16v Cars

[Eric Parham]

Thanks Jim. I went back and looked for their names in the results for the July 23 race at the Glen, and it looks like John turned a 2:23.9 on the long course (not too shabby ), but was down 2 laps for some reason. Do you know if his is a 2.0 or a 1.8 car? George DNS. Please PM me if you have contact info handy for either.

[Eric Parham]

Well, DUH!!!!

No offense, Eric, but am I in some alternative-VW world this week? You're telling me that you believe the Volkswagen 2.0L DOHC 16V FWD strut-equipped Golf should weight **300 pounds less** than my Nissan 2.0L DOHC 16V FWD strut-equipped NX2000 (and 150+ pounds less than a 1.8L Miata)? Why in the world would you think that? Do VWs really suck that bad...? I doubt it.

And this is coming from a guy that's owned, enjoyed, driven, and raced some form of VWoA product since I was allowed to! [/b]

Greg, I wouldn't say that VWs suck, but I would say that *ALL* of the current VW ITA options have signifcant drawbacks under the current IT rules when compared with other currently classed makes. Two very significant VW 16V differences come to mind: 1) a really terrible cylinder head design (exhaust has to turn more than 90 deg after getting past valve, thus eliminating most of the usual cross-flow advantages; and 2) CIS injection (basically a glorified carburetor), which is not only continuous, non-cylinder-specific and non-sequential, but provides significant intake restriction and cannot even be tuned very well under the current intake and computer/sensor/injector rules (that is, any computer is pretty much useless since there's so little that's controllable -- it's basically just a fuel pressure regulator for cold-start and emissions).

[shwah]

Greg, I wouldn't say that VWs suck, but I would say that *ALL* of the current VW ITA options have signifcant drawbacks under the current IT rules when compared with other currently classed makes. Two very significant VW 16V differences come to mind: 1) a really terrible cylinder head design (exhaust has to turn more than 90 deg after getting past valve, thus eliminating most of the usual cross-flow advantages; and 2) CIS injection (basically a glorified carburetor), which is not only continuous, non-cylinder-specific and non-sequential, but provides significant intake restriction and cannot even be tuned very well under the current intake and computer/sensor/injector rules (that is, any computer is pretty much useless since there's so little that's controllable -- it's basically just a fuel pressure regulator for cold-start and emissions).
[/b]

The 1.8 16V is CIS-E, which is not very hard to control fuel wise. No chipping required.

The 2.0 16V is CIS-E Motronic which can do more, but mostly via chip tuning, and there are few that can do this in a customized format as an IT racer would need.

[Eric Parham]

The 1.8 16V is CIS-E, which is not very hard to control fuel wise. No chipping required.

The 2.0 16V is CIS-E Motronic which can do more, but mostly via chip tuning, and there are few that can do this in a customized format as an IT racer would need.[/b]

CIS is great. It's virtually fool-proof, but it has no potential in IT when compared with the real EFI on most other modern cars. CIS-E and CIS-Motronic are really the same system for IT purposes (and haven't changed much since the mid '70s). Same inputs and outputs, just a fancier computer to control the fuel pressure on the Motronic (but same sized box). In fact, I've interchanged between CIS-E and CIS-Motronic fuel distributors (inputs and outputs are self-contained) on street cars without any problems at all. The tolerances may be slightly different, and occasionally they need readjusting, but it's the same system controlled by a slightly newer design "computer" (glorified fuel pressure controller). The one difference is that CIS-E has a separate ignition system box, while CIS-Motronic pulls it into the same box as the fuel pressure controller. The fact is that changing the fuel pressure for the entire system simply can't be done fast enough to even compare with real electronic fuel injection systems. The various chips simply allow it to run richer than the original lambda sensor output would allow with the stock computer. Most just yank the Lambda on these things and set it open-loop. The chips can't help it there (other than recurving the ignition on the CIS-Motronic). Some even disconnect the "computer", run a separate ignition, and adjust the mixture manually with fuel pressure or metering pin height.

