Questions from a MFI Newbie
 

Hi all,

So I’m new to the MFI world. I’ve been trying to read everything I can on these systems and have a few questions regarding Correlation and Measuring Exhaust Emissions in the Check Measure Adjust Manual.

Regarding correlation, I’ve saw the Porsche came out with a revised correlation process that does not require the protractors. I’ve also read Lee Rice’s article regarding correlation as well. From what I can tell, the procedure is as follows:

After checking all the previous items in the CMA manual, one needs to synchronize the throttle plates by using a Synchrometer (or something similar). Remove linkages, set all the air bleed screws the same, read all six stacks with Synchrometer, adjust the air flow by adjusting the throttle plate screw until they are all pulling the same air. Set the push rod between the MFI and the throttle crossbar to 114mm +/- 2mm, then adjust and reconnect the remaining push rods between each of the throttle butterflies and then to the throttle crossbar without an increase in RPM that was set in the previous set using the Synchrometer. Does that sound about right?

Regarding the measurement of Exhaust Emissions, it appears that a protractor is needed for this step. One needs to use the protractor attached to the bellcrank on the MFI pump so that the throttle can be set to an opening angle of 7 degrees. Is there another way to do this?

I’ve also read that you don’t really need an air-fuel gauge to perform the Part Load and Idle Speed measurement. All that one needs is an O2 sensor and a multimeter. Has anyone done this before? I would be setting up an MFI for a 2.2 1970 911E.

Thanks

Chris

Pm sent

No.

Doing it this way creates the real danger of having one or more of the throttle plates making contact with the throttle body bore when at rest in the idle position. They absolutely need to be resting on their stop screws, not the throttle body bores. The only reliable way to set the resting position of the throttle plates is with the throttle bodies removed so you can verify their position. It is imperative that they not rest on the throttle body bores. If they do rest on those soft magnesium bores, the vibration transmitted to them when the engine is running at idle will damage that bore.

It is true, however, that no one (or at least no one I know) actually uses the protractors. It seems a superfluous step. With the 114mm rod disconnected, and both the left and right rods to the throttle bodies from the cross shaft disconnected, and the fore and aft rods on both left and right disconnected, simply make sure all throttle plates are resting on their screw stops. In other words, simply disconnect all rods so the throttle plates move independentently from one another.

Then simply reconnect all rods, starting the two short ones on each throttle body. Make sure that when you connect any one rod that you have not introduced any preload in either direction - adjust each rod accordingly, until it snaps freely in place without moving anything.

Once the short ones on each throttle body are connected, skip to the 114mm rod and connect it, setting the cross shaft position relative to the pump. The pump needs to be on its idle stop. Now you have the throttle bodies with no preload, and pump to shaft with no preload. Once both of those are established, you can connect the cross shaft to the throttle bodies on each side. Again, with no preload. Don't be surprised if these two rods vary considerably in length - CMA allows up to 5mm variation.

The entire system should now be sitting with no preload anywhere, with the pump on its idle stop and each throttle body on its idle stop. From here, follow the Synchronizing Throttle Valves procedure outlined in CMA, using the air bleed screws.

I don't know anyone who does it this way anymore either. I think the CMA procedure was a product of the times, when all they had available were CO2 meters. I use exhaust gas analyzer that provides A/F ratios via an O2 sensor in the tailpipe. Mine is the old Innovate Motorsports LM1 model, but there are more modern versions available. I set the A/F ratio under load at mid to high RPM to what I feel is "safe" (between 12.5:1 and 13:1 at the most) and let the rest of the operating range take care of itself.

The idle mixture adjustment doesn't really have much of an affect. The main rack adjustment pretty much overwhelms it. I only use it after the main rack adjustment is where I want it, and only to smooth out the idle. I don't get too hung up on the idle mix - it's not all that important, and just as soon as the motor is up at any "driving" RPM range, it really doesn't have an affect anyway.

