CIS under the new rules

[Eric Parham]

Okay, I just read through the IT specs, and most of the GCR, for the first time in about 4 years. I see that we can now completely replace computers, add MAF and/or MAP sensors, and some other key bits. Unfortunately, I didn't see any provision to swap out mechanical injectors for electronic injectors. Did I miss it?

Assuming I didn't miss it and there's just no provision for electronic injectors yet, can CIS-basic (K-Jetronic), CIS-Lambda (KL-Jetronic), CIS-E (KE-Jetronic), or CIS-Motronic (KE-Motronic) cars take much advantage of the new rules in current form? If so, what might be considered within reason and worthwhile? The very early CIS-basic had a mechanical computer, namely the CIS metering unit. All of the later versions had metering units with some electronics attached to electronic computers. For all versions including the early basic, the cold start injector was electronic, but typically not well placed for even fuel distribution.

 

[shwah]

We can do anything we want with our computers too. No one can change fuel injection hardware.

I have found CIS-E to be very easy to achieve optimum air fuel ratio with, but will be testing a digifant/electronic injection system with an alternate ECU to see if there are any gains to be had. Progress is slow due to grad school and family commitments, but I expect to know the answer sometime this eyar.

 

[Eric Parham]

Thanks for pointing out that the CIS-E was easy to keep to optimum AFR. Maybe there's nothing more to be had there. Did you mean open loop and/or closed loop? I thought that CIS-E still went open loop at WOT, but don't have my Bosch manual handy this month. I'd also imagine that taking better control of the ignition timing might possibly be worth something over the OEM knock-sensing ignition.

I am fairly certain that there are some gains to be had for real EFI (e.g., Cali Digifant) vs. CIS, for those cars that can switch (e.g., A2 chassis 8V), but that wasn't really my question.

The question is more for those cars that have no choice (e.g., any A1 chassis Rabbit/Scirocco (non-Cabriolet) or any A2 16V), what can be done for CIS/L/E/Mot within the new rules and is it at all worthwhile? For example, I could imagine using a MSnS to control the CIS-E/Motronic differential pressure valve more easily while also improving ignition timing a bit. The big question is, is it worth the trouble?

 

[GKR_17]

The very early CIS-basic had a mechanical computer, namely the CIS metering unit.

Do we really need to define a "computer"? By that logic, a carburetor is also a computer.

 

[GTIspirit]

For example, I could imagine using a MSnS to control the CIS-E/Motronic differential pressure valve more easily while also improving ignition timing a bit. The big question is, is it worth the trouble?

I studied using a MS for CIS-E but didn't see a way to make it work with the existing software. I think that a single PWM driver would do it but the software wasn't configured for that.

As far as ignition, I see the four window hall effect sensor as the biggest limiting factor. It would be the equivalent of a two tooth crank wheel..... So while MS could give you the ability to change ignition timing across the map rather than rotating the distributor to make a global shift, the inherent inaccuracy of the four window hall effect sensor is a huge handicap.

One of the last Fastrack issues solicited for member feedback to allow "crankfire" ignition which would really help put cars from the distributor era on equal footing with newer cars that came from the factory with a crankshaft trigger wheel. It would also help to make it easier to adapt modern aftermarket ECU's to CIS and CIS-E cars. So hopefully everyone here made their opinion heard on this topic and wrote in.

 

[evanwebb]

I dunno, even though it would potentially help me I see the crank fire ignition allowance as a big departure for the IT rules. Not even limited prep Prod can add crank fire!

 

[pfcs]

You miss the point my dear Watson-

"As far as ignition, I see the four window hall effect sensor as the biggest limiting factor. It would be the equivalent of a two tooth crank wheel..... So while MS could give you the ability to change ignition timing across the map rather than rotating the distributor to make a global shift, the inherent inaccuracy of the four window hall effect sensor is a huge handicap"

The 4 slot sensor's as accurate as any crankfire (excepting any scatter in the intermediate shaft/timing belt/distributor bushings which in a well prepped VW is less than a degree)at 4 critical points-where it fires the ignition.
And in these IT apps, there's no need to vary ignition timing over the working (racing) power band at WOT-believe me.
Of course, without a 60(?) slot crank signal, you can't do some things with the more sophisticated aftermarket DIY injection systems, but if you're so smart, they shouldn't be needed!

 

[Flyinglizard]

The d fant AFM passes more air than the CIS door . But the D fant cam passes less air than the CIS cams. .
The advantage of the digi seems to be that , if there is a failure the whole rail shuts off. The CIS will burn off the top of a piston on one straight away, if the injector fails or the fuel dizzy fails/Plugs.
We have been over the 4 window thing a lot already. I support three oval cars that run 13-1 compression . We cant afford any spark scatter and have never seen any with the stock 4 hole dist.
The knock box can ruin an engine on over run, if the vac line is plugged in.
I have had the best results using a Diggy dist with no knock box, fixed advance. There may be gains to be had under our racing rev range or 4700 to 7300.
No crank triggers. bad road to go down. IMHO. MM

 

[GTIspirit]

No crank triggers. bad road to go down. IMHO. MM

While I agree with you in principal, and share your concerns about the Prod-car-ization of IT, the reality is that crank triggers ARE allowed in IT, if the car came from the factory with a crank trigger wheel and sensor.