Edit: CIS really is just a glorified (and spread out) carburetor, although it's more like a Zenith-Stromberg than a Weber. At least it's more reliable, though.

[shwah]

CIS systems also do a much better job of atomizing fuel than 'electronic' fuel injection systems.

CIS-E Motronic has a larger fuel plunger and can support a higher max hp than CIS-E; neither of which really matters for IT spec engines.

The mass air flow sensor in CIS type systems sure may seem like a big restriction when you look at it, but if you consider the size of the opening, and the cross sectional area exposed when you raise the flow plate, I don't think it is as restrictive as people expect.

[Eric Parham]

CIS systems also do a much better job of atomizing fuel than 'electronic' fuel injection systems.[/b]

That used to be very true, and may still be true to some extent, but many modern EFI injectors now have excellent atomization.

CIS-E Motronic has a larger fuel plunger and can support a higher max hp than CIS-E; neither of which really matters for IT spec engines.[/b]

Yes, a slightly longer rod with slightly longer fuel metering slits, as I recall, but no additional functions. They're the same animal for our purposes.

The mass air flow sensor in CIS type systems sure may seem like a big restriction when you look at it, but if you consider the size of the opening, and the cross sectional area exposed when you raise the flow plate, I don't think it is as restrictive as people expect.[/b]

Depends a lot on the control pressure for CIS-basic and CIS-lambda (that's another reason why we try to reduce it). I believe that CIS-E and CIS-Motronic may not be tunable in the same way, but I haven't reviewed it lately. If I'm correct, they may be even worse for our purposes than the older stuff.

[shwah]

It is painfully easy to tune CIS-E.

Get yourself a 0-10k linear potentiometer. Install it in line with the coolant temperature sensor. Go to a dyno and make some runs while measuring air/fuel ratio with a wideband O2 sensor. Turn the potentiometer until you find the optimum setting for power (typically 12.7:1 on these motors). Measure the resulting resistance, and replace the potentiometer with a fixed resistor of that value. People use different methods of switching this enrichment device into and out of the circuit manually or automatically based on WOT switch.

It is consistent, reliable and functional. One thing you are giving up to a programmable system is ignition mapping ability, but in the racing operation range you are typically at max timing regardless of the lower load/speed maps, making this a moot point.

[Eric Parham]

It is painfully easy to tune CIS-E.

Get yourself a 0-10k linear potentiometer. Install it in line with the coolant temperature sensor. Go to a dyno and make some runs while measuring air/fuel ratio with a wideband O2 sensor. Turn the potentiometer until you find the optimum setting for power (typically 12.7:1 on these motors). Measure the resulting resistance, and replace the potentiometer with a fixed resistor of that value. People use different methods of switching this enrichment device into and out of the circuit manually or automatically based on WOT switch.

It is consistent, reliable and functional. One thing you are giving up to a programmable system is ignition mapping ability, but in the racing operation range you are typically at max timing regardless of the lower load/speed maps, making this a moot point.
[/b]

I do agree that the limited adjustments that are possible with CIS-E are easy, but as with most things in life, I'm afraid that the gains are proportional to the difficulty (i.e., not much gain for the easy work). I actually had to do as you suggest with an added potentiometer on a street car once. I put a 16V engine into an 8V GTI and had to make do with the 8V fuel computer. Fortunately, I was able to use a digital multimeter to measure the milliamp current at the fuel pressure actuator and did not have to pay for any dyno time to adjust the pot.

The point I was trying to make in my last post is: 1) the control pressure can be reduced for CIS-basic and CIS-Lambda, and the reduction in control pressure reduces the downward force on the fuel-metering plate thereby reducing intake restriction, in addition to increasing injector pressure; but 2) the force on the fuel metering plate cannot be reduced in this manner for CIS-E and CIS-E-Motronic (collectively "CIS-E"). Also, any reduction to the force on these CIS-E systems would cause a reduction in injector pressure (the opposite of the older systems). Thus, while tuning the basic offset for the fuel mixture may be easy, these CIS-E systems simply cannot be tuned to reduce intake restriction like the older ones could.

