CIS fuel system flow rates (revisited)

2350 Views 21 Replies 7 Participants Last post by Butcher, Nov 29, 2001

Stephen Webb

Discussion Starter · Nov 26, 2001

I posted on this topic a week or so ago. To summarize, I tested the stock CIS system with full plate lift, for 30 seconds. I compraed the results with the manual, and then computed ballpark HP figures from the fuel flow.
The results were discouraging - I flowed as much fuel as the manual indicated (good), but my calculations put the max HP at about 100.
I have investigated further, and found out what the problem was (at least in part).
1. The values given in the book were not for full plate lift, but for some "full-load" test where the plate height is set by a special tool (ie it isn't necessarily at full lift). So, this value doesn't indicate the capacity of the fuel distributor, but more is an approximation of what the "full load" condition for the 1.8l engine.
2. The measuring device I was using is *WRONG*. (I wish I had a picture of it). It's a lab-looking graduated beaker with a glass handle stuck on it - it was given out as a gift/door prize at some science fair type of thing. I trusted it because it had the words "LAB GRADE" on the side of it.
I will give you my results now, but keep in mind that I had to estimate my fuel flow based on some half-baked "calibration" of the beaker. To calibrate I used a *measuring cup*, so my results could easily be 20% off. I tend to be conservative, so if anything the results should be less than actual.
Full load, 30 seconds, stock CIS fuel distributor:
Avg. of about 75cc per injector, or 150cc/minute per injector.
This is good for about 110 HP, by my calculations (0.5 lb/hr/HP N/A)
Full load, 30 seconds, Volvo 240 fuel distributor:
110cc per injector, or 220cc/minute per injector.
This is good for about 160 or so HP (0.5 lb/hr/HP N/A)
Turbo HP will be lower.
This is significantly below expectations (170 for stock, 220 or 260 for the upgrade).
Unfortunately I did not monitor fuel pressure during this test; I will next time.
I would be interested in hearing any comments on this.
-Steve

See less See more

1 - 20 of 22 Posts

89turbrio

· #2 · Nov 26, 2001

5

Re: CIS fuel system flow rates (Stephen Webb)

Mr Steven Webb. I think you're a genius. And you have way too much time on your hands- I would kill several people to have the time to test my injector flow rates empirically.
I need to know what the results of these tests are. To clarify: your test was at full throtle- so at partial throttle the yield is only lower?
And my next question is to the masses: HOW CAN THIS BEEE!!!
The CIS-e system I had thought was good to 200-210 HP as per hmmmm, some dyno result postd by an 8v turbo car on "show us your numbers" post. Also there's the bunnyboys who run 16PSI and 230HP on their CIS-e injected rabbit. Granted it has water injection.
Somebody comment on this mad scientists' experiment!!!

See less See more

89turbrio

· #3 · Nov 26, 2001

Re: CIS fuel system flow rates (89turbrio)

And what do we do about it!!!!

rocco2nr

· #4 · Nov 26, 2001

Re: CIS fuel system flow rates (89turbrio)

It must also be noted that this was an unmodified CIS system (correct?).
You tested the volvo distributor, but you tested it at the same static control pressure as the standard CIS one, right? I think a higher HP number could have been obtained by using a CPR with a pressurized boost inlet.

Stephen Webb

Discussion Starter · #5 · Nov 26, 2001

Re: CIS fuel system flow rates (89turbrio)

It actually didn't take too much time to do the tests - certainly not worth killing for.
The tests were done as follows:
Injectors pulled from head.
Each injector was put in a separate empty bottle. (Newcastle Brown Ale)
The fuel metering flap was raised as high as it could go and held in place.
The key was switched to the "on" position for 30 seconds. (fuel pump running continuously)
After the test, the fuel contained in the bottles was measured individually in an uncalibrated graduated beaker. Later the beaker was "calibrated" against a kitchen measuring cup by filling the beaker to observed level, and emptying the contents into the measuring cup. This yielded appx 440 cc of liquid in the measuring cup, so appx 110 cc of liquid was in the original sample.
All 4 injectors delivered 110 cc +/- 3cc in 30 seconds.
So, each injector (with this fuel distributor) is appx 220cc/minute injector. (again, keep in mind that this compares with a 260cc EFI injector at 85% duty cycle)
-Steve