While I also understand and agree with the philosophy that you pick your car, warts and all, the no crank trigger rule creates a rift between the newer cars and older cars.

I'm not proposing that the distributor can be changed out for coil on plug or similar, or that CIS injectors can be replaced by EFI injectors. IMHO that is a clear departure from class philosophy.

I'm just asking that older cars be granted an allowance to use a sensor that comes from the factory on newer cars. Is that not fair to ask? After all, engine output is still mechanically limited by intake, cams, exhaust, compression ratio, etc, all of which are tightly regulated.

 

[Eric Parham]

Anyone know the current email for the compboard (I tried [email protected] and it bounced)?

I just want to point out that we can already put a nice trigger wheel (e.g., more than 4 slots) inside the distributor. As for the less than 1 deg if everything's tight -- that's often true (at least if you haven't tanked the block and forgotten to replace the aux shaft bushings), but this can get worse for some engines due to a narrower drive gear with more inherent slop.

9.1.3.D.1.e "Any ignition system which utilizes the original distributor for
spark timing and distribution is permitted. Internal distributor
components and distributor cap may be substituted. Crankfire
ignition systems are prohibited unless fitted as original equipment.
Any spark plugs and ignition wires may be used. Ignition
timing is unrestricted."

 

[JoshS]

You can write to the CRB at http://www.crbscca.com.

 

[shwah]

I don't think trigger wheels are consistent with IT philosophy.

I also don't think there are gains to be had with them. 4 window works well. We do not need very much timing variation in the racing rev range. If someone wants to they can install a wheel inside the distributor. I know that motorcycle guys have even run smaller number -maybe 12-2 wheels in fuel injection conversions...

 

[Eric Parham]

How about:

9.1.3.D.1.a.6
"The engine management computer may be altered or replaced. A throttle position sensor and its wiring may be added or replaced. Crankshaft and/or camshaft position sensors and wiring may be added or replaced. A MAP or MAF sensor and its wiring may be added. Other existing sensors, excluding the stock air metering device, may be substituted for equivalent units."

9.1.3.D.1.e
"Any ignition system which utilizes the original distributor for spark XXXXXXXX distribution is permitted. Internal distributor components and distributor cap may be substituted. Crankfire ignition systems are XXXXXXXXXX permitted. Any spark plugs and ignition wires may be used. Ignition timing is unrestricted. Batteries . . .."

 

[Eric Parham]

I'm not proposing that the distributor can be changed out for coil on plug or similar, or that CIS injectors can be replaced by EFI injectors. IMHO that is a clear departure from class philosophy.

I'm just asking that older cars be granted an allowance to use a sensor that comes from the factory on newer cars. Is that not fair to ask? After all, engine output is still mechanically limited by intake, cams, exhaust, compression ratio, etc, all of which are tightly regulated.

I agree with your position that older cars should be allowed to add sensors that come standard on newer cars. Although I personally disagree that older cars should not also be allowed to add actuators (e.g., EFI injectors for CIS, or coil-on-plug for H.T. distributor, etc.) that come standard on newer cars, perhaps we're not quite ready for that yet.

 

[shwah]

What you are arguing would be to allow carbed cars to add efi as well then?

The sensor rule is only relevant to sensors, not hard parts or wheels. The distributor rule lets you do anything you want inside the distributor. With those 2 together you can make a position sensor in the distributor.

 

[Eric Parham]

At this point, all I'm arguing is for a crank or cam position sensor to be allowed as a potentially more accurate and cheaper input into the recently freed ECU. As it stands, I think we basically have another "within the box" rule like the old ECU rule, but the box limitation has just been moved to the distributor for cars that lacked an adequate crank and/or cam position sensor (or trigger wheel) from the factory.

Not sure what you mean by "the sensor rule". If sensors were completely free (which they are not), I think that would include trigger wheels. On the other hand, existing flywheel teeth could be equally useful if a proper sensor could be added.

We agree on the distributor position sensor being allowed, but all of my reasoning for that comes from the ignition section.

 

[GTIspirit]

At this point, all I'm arguing is for a crank or cam position sensor to be allowed as a potentially more accurate and cheaper input into the recently freed ECU. As it stands, I think we basically have another "within the box" rule like the old ECU rule, but the box limitation has just been moved to the distributor for cars that lacked an adequate crank and/or cam position sensor (or trigger wheel) from the factory.

Not sure what you mean by "the sensor rule". If sensors were completely free (which they are not), I think that would include trigger wheels. On the other hand, existing flywheel teeth could be equally useful if a proper sensor could be added.

We agree on the distributor position sensor being allowed, but all of my reasoning for that comes from the ignition section.

