals.
We do not have a source of plates and SDP gives us a cold shoulder when we
query after the plates or the fluid. But I would be interested to see if
the following part number yields them: 094 409 448 (or 44B cannot read
correctly).
We do not yet have a good source of the fluid but you can wheedle a can of
the fluid out of Wheedle for about $200. It would appear that you should
leave 14% air in the unit and that the correct fill amount is 290 grams of
the fluid. Tim says the fluid should cost more like $42. I do not really
trust that Wheedle is picking the correct fluid, so I would be cautious
about ordering it from them.
If the fluid in your old VC is of black color, this is said by Wheedle and
Oregon to be an indication that that fluid had gone bad. The fluid appears
to go bad after it absurbs a sufficient quantity of the microsopic burrs I
spoke of, so black fluid indicates the possibility you may wish to change
the plates as well.
There is a possibility that the following part number will yield from VW a
container of the correct fluid:
N 052 130 00
Can't be sure.
Most likely the part numbers I have provided must be ordered from a
European VW dealer because they do not show up in my USA microfiche.
After refilling and reinstalling everything, you will know you have done
the job right if your VC meets the following test:
Testing is being done by keeping the housing stationery and having the hub
rotate with a constant speed of 75 rpm. Temperatured is being measured by
means of a thermo coupling on the outside of the housing. Torque transfer
in Nm vs. temperature (C) at a speed difference of 75 rpm is the key
measurement. The torque level slowly *drops*! as the temperature goes from
10 degrees C to 120 degrees centegrade (from 250Nm to 140Nm) and then
shoots up to an infinite torque sharply at a point between a termperature
of 120C and 135C.
At 11:23 PM 8/4/99 -0400, you wrote:
>From: "karl" <thewestyman@mindspring.com>
>
>Today at my shop, I had my first run-in with the famed and often feared
>'Viscous coupling'. I was working on a Westy Syncro that I am getting ready
>to sell, and decided to test said coupling. Jacked the rear up, started it
>up, put it in gear, and........nothing! Not the slightest movement from the
>front. Well, I happened to be so lucky to have a 'spare' van here, so I
>pulled the front diff and transplanted it to the Westy. Works fine!
>
>Now for the 'fun' part. Having had a father who was a tinkerer (that's how I
>got my start in this business), I of couse had to pull the old diff apart to
>have a look at what troubled it. Wonderfully easy to remove the viscous
>coupling! On the bench, I removed the two small plugs from the side plate.
>Nothing ran out....hmmm.....it did have a somewhat foul odor though. Not
>knowing what to expect, I lightly placed the coupling in a vise, to hold it
>together as I removed the large snapring that holds it all together.
>Carefully loosened the vise, nothing popped apart. A couple taps with the
>plastic mallet, and the thing seperated. I have always been under the
>impression that these viscous coupling were some really complicated device,
>but it is not so. Basically, a cup-shaped housing, a side plate, a hub, and
>a stack of thin plates, alternating ones having holes or slots. Quite a
>simple assembly, IMO! This particular unit had very little fliud in it,
>however all the plates appear in very good condition, free of warpage. There
>was a small amount of fluid (more like sticky 'motor-honey') on the plates.
>The failure that allowed the fluid to escape was one of the seals on the
>hub. That's it--one little seal. This brings me to my next idea: if we can
>somehow get our hands on the fluid, as well as the two seals and one o-ring,
>we (or I) could rebuild these things at a very reasonable cost. Am I missing
>anything in my thoughts? Is there any reason, outside of finding the bits to
>'rebuild' them, that this is not a possibility? Please pardon me if I missed
>any pertinent postings recently, I guess I am guilty of using the delete key
>excessively. So, lets talk about this. Any thoughts?
>
>Karl Mullendore
>'87 West Syncro x 2 (1 FS)
>'87 Syncro 7-pass x 2 (both FS)
>
>
>--------------------------- ONElist Sponsor ----------------------------
>
>ONElist: home to the world's liveliest email communities.
>
>------------------------------------------------------------------------
________________________________________________________
Derek Drew New York, NY & Washington, DC
drew@interport.net
'90 Syncro Westfalia...
...seen off-road at http://www.tiu.net/~des/vw/drew/index.html
Also seen at: http://www.pipeline.com/~tforhan/swrange.htm
http://4wd.sofcom.com/VW/snow.html
http://www.pipeline.com/~tforhan/swmassn.htm
http://4wd.sofcom.com/VW/Campers/Campers.html
Note: most valuable Vanagon sites on the planet (for owners) are:
http://gerry.vanagon.com/cgi-bin/wa.exe?S1=vanagon
ftp://gerry.vanagon.com/pub/