See less See more

Stephen Webb

Discussion Starter · #6 · Nov 26, 2001

Re: CIS fuel system flow rates (89turbrio)

RE: How can this be, and What can we do about it?
1. I have been claiming 220 or 260 HP in the past for this fuel distributor. I may have been mistakenly quoting the numbers from the Audi 5 cylinder unit. 220 * 4/5 ~= 175 HP, so it is plausible that with only 4 injectors I am only getting 4/5 of the audi 5 cylinder flow rating. After revisiting the www.turbobricks.org website, it seems that this unit is good for about only 180 HP or so before they start talking about fueling upgrades.
2. What can we do about it?
If you want more than 180 HP, the turbobricks website offers a few suggestions.
Run the 5th injector (or some other small injector) with a Hobbs switch at 10 psi of boost or so. (not favorable in my book)
Add an extra injector controller (and injectors) to get better control over the additional fuel. Better, but money.
Increase the system pressure by shimming the fuel pressure regulator. This is probably what I will do. Shim the fuel distributor so the pressure is at the high end of the factory spec.
They also mention using a V8 fuel distributor from a 928 or a mercedes V8 (someone else was talking about this on Vortex - my apologies to whoever it was, I forget). At any rate, the turbobricks guys talk of running two of the ports into a Y adapter, that feeds one injector. As long as the injector can handle the additional flow, I guess it would work.
Next weekend I will play around with the system pressure and see how much additional fuel I can get out of the system.
-Steve

See less See more

Stephen Webb

Discussion Starter · #7 · Nov 26, 2001

Re: CIS fuel system flow rates (rocco2nr)

quote:[HR][/HR]It must also be noted that this was an unmodified CIS system (correct?).
You tested the volvo distributor, but you tested it at the same static control pressure as the standard CIS one, right? I think a higher HP number could have been obtained by using a CPR with a pressurized boost inlet.[HR][/HR]

Yes, it was an unmodified CIS system, using the stock CPR from the VW. I would argue, however, that this doesn't matter. The reason I say this is because of how the control pressure affects mixture is by making it more ore less easy for the flap to raise (if it raises more easily, then for a given amount of air there will be more fuel). Since I physically lifted the air flap with my finger, and forced it to be at full lift, I effectively eliminated the control pressure regulator. This is my understanding, at least.
There are two things that I was not diligent in monitoring. The first is the system pressure. It is possible (though I think unlikely - I will check and let you guys know) that the fuel pressure dropped off under this full load condition because the engine was not running, and hence the fuel pump voltage was lower than it would normally be.
The second thing that I did not monitor was the frequency valve duty cycle. My understanding is that this controls the differential pressure and affects how much fuel is injected at a given air flap lift. I believe that in the tested configuration (engine cold, key on, engine not running) that the duty cycle would be pretty high (70%?) which would cause more fuel to be injected, not less. I can't be certain about this without testing, though, so I will.
Shimming the system pressure regulator could give a 15% gain with very little effort. I plan to check this as well.
-Steve

See less See more

Peter Tong

· #8 · Nov 27, 2001

Re: CIS fuel system flow rates (Stephen Webb)

