Finally driving again!, Type IV CIS conversion Options

[dflesburg]

whats with the long pipe on the air box?

Why didnt you just put a k&n cone on there instead of the pipe?

[dlee6204]

I was drooling over Dan & Jeff Dixon's CIS setup on Dan's Type IV at WCR and asked for detailed info. Jeff sent me this great site Ferrari400parts.com. The is a lot of documentation there.

Volvo 1800s also used this system and I've found some info by searching for it.

This is definitely something that is in my future since I've got my own 914, my son's 914 and my other son's VW Bus to think about. It certainly looks superior to anything VW/Porsche ever put on a Type IV!

Are there particular years of Rabbits that you want to find for this?

Link saved! Thanks! The Rabbit I got my stuff off of was a '77. I chose to go with a K-Jet Basic setup (no electronics/ O2 sensor) but I don't know all the specific years or models it came on.

[dlee6204]

whats with the long pipe on the air box?

Why didnt you just put a k&n cone on there instead of the pipe?

The pipe goes from the Fuel Distributor/Airbox to the stock 1.8L throttle body. Fresh air gets sucked up from under the airbox.

[scotty b]

I would think CIS would also be great as the basic system should do well with minor engine mods ( cam ) and even small displacement changes ( 2056 ) (IMG:style_emoticons/default/confused24.gif) I too am considering this setup. I've had a couple 924's and my dad had several rabbits when I was young. CIS is stupid simple and brilliantly effective (IMG:style_emoticons/default/first.gif)

[ArtechnikA]

It's "basically" a mass-flow system so displacement and altitude correction are built in, up to the excursion limits on the flapper.

However, like D-jet, it hates intake reversion and needs a mild cam with not a lot of overlap. Pulses in the intake drive the flapper batty...

[Dr Evil]

There has been proof that CIS actually does fine with radical cams....I need to find where that was. I am putting CIS on my tweaked out corvair engine in the bus (IMG:style_emoticons/default/smile.gif) Already have it in the 914 on the 2.7 (stock).

[ArtechnikA]

There has been proof that CIS actually does fine with radical cams...

There's a whole spectrum of meanings to 'radical'... And 'fine' for that matter ;-) Lift is OK, but when you get into lots of overlap there are going to be issues. Are they issues that have workarounds? probably...

Porsche ran CIS on the 934's because the FIA made 'em do it, and they ran just fine, but not with as much power as the Kugelfisher injection they could run in Group 5. They certainly didn't have 'street-car idle' either. I suspect they gave up a lot in fuel economy at idle and part-throttle - but no one much cares about those in a race car.

...I am putting CIS on my tweaked out corvair engine in the bus (IMG:style_emoticons/default/smile.gif)

this being the bus that weighs half as much as my 5,6-liter Titan but gets just as good mileage?

Never did I say it couldn't be made to work. As Scotty says, it's a dirt-simple mechanical injection system, but it does have a few fundamental underlying principles, chief of which is that the instantaneous position of the metering plate is proportional to the fuel requirement at that moment. If the plate is fluttering (and it does have some hydraulic damping to it) that principle isn't true. 6-cylinder engines have pretty much continuous intake flow - 4's are a little choppier by their very nature. And yes, I ran a Techtonics 268º cam in the ur-GTI and it ran great. That was generally accepted as radical as you'd like in an emissions-rated street engine - and I think it might have given the cat a workout if I'd still had one...

[Dr Evil]

There has been proof that CIS actually does fine with radical cams...

There's a whole spectrum of meanings to 'radical'... And 'fine' for that matter ;-) Lift is OK, but when you get into lots of overlap there are going to be issues. Are they issues that have workarounds? probably...

Porsche ran CIS on the 934's because the FIA made 'em do it, and they ran just fine, but not with as much power as the Kugelfisher injection they could run in Group 5. They certainly didn't have 'street-car idle' either. I suspect they gave up a lot in fuel economy at idle and part-throttle - but no one much cares about those in a race car.

...I am putting CIS on my tweaked out corvair engine in the bus (IMG:style_emoticons/default/smile.gif)

this being the bus that weighs half as much as my 5,6-liter Titan but gets just as good mileage?