^2. Well said, exactly my position just stated differently.

We've already agreed in separate threads that a toothed crankshaft wheel is allowed under the pulley allowance. And I thought we already agreed that there is no provision for the crankshaft position sensor, sensors are only free for gauges and data acquisition, not for engine management.

For example, WB02 is allowed for gauge display and data acquisition purposes, but not as an input to the ECU, unless of course you have a newer car that came from the factory with a WB02 input to engine management.

P.S. I just received word back from the SCCA that the letter I submitted on this subject a few weeks ago, in response to Fastrack asking for feedback on this topic, was reviewed and a proposal set forth to be voted upon in the next board meeting and published in the next Fastrack.

 

[Eric Parham]

Good stuff. Can't wait to see how it turns out!

Now let's get back to the CIS issue. I'd love to control it with MegaSquirt. I recall from an older thread that someone made CIS-Lambda work with a proprietary box, but that wasn't open source and wouldn't even work for CIS-E or CIS-Mot. I'm also unconvinced that it would be fast enough for transients, although steady-state would have to be better. It's too bad that no one seems to have been able to get the CIS-E/Mot differential fuel pressure regulator (DPR) working with MegaSquirt. Maybe I'll make a project of it, but first let's consider an alternative.

While most EFI systems primarily try to regulate fuel pressure and modulate injectors to control fuel flow, CIS generally controls fuel flow by changing fuel line pressure to the injectors. I wonder if this line pressure could be modulated (i.e., turned off and on) to control flow using off-the-shelf MegaSquirt in a non-sequential mode as an alternative to controlling the DPR. The later CIS-E/Mot systems might be physically capable of it (and already have a built-in airflow plate potentiometer), but would probably not be accurate enough due at least to the uneven lengths of the injector lines. Fortunately, "fuel lines may be replaced, relocated" (see 9.1.3.D.1.b), presumably with equal lengths

Also, 9.1.3.D.1.b (final sentence) seems to permit installing "an external fuel pump pressure regulator". Is that different than a fuel pressure regulator (FPR)? Does "an" limit this allowance to one? If the car already has one, can a different one be substituted or added? Could an original one be substituted AND an additional one be added? I don't see any limitations on placement other than that it be external to the fuel pump. I also don't see any limitations to a purely mechanical device, so I think it can be controlled by vacuum and/or electronically.

Are we there yet? Worth a try?

 

[GTIspirit]

Back on topic.
I've been thinking about this very topic for a long time now and don't have a good solution. I believe that all you should need is a PWM output to control the DPR valve. If you can control the duty cycle to the valve then you are in control of the current to the valve. And not very much power is required, the DPR valve is around 18-20Ohms and needs at most 150mA, around 0.5W.

I don't think it will work to just tie the injector outputs together to get a periodic signal, probably not fast enough and the DPR might actually react to the individual pulses of the injector.

So to do this the injector pulsewidth output from the MS would need to be converted to a duty cycle, 0-100%. The trick will be getting enough resolution, on my PL engine at WOT the DPR current was around 22mA and idle is around 5mA. So not much current is required, and AFR definitely responds to a 2mA change in DPR current.

The other thing to consider was the good point made above about overrun fuel cutoff. If the DPR current is not adequately reversed and there is some fuel dribble on overrun that could be a very bad thing to have such lean burn. If the injectors are completely closed off and the engine is just pumping air on overrun then no problem. But I can see how some small fuel flow could create big lean burn problems.

 

[Eric Parham]

I believe that all you should need is a PWM output to control the DPR valve. If you can control the duty cycle to the valve then you are in control of the current to the valve. And not very much power is required, the DPR valve is around 18-20Ohms and needs at most 150mA, around 0.5W.

Is the OEM DPR input pulse width modulated (PWM) or continuous? I had been thinking that we needed to give it a continuous mA signal.

I don't think it will work to just tie the injector outputs together to get a periodic signal, probably not fast enough and the DPR might actually react to the individual pulses of the injector.

Probably not, but still might be worth a try.

So to do this the injector pulsewidth output from the MS would need to be converted to a duty cycle, 0-100%. The trick will be getting enough resolution, on my PL engine at WOT the DPR current was around 22mA and idle is around 5mA. So not much current is required, and AFR definitely responds to a 2mA change in DPR current.

The Power Module ups that DPR current a bit. I think there are half a dozen settings, but the higher ones are in the 40mA range, IIRC.

The other thing to consider was the good point made above about overrun fuel cutoff. If the DPR current is not adequately reversed and there is some fuel dribble on overrun that could be a very bad thing to have such lean burn. If the injectors are completely closed off and the engine is just pumping air on overrun then no problem. But I can see how some small fuel flow could create big lean burn problems.

Agreed. I believe that a sufficiently negative input (don't remember how much, maybe -10, maybe -50) to the DPR does cut fuel completely, but this would have to be verified.
On the other hand, it would probably run better under racing conditions (if fuel usage wasn't an issue) if the AFR was simply maintained.