Pressure Drop through Metering Plate?

[911st]

Is anyone in a position to measure and document the pressure drop through the 930 CIS metering assembly, intercooler, air filter, and or throttle body?

This could be a very important contribution and data point.

Any restrictions in these areas, just like on the exhaust side, hurts our performance potential.

For example if we find the throttle body creates a pressure drop at some point up the HP ladder, we might be best served to enlarge it. If not we should not wast or efforts on it.

However, my biggest interest is in restriction created by the metering assembly.

It might be interesting to see if the boost triggered lowering of Control Pressure is measurable or significant.

On my car I found that if I triggered an immediate drop in Control Pressure with acceleration off idle or from cruse (using my Programmable Fueler), I found this gave me better throttle response and preboost TQ.

I should note that the 3.0T lowers control pressure with the vacuum loss that comes with acceleration and the 3.3 dose so with boost.

I experimented with immediate lowering of control pressure with acceleration on my 91T using a custom programmable fueler and did find that a noticeable improvement in throttle response.

Just a thought.

[David]

I wouldn't worry about the pressure drop on a turbo car since you can just add more boost.

When I built my engine I was more worried about throttle control so I choose the smallest throttle body I could and still stay below (or close to) 300 ft/s air velocity.

As long the air velocity is below sonic it can still flow more.

[911st]

Thank you Daivd,

I am not sure I understand the sonic thing.

Maybe that is why improving differential pressures about the hot side turbine, like with headers or better muffler, improves response and power pretty well where you are thinking taking the same approach to the cold side compressor might not as we can just turn up the boost.

Here is my thinking.

If we lose say 2.5 psi between the CIS metering plate, air filter, plumbing, TB, and intercooler -- I suspect that means the turbo has to work at about 17.2 psi boost to achieve 14.7 psi of boost.

Anything that makes the turbo work harder than it needs to or that requires a larger turbo with more mass to hit ones target boost would seem to work against spool time, efficiency, and add unnecessary heat to the charge air.

I suspect this is part of the reason EFI works better than CIS. Yes EFI is more accurate and adaptive but it is usually less restrictive. Especially if it is a MAP system with no metering to cause restriction.

I do not think we are going to find a lot of opportunity but I believe we can help the boost response. However, if the metering assembly is making for a significant pressure drop then that supports what many believe that CIS is not fuel limited but air flow limited.

Seems getting accurate info on what is actually happening is better than shooting in the dark or taking someone's opinion or sales pitch that a different mod will make a difference.

Just a thought.

Any help is appreciated.

[David]

I seriously doubt CIS is air flow limited. Maximum air flow is reached at sonic flow and we shouldn't be any where near that on a car engine.

I'm no expert on this, but the biggest issues I see are too low air velocity where the cylinders aren't filling and throttle body can't control the air flow and too high an air velocity where the air is so turbulant it also can't fill the cylinders or be metered properly. I forget what the proper velocity is but I think it's between 200 and 300 ft/s.

Roughness, changes in area, throttle plates, intake valves, etc don't necesarily limit air flow but rather reduce air pressure or head as the air passes the obstruction. In a NA engine this affects performance because you only have the vacuum of the engine drawing in atmospheric air so every little bit helps as long as you don't go so big that velocity slows to the point of not filling the cylinders. In a turbocharged engine you're setting the boost based on manifold pressure after most of these obstructions so the turbo will make up for the lost pressure which is probably much less than 1 psi anyway.

Once again, I'm no expert but I felt like typing . Maybe someone else can chime in with more real world experience.

I recommend all the A. Graham Bell books for engine performance. They're well written and easy to understand for anyone from beginner to engineer.

[patkeefe]

The throttle plate is a means to dropping pressure, which is what a throttle is. My guess is that the TB was engineered to perform in this flow range and flow regime by Porsche/Bosch. I kind if know where you are going with this...you need throttle at off boost. If one could devise a system which used throttle at off boost, and some sort of inlet vane restriction on boost to control flow, then you could ditch the throttle at boost conditions.

I concur with David's assessment of the situation, which was pretty good. Fluid mechanics is a difficult topic. Maybe when I get my car running again, I can measue the differential pressure.
Pat

[aerodyn930]

The bottleneck gets moved around, But it can never be removed from the equation,

Achieving the necessary air flow capacity upstream of the turbo ( on modified engines) can be addresed by :

1- Re designing the ducting between the A/F meter and the turbo
2- Modifying the A/F meter (cone and Plate)

When these components where designed the target output in conjunction with other relevant factors gave us little room for increased capacity.