This was all written before the invention of AFR meters. I recommend you install one and make adjustments accordingly. Once you understand how the MFI works you can dial it in for idle, off idle, part and full throttle. There is an inherent off idle lean spot that you can address but is just part of the car.

Make sure everything is connected and working properly (like the hose to the thermostat).

AFR takes all the guesswork out.

Chris
73 911 E

Jeff,

When you set your MFI at A/F ratio under load at mid to high RPM between 12.5:1 and 13:1, is this at partial or full throttle?

When going up hill partial throttle, how high does your AFR go?

Your MFI open heart surgery post is one of the most interesting and informative topic on MFI

Thanks
Marlon

I'm sorry Marlon, I could have been a bit more clear. 12.5:1 to 13:1 is as lean as I will let it go under any load, at any RPM, at any throttle position. So, full throttle high RPM, part throttle going up a hill - no difference, I consider that my limit. In setting mine up under this parameter, I accepted the fact that under some conditions (low RPM full throttle), it would go a bit richer, like as rich as 10.5:1. This is, however, for an exceedingly brief instant, and it will only do this if I "allow" it to happen by being lazy on the gear shift.

Trailing throttle is a different story. Even though I have removed the decell cut off solenoid, I still see A/F ratios well over 18:1 when I am, say, coasting down a hill on the freeway or something. Just as soon as I even think about tickling the throttle, however, the A/F returns immediately to a proper range to be under load.

Jeff, No worries and thanks for the clarifications. Time to play with the AFR again. Marlon

Currently, I know a few of the throttle plates are making contact with the throttle body because they are not resting on their stop screws. So what is the proper way to setup the throttle plates before syncing the throttle valves? Is the only way to remove the throttle bodies? Or is there a certain number of turns on the throttle plate screw I should make once it makes contact?

Chris

1/8-1/2 turn, depending on how much wear is in the system. Once you are up to operating temp, loosen the rods and see if the throttle plates are still free, not sticking. A bit of expansion from the heat can cause the plates to bind. A lot of laying on of hands and praying to get it up and running. But once dialed in, they are beautiful. Have 3 engines with that system, one on a street car that has been virtually trouble free for 25 years.

Be careful. Lean is fast but too lean will cause detonation. Its tempting to go as lean as possible. I ran higher compression with a DC 30 cam and struggled with detonation as I tried to get everything out of it. Would log AFR every drive and make changes to try to get the cam matched to the MFI. That engine was a screamer, FAST!

Chris
73 911 E MFI

Yes, indeed - be careful. I was not as careful as I should have been and paid the price. My initial build ran a single plug, too much advance, and too lean of a mix. No single factor all by itself would probably have damaged anything, but the combination of the three proved fatal. I had several thousand happy miles on it on the street before everything went "critical" at a track day. Just running it that much harder, that much longer... Undetected high RPM detonation... Don't let this happen to you:

http://forums.pelicanparts.com/uploa...1601074943.jpg

Ouch! I'm doing a MFI conversion on mine this winter. Will read your advice carefully. I'll probably have Gamroth do the tuning on his dyno.

Jeff will get it right.

Happy ending, though - today marked 90,000 trouble-free miles since I repaired that damage. :)

One of the checks is to check fuel pressure to the pump. Can someone tell me the symptoms of incorrect fuel pressure especially to high? I've checked my pressure and I'm at the upper limit of the allowable but I don't really know the accuracy of my gauge.

Thanks

Chris

Unless it's very, very, high, you should not have a problem. If you are within the limit,call it good.

It's really difficult to generate too much fuel pressure from the electric pump. There really is nothing for the pressure to "back up" against - it's a recirculating system, with unused fuel simply returning to the tank. If pressure is too high, there almost has to be something restricting that flow back to the tank.

1970 911E has steel fuel lines. I have seen them partially blocked so worth checking if pressure is above 15psi or so.

What are your fuel pressure numbers. You are checking between the fuel filter console and the inj pump. Volume is a liter in 30 sec. All of these issues are addressed in ''Open heart surgery'' I think that is the files name. A really well done piece with much info and I think can be found here on PP