Re: Improved S-Knob

[ptimmerm@mashtun.JPL.NASA.GOV]

the industrious Russ Wrote

>> When we spoke on the phone a few months ago I mentioned that I was going
>> to try putting a check valve inline with the W/G vacuum line to prevent
>> engine vacuum from being applied to the W/G. 
>> 
>> At any rate with the check valve installed I think there is a noticable
>> difference in how fast the boost pressure comes up during gear shifts. I
>> have a mechanical boost gauge and can see that the pressure comes right
>> up to about 6 psi when shifting before it lags for a second or two and
>> then takes off for 15 psi. Without the check valve the boost only comes

Graydon Cherfully pitched in with:

>Hmm, I'm not sure a check valve is the right thing to do here.   I used a 
>pressure regulator in the lline to the top of the wastegate, and that 
>worked well.   The thing that makees me wonder though, is why does your 
>boost sit at 6psi and then rise to 25?   My car has alaways gone directly 
>to the max pressure set point with out any hesitation at all.   
>Occasionally it will rise one psi or so higher than the setpoint, but it 
>generally gets to the setpoint very smoothly.


A few comments from Dr. Knob ( does that have a good ring? )

I have to admit that the leak was my idea.  Heres my reasoning.  

If you put a check valve in the circuit to the cap you will build up 
pressure and hold it, causing strain to the wastegate diaphram. Well, 
likely some leak, but you really don't want to break that hard to get 
bit of plastic (rubber?).

I have seen that one unnamed lister has run a knob wastegate controller 
turned up to 22 psi output, to compensate for a leakey WG cap.  
Normally they are adjusted to less than half that value. I did not 
recommend this, but he was very happy with the results, i.e. 
didn't have to fix his car :-) The car ran fine providing good boost and 
normal idle too.  Thus I knew a little leak would not hurt.  

Thus it was my suggestion that if Ross wanted to play at this, he add
the some leak ON PURPOSE.  I hope it sounds a little more logical now.
The only time the thing will leak is when you are running boost, and
the total amount of leakage in the system will be MUCH less than the
accuracey of the air measuring ability of the system.  Say the turbo is
pumping 100 CFM at some level of boost, what fraction of that would be
being lost through a pin hole 0.1% ?  Not much.

A more elegant solution could be rigged up using multiple check vavles
and a precision orifice.  Something like this

	   +--0---|<----+
	---|		|-----
	   +------>|----+

where + is a connection  |< is a check vavle and 0 is an orifice.

The advantage would be it would still be closed loop, and the 
orifice could be matched with some presision to the job.  In
terms of air balance in the closed system, you still would have
some small amount rushing in and out of the cap creating tiny disturbances 
to air measurement.  That happens already with this system, no problem.

I hope I didn't beat this to death :-) 

paul timmerman