Steve is that the 114 FD? That one is rated at 236cc's per minute 50% duty cycle. High system pressure setting for that one (B21FT motor) is 5.9 bar... as soon as I get a chance I'm going to set my 924 turbo one to 6.2 bar... glad to hear you figured out the problem...I know I mentioned this before but did you check that the duty cycle was in the 70-80% range?
Oops I guess you didn't check...if you could ground the O2 sensor wire and retest that would be good...
regards,
Peter Tong

quote:[HR][/HR]It must also be noted that this was an unmodified CIS system (correct?).
You tested the volvo distributor, but you tested it at the same static control pressure as the standard CIS one, right? I think a higher HP number could have been obtained by using a CPR with a pressurized boost inlet.
Yes, it was an unmodified CIS system, using the stock CPR from the VW. I would argue, however, that this doesn't matter. The reason I say this is because of how the control pressure affects mixture is by making it more ore less easy for the flap to raise (if it raises more easily, then for a given amount of air there will be more fuel). Since I physically lifted the air flap with my finger, and forced it to be at full lift, I effectively eliminated the control pressure regulator. This is my understanding, at least.
There are two things that I was not diligent in monitoring. The first is the system pressure. It is possible (though I think unlikely - I will check and let you guys know) that the fuel pressure dropped off under this full load condition because the engine was not running, and hence the fuel pump voltage was lower than it would normally be.
The second thing that I did not monitor was the frequency valve duty cycle. My understanding is that this controls the differential pressure and affects how much fuel is injected at a given air flap lift. I believe that in the tested configuration (engine cold, key on, engine not running) that the duty cycle would be pretty high (70%?) which would cause more fuel to be injected, not less. I can't be certain about this without testing, though, so I will.
Shimming the system pressure regulator could give a 15% gain with very little effort. I plan to check this as well.
-Steve[HR][/HR]

[Modified by Peter Tong, 6:57 PM 11-26-2001]

See less See more

Peter Tong

· #9 · Nov 27, 2001

Re: CIS fuel system flow rates (Stephen Webb)

Even though the fueling is controlled by the fuel distributor I think there is a practical air flow limit for these 80mm flow meters. Sure the profile of the cone could vary between different 80mm plates but its still got to stay within a band of acceptable a/f ratios with regards to the metering plate lift. I'm going to guess that the top end flow limit for these things is something just over 210-220hp... I've corresponded with a few 924 turbo folks running the european injection that have dynoed 220 bhp worth... beyond that and it appears you have to change the metering plate to the V8 one.
Just my $.02 worth. and yes as soon as I can flow the 924 turbo setup I'll do it.
Peter Tong

quote:[HR][/HR]RE: How can this be, and What can we do about it?
1. I have been claiming 220 or 260 HP in the past for this fuel distributor. I may have been mistakenly quoting the numbers from the Audi 5 cylinder unit. 220 * 4/5 ~= 175 HP, so it is plausible that with only 4 injectors I am only getting 4/5 of the audi 5 cylinder flow rating. After revisiting the www.turbobricks.org website, it seems that this unit is good for about only 180 HP or so before they start talking about fueling upgrades.
2. What can we do about it?
If you want more than 180 HP, the turbobricks website offers a few suggestions.
Run the 5th injector (or some other small injector) with a Hobbs switch at 10 psi of boost or so. (not favorable in my book)
Add an extra injector controller (and injectors) to get better control over the additional fuel. Better, but money.
Increase the system pressure by shimming the fuel pressure regulator. This is probably what I will do. Shim the fuel distributor so the pressure is at the high end of the factory spec.
They also mention using a V8 fuel distributor from a 928 or a mercedes V8 (someone else was talking about this on Vortex - my apologies to whoever it was, I forget). At any rate, the turbobricks guys talk of running two of the ports into a Y adapter, that feeds one injector. As long as the injector can handle the additional flow, I guess it would work.
Next weekend I will play around with the system pressure and see how much additional fuel I can get out of the system.
-Steve[HR][/HR]

See less See more

Stephen Webb

Discussion Starter · #10 · Nov 27, 2001

Re: CIS fuel system flow rates (Peter Tong)

quote:[HR][/HR]Steve is that the 114 FD? That one is rated at 236cc's per minute 50% duty cycle. High system pressure setting for that one (B21FT motor) is 5.9 bar... [HR][/HR]

Not sure what you mean by 114. Part number suffix, perhaps? I will have to check.
It is from 84/85 (I forget which year for sure. Sealed beam headlights...boxy