[shwah]

OK. I see your point. I honestly just don't have enough CIS 'basic' experience to know that you tune the control pressure for fueling, which in turn reduces the resistance provided by the airflow plate.

FWIW that simple CIS-E tuning method can offer 20+ hp in situations. The factory system runs lean, especially at higher rpm.

[shwah]

I must have been asleep at the keys last night.

I should have mentioned that the CIS-E system also reduces the control pressure to affect mixture, which reduces the tension on the airflow plate. It just uses a different mechanism to do so - this is the DPR that you measured the miliamps at when tuning your street car.

[Eric Parham]

I must have been asleep at the keys last night.

I should have mentioned that the CIS-E system also reduces the control pressure to affect mixture, which reduces the tension on the airflow plate. It just uses a different mechanism to do so - this is the DPR that you measured the miliamps at when tuning your street car.
[/b]

I believe that is incorrect. I seem to remember that the plunger and plate are controlled by SYSTEM pressure in CIS-E, and not by control pressure as in the slightly older CIS. Yup, verified. See ch. 5, p. 24, Fig. 5-10 of Bosch Fuel Injection and Engine Management, Probst 1989 (Robert Bentley Pub). The so-called "pressure actuator" in CIS-E only controls the fuel flow to the injectors, and this controlled flow cannot affect the force on the plunger or plate. It might be a slightly more accurate way of regulating injector pressure, but it sure takes away one of the greatest tuning features of the older CIS.

[shwah]

The differential pressure regulator is what varies the fueling in CIS-E. As the name implies it varies the differential pressure across the system, it does so by changing the control pressure.

[pfcs49]

Sorry Swah, but he's right. I even consulted my Probst to be certain. phil

[shwah]

OK. Chalk up one more in the wrong column for me.

[Eric Parham]

FWIW, I wish you'd been right.

[shwah]

I think I am going to rig up a flow bench type test with my shop-vac and compare the Digi and CIS-E air metering systems at different positions of the metering 'doors'. Maybe later in the winter I will get around to this.

[Bildon]

>> I think I am going to rig up a flow bench type test with my shop-vac

As accurate as any Superflow. And that's not my opinion. It was proven at a Ford engineering facility.

http://www.flowperformance.com

[shwah]

Thanks for the link Bill.

[16v]

You boys just gotta get outta your pouty "going to the garden and eat worms" attitudes and MAKE IT HAPPEN. Show me a max-out build on a Golf (and that means probably spending $25k minimum and running serious shocks instead of off-the-shelf Bilsteins, 9# wheels instead of stock VWs, new Hoosiers, etc) driven by an experienced, skilled driver that's failed to consistenly run at the front, and we can chat. Otherwise...
[/b]

hey hey.. didn't Brian do that in the A car? It was no match for the field of Acuras, Nissans

[shwah]

Didn't catch that until it was quoted above. $25k?!

Some of these $$ numbers people assume are required to build a car still floor me. Sure if you pay for all the work, it takes a big checkbook, and for many that is the only way they have time to race. I totally respect that. However, what I see at the race track is that most IT racers are of the more budget minded type, who are more than willing to figure out how to do much of this stuff themeselves, or do things a different way - think for themselves and try the 'wrong' way that does not invovle buying the 'trick' parts offered by race shops - and succeed.

I don't believe for a second that $25k is any kind of threshold or standard to define a 'fully built' ITA car. Especially with the A2 Golf, the existing chassis development is extensive, the parts are available, they are not outrageously expensive, and they are tested in almost every ITB race across the country every year. The only missing peice is the motor for the 16v, and if you are putting $10k into that motor, you are spending too much IMO.