cis- megasquirt?

[plynch]

i dont want to sound ignorant, but i am! how does the cis /mega squirt system work? what parts do i need ? can any one give me a list? i had a stock fuel injected 77std beetle but have long since sold the 48 hp low performance stuff off.... i did under stand somewhat the system and trouble shoot the cold start valve...but i really have no experience at the fi stuff any help? in the dark here.....

[raygreenwood]

There is a CIS megasquirt system? . CIS is factory mechanical injection (no electronics at all) ...except in versions like CIS-e and CIS motronic with Lambda. Those simply have assistance from 02 sensors and electronic fuel pressure control to make them more adjustable.

Megas squirt is an aftermarket electronic fuel injection system.......unless they now have a version to slave onto your standard CIS-e, CIS Motronic or CIS lambda to make it self programmable. Ray

[plynch]

yeah this is what i wanted was an example of each... is there a web site that list a diagram of how the system works?mega squirt... that is.....

[panel]

is there a web site that list a diagram of how the system works?mega squirt... that is.....

http://www.bgsoflex.com/megasquirt.html

http://www.msefi.com/

[Piledriver]

I have THREATENED to use a MS to control the mix on my lobotomized CIS-E setup, and to control ignition.

It's sort of low on my list right now as the rediculous setup I have runs awesome...
...and I bought 65panels early Vanagon hall-sensor style ignition setup.

Regardless of what you install, a CIS-style(and rated) fuel loop is HIGHLY recommended.
(sump/feeder pump/pump)

[R2.0]

Here's a question: Can the frequency valve on a CIS-E be used to control mixture without the plate and lever arm? Or, replace the warmup regulator on a regular CIS with an MS controlled pressure regulator. If so, it could allow 2 things:

1) No more air box, allowing far more mounting options in the engine compartment
2) Better flow w/o the air flow meter (Yes, I know it's not that restricive, but it's there)

Remember, this would be using the frequency valve/warm up regulator for full range, not just trim. Anyone know the duty cycle on these?

R2.0

[raygreenwood]

The airflow meter is absolutely "0" restriction. It is roughly 4 times the volume flow of most of the cars it is on. Many people get this mental image of air "struggleing" to lift the plate...but its not true at all. The surface area is huge. Its also delicately balanced.

No, the frequency valve cannot be used to control the fuel mixture by itself. It would be the equivalent to trying to gauge the entire range of fuel mixture by a TPS.

But...the frequency valve CAN be used to replace the fuel pressure regulator.....and do a much much better job of it.....and allow tighter control of mixture across the board...on electronic injection systems.

I have been playing with something similar off and on. It uses a bank of frequency valves...say three....to constantly bleed high pressure back to the low end of teh needed range. They can be either cycled up or down...or shut off..to give instant increases or variations.

The duty cycle at idle for instance...on a Saab 900, is around 45-50%. Ray

[R2.0]

What I'm picturing is using the frequency valve to replace the function of the AFM. The frequency valve would get its input from a controller that measures airflow in an arbitrary manner - AFM, MAP, even alpha-n. In the simplest sense, that is what all electronic FI's do - take a series of inputs and calculate to get one output signal: on-off-on-off-on-off-on-off-on-off. That signal goes to injector(s), but what if it was sent for a frequency valve? That's why I was asking about the capability to eliminate the AFM in a CIS - it would be redundant. And space isn't a small concern in teh cars we tend to deal with. If I could reduce the AFM assembly and hoses to a 4" square hockey puck, I think it would be a good idea.

[Piledriver]

I like the CIS fuel delivery system... The "optimum" hack would be to drive a custom "voice coil" and directly drive the control piston, replace the airmeter.

OTOH doing that eliminates ALL of CIS simplicity, and eliminates any "limp home" operation.

[raygreenwood]

Also...and this you will not get with ANY computer driven device.....the airflow plate meter....has "infinite " positions and settings. it is a direct reaction to incoming air.
Other than space problems....I cannot see why anyone would want to replace the plate metering unit. It is a work of automotive art.
The range of adjustability on it surpasses all but the most complex EFI systems. Most of the available adjustments to the metering unit.....I have seen few people even contemplate here...cept pile of course.

the correct thing to do with metering units when stuffing them into a bug...is to get them as close to teh TB as possible. If you can all but eliminate the trunking....the accuracy rises a small amount. Ray

[Piledriver]

I said "simplicity". I meant "elegance" as well...

Sadly the CIS airbox really should be fixed to the chassis somehow, not the (rocking/vibrating) motor.