When Porsche began testing the 930/71 (934 motor) bosch engineers approached the restriction of the A/F meter by enlarging the diameter and changing from a metering disc to a cone, This BTW had the side effect of poor fuel mixture progression, The location of the unit was selected so that a short duct could be used into the turbo inlet.
These mods where deemed necessary for a 485bhp output

lowering control pressure should be used to allow the engine to accelerate not as a way to reduce the restriction of the A/F meter, 86-89 engines used this strategy by changing duty cycle of the fequency valve during acceleration

As we look at flow capacity in these components we also stand to gain compressor efficiency while reducing lag,

I personally would first focus on the ducting, soon hope to have a solution, then slowly realize that it is very easy to reach the point where EFI becomes the solution to most of these gremlins,

Lou/ AERO DYNAMICS

BTW, thanks , as I see this becoming a very interesting thread.

[911st]

Lou,

Thanks for the informative thread. We might be thinking some of the same things.

Yes, EFI is better but I love CIS.

I suspect you are on the right tract about the intake turbo plumbing and metering assembly.

I know of a really good tuner in Florida that used to reroute there turbo pipe thru the wheel well on there high HP CIS motors. Also, Porsche made some changes with the C2 turbos in this area for some reasion.

Also note that the exit ID on the metering assembly might not be as big as many think. Maybe if this could be enlarged as it might help.

However, it would be really cool to document the potential of each with proper instrumentation / data logging so it is not so much of a guess and mod approach.

For example, I just wonder how significant the restriction of the metering system might be? Say compared to something like .5 to 1 psi drop for a very good intercooler.

Could it be in the 1-2psi range pre-boost when control pressure is higher? Could it be 1 psi at 350chp, 2psi at 400chp and say start increasing dramatically to somthing like 4psi at 450chp? If so the metering section would warrant a lot of attention as it could yield a good HP return.


I guess I have two agenda's with this thread.

First is to look for enlightenment and ask for help using the wonderful data logging that many now have with the newer wide band systems.

The Second is to verify if converting a 930 WUR so it lower's control pressure with acceleration might help reduce lag as I found it did with my car.



ps: Pat, good to hear from you. I love your SC conversion.

[mark houghton]

Get rid of the airflow metering plate and arm, and use a quick acting stepper motor tied to a MAP to actuate the fuel plungers in the fuel head....or some such magic that is way beyond me. That would get rid of the air flow restriction in the metering assembly anyway. In affect, turng the system into a quasi-electronically operated mechanical fuel injection.

OK, I'm way out there today, in over my head...

[911st]

Mark,

Always a pleasure.

That is to easy.

Just take the metering arm off, put a spring under the metering pin pushing up. Use the existing fuel-distributor's control pressure circuit to push down on the metering pin. Replace the WUR totally with an Frequency Valve. Then the metering pin's position will be 100% controlled by the Frequency Valve duty cycle. Then drive it with the simplest one cylinder programmable Fuel Injection ECU you can find. If you wanted you could even use the later CIS lambda system's existing capability to fine tune the AFR's.


Thought about it (and I think we talked about it) but at that point EFI is a better solution.

I think it is more fun and challenging to get our vintage fuel delivery system to work better without added electronics.

[mark houghton]

Originally Posted by 911st

Mark,

Always a pleasure.

That is to easy.

Just take the metering arm off, put a spring under the metering pin pushing up. Use the existing fuel-distributor's control pressure circuit to push down on the metering pin. Replace the WUR totally with an Frequency Valve. Then the metering pin's position will be 100% controlled by the Frequency Valve duty cycle. Then drive it with the simplest one cylinder programmable Fuel Injection ECU you can find. If you wanted you could even use the later CIS lambda system's existing capability to fine tune the AFR's.


Thought about it (and I think we talked about it) but at that point EFI is a better solution.

I think it is more fun and challenging to get our vintage fuel delivery system to work better without added electronics.

Yeah, I think we did talk about this awhile back, but it's fun to be creative. You're out there a lot farther than I! Though I tend to agree, CIS is not at all a bad fueling system...just needs a touch of refinement.