)
Anyway, it's a 240 turbo, and from what I remember it is the B21FT. (Sorry, don't really know much about Volvos)
I will check the duty cycle next time.
Thanks for the info.
-Steve

See less See more

Stephen Webb

Discussion Starter · #11 · Nov 27, 2001

Re: CIS fuel system flow rates (Peter Tong)

So are you saying that the 80mm diameter plate is the restriction for airflow, or just that at higher airflow rates the plate doesn't meter fuel correctly?
If it were a perfect tube 80mm in diameter, I think it could support 1000 CFM without issue. Of course it's not even close to being ideal. It would be interesting to put a fuel distributor on a flow bench..

How big is the 8 cylinder fuel distributor? Does it bolt in to the factory airbox like the Volvo one does, or no such luck?
I'm keeping things mild, so I really don't need more than about 170 HP, which it looks like the Volvo unit will handle.

Thanks again for the info.
-Steve

See less See more

psi

· #12 · Nov 27, 2001

Re: CIS fuel system flow rates (Stephen Webb)

If you take you original estimate of plus or minus 20% then your estimate of 180 hp is in the range of the original 220+ hp.You also say that the flow was equal to some 260cc injectors.The 260cc injectors will make close to 200 wheel hp in a turbo G60.Keep working on this set up,I love this idea

Chuck B

See less See more

Peter Tong

· #13 · Nov 27, 2001

Re: CIS fuel system flow rates (Stephen Webb)

The larger V8 metering plates are 110mm...
What I'm guessing at/proposing is that with the control pressure counterforce being what it is, and the flow cone being what it is that to maintain the proper a/f ratio that there has got to be some kind of upper limit at which the plate hits the limit. Whatever that air flow is I'm guessing its around 210hp or so worth of air or thereabouts. Just my own theory however...
The highest flowing CIS-Lambda fuel distributor I know of on a stock car is the one on the 911 turbo 3.6 CIS.
Peter Tong

quote:[HR][/HR]So are you saying that the 80mm diameter plate is the restriction for airflow, or just that at higher airflow rates the plate doesn't meter fuel correctly?
If it were a perfect tube 80mm in diameter, I think it could support 1000 CFM without issue. Of course it's not even close to being ideal. It would be interesting to put a fuel distributor on a flow bench..

How big is the 8 cylinder fuel distributor? Does it bolt in to the factory airbox like the Volvo one does, or no such luck?
I'm keeping things mild, so I really don't need more than about 170 HP, which it looks like the Volvo unit will handle.

Thanks again for the info.
-Steve[HR][/HR]

See less See more

Peter Tong

· #14 · Nov 27, 2001

Re: CIS fuel system flow rates (psi)

Hey chuck,
If you can locate a V8 FD that is in good working condition I'd be willing to give it a go on my Lysholm Cabby...this would be called taking old school stuff to the limit... I believe some of the V8 flow meters would bolt right up to the bottom section of my 924 turbo downflow flow meter section
regards,
Peter T.

Stephen Webb

Discussion Starter · #15 · Nov 27, 2001

2

Re: CIS fuel system flow rates (psi)

quote:[HR][/HR]If you take you original estimate of plus or minus 20% then your estimate of 180 hp is in the range of the original 220+ hp.You also say that the flow was equal to some 260cc injectors.The 260cc injectors will make close to 200 wheel hp in a turbo G60.Keep working on this set up,I love this idea

Chuck B[HR][/HR]

