Re: Bypass valves and turbos-long

[Scott Mockry <scottmo@teleport.com>]

Jeff and Scott,

The Dodge Turbo site mentions the use of a one way check valve which allows
vacuum to go to the bypass valve small fitting, but bleeds off boost to the
bypass valve small fitting. The reasoning is that the boost pressure acting
on the diaphragm will somehow "damage" the valve. 

This " questionable" approach IMHO, will certainly cause the Bosch Bypass
valve to leak prematurely under boost, and is NOT how the factory bypass
valve is installed in the 1991 200TQ 20V and the 1992> S4/S6. Having boost
pressure acting on both sides of the bypass valve diaphragm as was done by
the Audi factory, one would think, would be less stressful on the
diaphragm, than having atmospheric pressure on one side, and boost pressure
on the other.

The OEM Audi Bosch turbo bypass valves REQUIRE boost pressure from the
intake manifold acting on the diaphragm though the small vacuum/boost
fitting in order to remain closed under boost. 

As Scott J. mentioned, the stock 20V Bypass valve as used on the 1991 200TQ
20V and on the S4/S6 use a spring that will crack open at about 6 psi IF no
pressure is applied to the diaphragm via the small fitting. But of course,
the valve has boost pressure acting on the diaphragm to assist the spring
pressure.

The other overlooked fact  is the Bosch bypass valve diaphragm AREA which
receives the intake manifold vacuum/boost is larger and provides more force
to hold the diaphragm closed than the turbo exit boost pressure acting on
the other side of the diaphragm to try and push it open. You also have a
slight vacuum acting from the turbo inlet side which further helps to keep
the bypass valve diaphragm closed under boost.

Here is a post to the main qlist I did back on 3/27/99 which may help to
fully understand how the Bosch Bypass Valves really operate. Please let me
know if I am missing something here.

From: "Scott Mo." <scottmo@teleport.com>
Subject: More Bypass Valve stuff
Date: Sat, 27 Mar 1999 11:50:03 -0800

>Scott Justusson said:
>I measured several valves at my shop a while ago.  The valve with upper
>connection open to atmosphere (sea level) it will crack open at 6psi.
>The sport quattro valve will crack at ~7-8psi.  The sport quattro valve
>uses a slightly stiffer spring and has a thicker diaphragm.  It also
>costs almost twice as much as the stock 20vt. 

Scott Mockry Said:
That jives with the two bypass valves I took apart, one from the Porsche
944T and one from a 1991 200TQ 20V. Both valves are identical except for
the diaphragms, I was wrong about my previous comment that the 944T BP
valve spring was different. Both springs are identical and provide about
6-7 lbs of force against the diaphragm. The Audi diaphragm is the same
size, but has the round metal plate riveted on both sides in the center
of the diaphragm as Scott J. has commented about before. 

The 944T diaphragm has no metal in the center of the diaphragm, it just
has the rubber, but one thing I did notice, is there is a hole through
the back plastic support which apparently adds some pressure between the
plastic support and rubber to help seat the diaphragm.

Bench tests I did on a new replacement bypass valve for the Audi S4/S6
(034 145 710C) where I applied boost pressure to the small vacuum/boost
fitting on the bypass valve, and applied the same pressure to the turbo
exit side, indicated that the valve would not open even with 2.7 bar
absolute (~24.5psi gauge) applied.

Because there is always a slightly lower pressure in the intake
manifold, compared to the pressure exiting the turbo, I always assumed
the spring was in the bypass valve to return the valve to the closed
position, and to help keep the diaphragm closed even when there is the
pressure differential. Pressure losses through the dual pass 5kTQ
intercooler and throttle valve drop at least a couple of psi at 14psi if
I remember correctly.

If you take the bypass valve apart, you will find that the area of the
diaphragm which connects to the intake manifold (vacuum/boost) is about
2.3 square inches, the area of the diaphragm which connects to the turbo
exit (pressure) side is only 1.2 square inches once you subtract the
center area of the opening. Pressure from the intake manifold that is
applied to the diaphragm "should" always win and keep the diaphragm
closed:

For example, 14psi intake manifold pressure times 2.32 square inches
equals 32.5 lbs force, plus the spring pressure of 6lbs equals a total
38.5 lbs holding the bypass valve diaphragm closed, versus the turbo
exit side with lets say, 16 psi times 1.2 square inches equals 19.2 lbs. 

There is also some additional help to keep the diaphragm closed from the
side of the valve opening which connects to the turbo inlet (slight
vacuum) which is 0.57 square inches. It appears from the above numbers
that the size of the diaphragm that receives intake manifold pressure,
is more important than the spring tension to keep the valve closed under
boost.

The dynamic behavior of this valve is likely more complicated because of
the difference or time delay in pressure rise between the turbo exit and
the intake manifold, but it does give some insight to how this valve
"should" work. 

I have a diagram on my web site which illustrates the Bypass valve
installation from the 1991 200TQ 20V and shows the pressure applied to
the relative areas described above.

http://www.sjmautotechnik.com/turbo.html#bypass

HTH
Scott Mockry
scottmo@teleport.com