Never did I say it couldn't be made to work. As Scotty says, it's a dirt-simple mechanical injection system, but it does have a few fundamental underlying principles, chief of which is that the instantaneous position of the metering plate is proportional to the fuel requirement at that moment. If the plate is fluttering (and it does have some hydraulic damping to it) that principle isn't true. 6-cylinder engines have pretty much continuous intake flow - 4's are a little choppier by their very nature. And yes, I ran a Techtonics 268º cam in the ur-GTI and it ran great. That was generally accepted as radical as you'd like in an emissions-rated street engine - and I think it might have given the cat a workout if I'd still had one...

Um, the bus comment was separated from the radical cam comment for a reason. They are not related. So, no need to argue (IMG:style_emoticons/default/wink.gif) The bis was related to the topic of adaptation, the cam remark was related to the concept that there is more cam variability available with CIS than more people think (IMG:style_emoticons/default/rolleyes.gif)

[ArtechnikA]

...So, no need to argue

twasn't an argument. trust me - if I am arguing with you, you will know it ;-)

...

the cam remark was related to the concept that there is more cam variability available with CIS than more people think...

I agree. So you don't need to argue ;-) ...
But it is something to be aware, especially if you're looking for efficiency (i.e. mileage) or need to be concerned about emissions (some do...). Overlap is the killer. I don't recall all the specs for the Techtonics 268º cam, other tha the signature duration. For 911's the sweet spot is somewhere between the '75 'S' cam and the 964 cam.

But the reality is that it's way easier to change cams in a 911 or Rabbit than a T-IV. I just want people to do their homework, think about what they're doing, and make an informed choice. CIS is a thing of beauty in its ability to provide efficient, smooth idle, broad smooth power band, and decent top end - with a minimum of constant tinkering. I used to go from 6000' to sea level twice a year in a 10,3:1 engine with no other attention paid than running premium gas at sea level - which was pretty much necessary only because I ran so much spark advance at altitude.

If you go in thinking you can run any cam you like because the CIS will adapt, you will probably be faced with driveabilty issues. Just look at the changes Porsche made between the last of the carb/MFI cars to the CIS cars. To be sure, a lot of those were driven by emissions requirements, especially in a leaded-gas no-cat environment.

[TargaToy]

Ok, Jake says it's a good system. He's the man who's done the testing.

Will this system let a 2056 make the kind of power that one with carbs would? What I'm getting at is that I often read a 2056 with D-Jet will be limited to 100-115 hp whereas a carbed motor can make 130.

I like the thought of injection AND the bigger power.

[scotty b]

Ok, Jake says it's a good system. He's the man who's done the testing.

Will this system let a 2056 make the kind of power that one with carbs would? What I'm getting at is that I often read a 2056 with D-Jet will be limited to 100-115 hp whereas a carbed motor can make 130.

I like the thought of injection AND the bigger power.

(IMG:style_emoticons/default/agree.gif) and IMHO yes. But Jake would cetainly know better than I

[VaccaRabite]

I have thought about CIS for my 2056, but have held back as I think my cam is too aggressive for it. I'd love to find otherwise.

Zach

[scotty b]

I have thought about CIS for my 2056, but have held back as I think my cam is too aggressive for it. I'd love to find otherwise.

Zach

Me thinks snot. CIS works on a basic principal. Throttle ( airflow) opens plate , position of plate increases fuel flow via plunger. Therefore the mid range cam you have MAY require a sight adjustment, but would not cause any real issue

[TargaToy]

Is this hard to integrate with a stock ECU? What are the drawbacks, if any?

[dlee6204]

I just thought I would post some links that I found helpful before I started down this road...

http://www.shoptalkforums.com/viewtopic.php?t=114309

http://members.rennlist.com/evill/ed1.htm

http://www.thesamba.com/vw/forum/viewtopic...c9d130f4f1aa114

http://www.stits.org/DMCVEGAS%20DMC%20Arch...CIS%20Thing.pdf

http://www.aircooledtech.com/CIS_injection/

http://www.thesamba.com/vw/forum/viewtopic...cis&start=0

[Prospectfarms]

I'm not sure how that system would interface with an ECU:

"K-jetronic is a mechanical continuous fuel injection system which does not require any form of drive whatsoever." "K-jetronic: a fuel injection system from Bosch"

Its not simple -- rather mechanically complex, but it is modular and not electronic so in that way it might be simpler to work on:

Here's a picture of the air flow sensor that controls fuel:

Here's a diagram describing how it enriches the mix when you stomp on the gas:


I'm not qualified to comment on how a high duration camshaft might affect the operation of the sensor, but it is clear that strong pulsations in manifold air flow might. I thought that valve overlap acted to mitigate exhaust gas resonance, permitting more fuel to enter the combustion chamber? In the Bosch K-jet material they mention overlap as an important part of their performance/economy/emissions strategy.

I want to know what those who have actually worked on it (K-jet system) feel about the durability of all those the mechanical movements over 25-30 years. Are the necessary components obsolete? Are used parts reliable?

One thing I found very interesting in my research is that later K-jet usedan overlay electronic feedback loop with a lambda sensor.

Thanks to Thinair for providing the informative links.

[jcd914]

I can't speak to the overlap limits with CIS (K-jet) but I spent a number of years working 5 days a week on cars with CIS FI.
The system as a whole was relatively trouble free. The common issues were typical of most FI problems, vacuum leaks, injectors wear or get dirty, fuel pumps fail, fuel pump relays fail and of course vacuum leaks.
Also there seemed to be a number of techs that would "adjust" the pressures to improve something (drivability, performance, mileage, etc...), this was usually a huge disaster that cost a lot to fix. Fuel pressures are used throughout the system to control everything and they all have to work together.
Fuel contamination (water/rust or dirt in the fuel) could be a huge problem and required replacement of most components that had fuel going thru them. We had quite a few cases of fuel contamination in the late 70's early 80's but after gas stations started paying some big repair bills there seemed to be a big drop in contamination cases.

It was just a few years ago I sold my last Audi quattro with CIS and it worked fine with an unknown number of miles. It had 190K miles when I bought it and the speedo cable was broke, after I fix the speedo cable I found the odometer only counted miles sometime. I drove it 7 years after that and only replaced a fuel pump relay and an ignition module, well those were the only parts related the engine running at least.

Used parts seem to be available but watch out for dirty parts.

Jim

[Prospectfarms]

Thanks Jim, I value your "narrative of experience." I take that the major cans/components are not normally serviceable and the typical cause of failure is contamination of some sort.

I can't speak to the overlap limits with CIS (K-jet) but I spent a number of years working 5 days a week on cars with CIS FI.
The system as a whole was relatively trouble free. The common issues were typical of most FI problems, vacuum leaks, injectors wear or get dirty, fuel pumps fail, fuel pump relays fail and of course vacuum leaks.
Also there seemed to be a number of techs that would "adjust" the pressures to improve something (drivability, performance, mileage, etc...), this was usually a huge disaster that cost a lot to fix. Fuel pressures are used throughout the system to control everything and they all have to work together.
Fuel contamination (water/rust or dirt in the fuel) could be a huge problem and required replacement of most components that had fuel going thru them. We had quite a few cases of fuel contamination in the late 70's early 80's but after gas stations started paying some big repair bills there seemed to be a big drop in contamination cases.

It was just a few years ago I sold my last Audi quattro with CIS and it worked fine with an unknown number of miles. It had 190K miles when I bought it and the speedo cable was broke, after I fix the speedo cable I found the odometer only counted miles sometime. I drove it 7 years after that and only replaced a fuel pump relay and an ignition module, well those were the only parts related the engine running at least.

Used parts seem to be available but watch out for dirty parts.

Jim

[Rex-n-effect]

My 81 Scirocco, daily driver, has 417,000 miles on it and i have never had a problem with the FI system. I would think that qualifies for dependability.

The later CIS did use a fequency valve (lambda control) system that you can still use. Need to set it to 50 pulses or minimum to negate the 02 sensor but still it would work without the 02.

I had been thinking of doing this also as the system is trouble free and i have access to about 50 of the cars for parts. This thread was just what i needed.


Rex

[Rex-n-effect]

I have to take that back. I had to replace the Lambda relay a couple of years ago. Damn FI anyway. (IMG:style_emoticons/default/smile.gif)