If I could pin down fuel density at about 65 deg F, and fuel consumption numbers, I could get a bit better estimate of power, perhaps. I have been assuming 6 lb/gallon for gasoline - I have seen as low as 5.85 and as high as 6.2. I have also been using .55 lb/bhp/hr fuel consumption (I think). I have seen as low as .45 (NA, I think) and as high as .7
If I use 6.2 lb/gallon, 236 CC/minute, and .45 lb/bhp/hr, I get as high as 206 HP.
If I am more conservative and use 5.85 lb/gallon, and 220 cc/minute and .6 lb/bhp/hr I get only 136 or so!
Looking at SDSEFI website, 231 CC inectors (at 100% duty cycle) assuming .55 lb/bhb/hr is good fo only 160 HP on a 4 cylinder.
Sort of depressing, but when I red that the STOCK volvo engine (intercooled, 10.5 psi) pushed into the 160 HP range, I have to believe that the fuel distributor can handle more than that.
As for the 260cc injectors makign 200 WHP, that is encouraging. My comparison was for 85% duty cycle, though, and maybe the guys getting 200 WHP are running near 100% duty cycle?
As far as this being an interesting idea, it's more than an idea - I'm doing it

I've got pictures to prove it.
-Steve

See less See more

rocco2nr

· #16 · Nov 27, 2001

Re: CIS fuel system flow rates (Stephen Webb)

quote:[HR][/HR] My comparison was for 85% duty cycle, though, and maybe the guys getting 200 WHP are running near 100% duty cycle?
[HR][/HR]

ROFL they made EFI back to CIS

this is all very interesting and I hope we can gain something from it.

See less See more

89turbrio

· #17 · Nov 27, 2001

Re: CIS fuel system flow rates (Stephen Webb)

So really the quesstion is one of injector size. We could just call up RC engineering and get some large injectors.
However- is this necessary? I have a 5th injector that is triggered to boost- BUT I also have a large exhaust leak at the manifold- unfortunately causeing my mixture to run rich all the time. The FD says bzzz bzzz bzzz because the oxs reads the sucked-in oxygen as lean mixture. The FPR richens the mix leans it out richens leans it out Constantly!
The a/f meter reads so rich, it's way rich. If these injectors don't flow enough fuel, how is it that my car runs so rich? I'm up to about 9 lbs. boost and no leaning This is in part due to my 5th injector.
After I refresh the headgasket with a copper one, drill out the broken exh studs -fixing the exh leak, I plan to go for a 9-11 lb. spring and probably to a 10-14afterwards. I'll let you know if it leans out at all, but I doubt I'll have a problem, b/c the thing is dumping fuel by the bucketfull with no signs of lettng up.
Tong- what size injectors are you running in the 928? Some had made a comment weeks ago that the injectors were the same size used in the turbo posches.
Steve- Mr. Freetime

- why not just go with a 5th injector and injector controller to trigger to coil pulses at like 4 lbs boost? The alternatove is to get some turned at RC engineering- any flow rate you desire. Bout 300 for a set.

See less See more

Stephen Webb

Discussion Starter · #18 · Nov 27, 2001

Re: CIS fuel system flow rates (89turbrio)

The reason I'm not into the 5th injector setup is because I don't know enough about it. Is the fuel pattern good enough for using as a fuel injector, or just good enough to get the damn car to start...? Is the actuator intended to be pulsed 6000 times/ minute? How long of a pulse does it get per revolution, anyway? The pulse should really increase with boost and RPM, I woudl think (ie when your stock system can't keep up).
I could do it, I guess, but I'd prefer to do it with a "real" injector, and I'd probably want it to be a variable amount of fuel being injected, depending on conditions.
-Steve

Butcher

· #19 · Nov 28, 2001

Re: CIS fuel system flow rates (Stephen Webb)

The best way to go is to have the fuel metered to each port. Yeah, you can do all kind of cheaper routes but the best way is not an extra injector. I really get mad at all these comments since I have been hording a bunch of parts to complete my turbo project then someone talks about V8 fuel dist and that wants me to get rid off all the stuff I bought and start over with the V8 stuff. As a Mercedes tech for 14+ years I am sure I can find all kind of stuff to butch together a perfect system. I hate this new forum.

sirhc

· #20 · Nov 28, 2001

Re: CIS fuel system flow rates (Butcher)

quote:[HR][/HR] I hate this new forum. [HR][/HR]

See less See more

1 - 20 of 22 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.