How did they mount the L-Jet meter in a bug? (I don't recall)

[raygreenwood]

The L-jet on bugs was part of the air cleaner box assembly. It was essentially bolted right on top of the TB with no trunking as I recall. That may have been one of the potential failure issues. As you note...it moves and vibrates with the engine. That can't have been teh perfect situation for the flap.
On CIS....as you note, the floating plate unit was simple and elegant. I think a modern version of this could....get away from teh control plunger altogether...and use the plate along with a very fine...either magnetic or strain gauge based measuring unit that senses plate position.

What I think we are both getting at...is that the actual reaction movement of the plate to incoming air...is a very direct and accurate dynamic. Its an excellent meter.

In certain ways I have commented...that the MPS of D-jet has some of the same advantages. It is NOT just a MAP sensor. There are newer and simpler MAPS that do a more accurate job of just metering manifold absolute pressure. But more precisley...the D-jet MPS is a differential pressure unit with a mechanical reaction linkage. It incorporates MAP and external barometric. It more accurately follows the dynamic of change that foloows throttle movement. Bits downfall was the ECU its attached to and its mechanical and connective reliability. I think the CIS metering plate is also a solution of that level of elegance. Ray

[R2.0]

Please don't get me wrong; I think the CIS system is a piece of audacious engineering, which qualifies as "elegant" to my engineering sensibility. When I first looked at the flow diagrams of the CIS fuel distributor, it took me a minute but then I had one of those "Eureka" moments.

I pose my questions for 2 reasons:

First, the whole reason we are here is that we like to play with our cars, and most likely modify them. I want to see if I can merge the MS with a CIS fuel didtributor not necessarily to create something "better", but to see if it can be done.

Second, I view the CIS airbox assembly as 2 separate systems.

a) the AFM system, which is remarkably simple, but also inflexible. All it does is translate airflow into linear motion, and it's calibration is locked in by the shape of the cone and the size of the plate. It's kind of a one trick pony.

b) the fuel distributor, which is where I think the really interesting stuff happens: individual fuel trim, etc.

Theoretically, one should be able to isolate these systems; I want to see if it's possible.

Another comment regarding resolution. I view the CIS AFM like a vinyl record, and MS like a CD. A record, like the AFM, is analog and theoretically has infinite resolution; the CD, and MegaSquirt, is digital with finite resolution. So, theoretically, the analog systems should offer better resolution and "precision". However, as a practical matter, this may not be the case. A stereo system can only reproduce the accuracy of the analog record IF everything is in excellent condition and maintained that way; otherwise the system isn't reproducing the signal correctly and the distortion goes up. The CD, on the other hand, produces an inherently "noisier" signal, but a CD player can generall reproduce it every time, exactly the same way. (It's also been argued that the resolution is small enough that the distortion cannot be distinguished by human ears, but I digress.) So we have an instance where, by the time the sound leaves the speaker cones, the CD sounds better than the album most of the time, because the album can only hit perfection a small percentage of the time and it rapidly degrades to poorer signal fidelity than the CD.

How does this translate to CIS vs. MS? The way I see it, a CIS AFM can be more accurate if it is in excellent condition, mounted on an engine for which it has been sized. If either of these go out, the AFM "fidelity" will degrade. A digital system, while not being able to match the fuel curve exactly, I believe has the capability of getting closer than the CIS AFM if the engine size is different or there are otehr circumstances (turbo, nitrous, etc.). And as the system degrades - injectors wear, etc. - the digital system can be reprogrammed to compensate; this is (I believe) harder to do with the CIS AFM.

There is another reason to look at an MS/CIS hybrid: features. Currently, the MS&S-e version of the software can do the following:

Rev limiter
Overboost protection
Mappable Air:Fuel Ratio target for use with wideband O2
Coolant temp timing advance
Inlet temp timing retard
MAPdot accel/decel enrichment
Anti-rev / Traction Control feature
Over-run fuel cut
Afterstart Enrichments
Cranking /Priming settings
Hybrid Alpha-N
Air Density Correction

True, some of these are ignition related, but most are fuel related, and should be able to be implemented with a CIS fuel distributor.

In short, I just want to play with this stuff, but before I do, I want to know if its possible, not necessarily if its a good idea. Even if I try some experiments and it doesn't work out, even failure is a useful data point.

R2.0

[mschilling]

You can do individual fuel trim (per injector) with efi also, and individual cylinder timing adjustment adv or retard on top of that.

[Piledriver]

One of the main reasons I converted over to the CIS-E metering unit was that it is current controlled... It should be a piece of cake to drive with an injector driver and appropriate sized filter, and just "fine tune" the CIS, along with all the nice other features MS brings.
(unless MS has an analog output I don't know about, which would be best)
(E also happens to be able to flow >2x the fuel of CIS basic, but I don't need that feature, yet)