Need advice from MFI Gurus

[drwhosc]

Ok intalled my newly rebuilt MFI pump today. Mostly went well, but not quite. I can not seem to get rid of an idle lean misfire, and when I accelerate, the car lacks power but does alringt at about 3K rpms. It wil complain with some more lean misfires and some jerling when asked to accelaerate at the low end.

All my reading tellmes me that this is most likely a pump adjustment tocorrect the fuel delivary. Make sense to me, but I have made some big time adjustmets. Here is what we have done thus far.

Made sure the z1 marl and the pump mark were matched up. When we first started the car, the main rack adjustmet was lean as hell, we came counter clock wise about 4 clicks and got it to fire off. Seemd to do oK at the top end, but the idle was faurly smooth, but losts of popping. Adjusted the idel mixture, but could never get the popping completely out.

Did a test drive, and I would say the car was OK, but not a porsche. It had an identity crisis that I described above. Brought it back and adjusted the main rack with two more counter clock wise clicks, and this may have made things worse.

I have a gunston CO meter, but I can not seem to get the idle mixture below 10%. Unfortunatley I think I might have an oil burn that may be making that thing usless. Not 100% sure about that, but I have my suspicions.

OK I am reluctant to go move aggressive on the main rack. I have a 71E and I have turned it 6 clicks already. I read that I have 12 on this model year. I also think the pump may have been shipped all the way lean, but I will have to wait until tomorrow to confirm. Supertec rebuilt this for me, and they have been great and told me they will help tomorrow. I will let you know what they say.

Is there something else I could be missing. DID the CMA, and all that is OK. I have the stacks synched fairly well. I have no way to test the injectorsm and I am wondering if a bad injector could be causing my issues because most of my problems appear to be coming from one or two cylinders. Fianlly what does a rich MFI engine look like. Because I have lean misfires and poor acceleration, I am assuming that this is a lean issue. I have been 180 degrees wrong in the past

Thaks for advice if you offer it. As noted earlier, I will post a follow up.

[kenikh]

It might be a faulty microswitch.

[edmayo]

There are a hundred mistakes that could cause your problems, none of which are complicated, but all require going in an orderly sequence and minding the details. Your first error from your description is the basic pump timing, it is NOT pump mark at Z1. Rotate the engine at least one full revolution (this gets all the slack of gears chains and belts settled) Now rotate engine until you see the ignition rotor line up with the #4 spark plug wire in the cap. Now rotate about 1" further on the crank pulley until you see the FE mark line up at the case half split, NOW check to see if the pump gear lines up with reference mark. Depending on what that does we can go further or quit.

[drwhosc]

tested the micro switch and rpm transducer. they are working like the service manual says they should. but i will recheck.

about the FE mark i will check that again. we did do the rotation one full turn. not sure which mark we used, but i will respin the crank shaft and look at the mfi mark. also have to search for the timjng mark locations. the rotoron # 4 is a good tip. first i have heard of that, and it makes it easy.

[orange911]

Double check that you do not have a vacuum leak. I had this problem after reinstalling the TB's. I failed to make all the nuts tight being scared of cracking the flanges. Leak was found using carb cleaner sprayed around the flanges - little at a time. Ditto what edmayo said about timing

[304065]

Did the CMA? Pop quiz time

Any competent Porsche driver tracks faster if brakes come smoothly.

Air filter
compression loss
plugs
dwell
timing
fuel pressure
injectors
belt
correlation
smog.

What is your report on each?

Ed is right, READ CMA for the exact instructions on how to set the pump timing. You go 40 degrees beyond TDC, the instructions are very specific.

Sounds like too lean to me, what does the gunson say? I use an LM_1 with 02 sensor, Gunson takes too long for me.

The answer is not too far off, it just takes patience. Good luck!

[drwhosc]

Thanks to all who have replied.

Upper end is nice now, the pump came to me just lean as could be. After 14 clicks on the upper rack, the top end is sweet, Bottom end is still shaky, but improved.

I am going to spay the carb cleaner becase i think that is just a good idea. I did have the tbs off, and was afraid to over do the nuts. When I had the TBs off, we did resurface the bottoms to a nice uniform flat and even surface.

Checked the belt, and we did use the FE not the z1 mark, We reverified that today, that is goond news that I am not that big of a mo head. The Factory service manual was a help there, and we did follow those directions.

Compression is good. Don't have the numbers here in front of me, but not a big difference between each cylinder. I remember the numbers being betwen 135 and 140s. My number 5 cyl was lower but not off by more the 10% from the top number.

I actually decided to change to plugs tomorrow. It has been about a year and I am told the engines eat plugs for lunch

I have no idea how to check the dwell. Would love a procedure. I do have a dwell meter that hooks to the coil, and an MSD system. I am plannng on going to pertronix system after I get the gremlins out.

Fuel pressure was 14 sustained lbs.

Have no idea how to check the injectors, but a cleaning may be in order.

Belt is brand new and timing marks are there.

Stack are drawing 5 and balanced using my little synchrometer tool RPM is right at 1000. A tad bit high

Smog is an issue and yes the gunson take a long time. that is the one thing I can not get in spec yet.

Right now the engine is running better but still lean on the bottom end.

Thanks all for the help, I will try to rich it up, change the plugs and check for an unwanted air path.....

Bottom line patience is a virtue. i wish I had an ounce or two...

[orange911]

Get the flow down to no more than 4 kgs/hr with all the linkages on their stops. I like to do this with the linkages OFF so that I know they are closed. Recheck after all linkages are snapped back on EXCEPT the one from the foot throttle. If they are all still at 4, then you are good. If not, adjust till it is. You can isolate the troubled one by checking each time you connect one. 4 will reduce your idle, but I would get it as low as it will go (with all in sync) and still be around 900.

[drwhosc]

cool i like that advice. great trick - easy to do - makes sense. so one of my goals should be lowest possible synched air flow at idle.

[orange911]

so one of my goals should be lowest possible synched air flow at idle.

yes, but if the engine is starving because it is too low, give it a tad more.

Also, I am assuming that all the carbon was removed from the vacuum ports and air by-pass balance screws. Also, make sure, no make that triple sure that your dizzy is advancing per the advance curve for your car. Check it at idle, 2000 rpm, 4000, and 6000 rpm, if you can take it. I know that mine was not advancing right and I discovered lots of wear in an original 90k mile dizzy.

My advice to eliminate an extra variable, set the timing at idle with the vacuum retard removed if you do not see at least 8 degrees advance when disconnected. The vacuum if strong enough is suppose to retard the timing about 10 degrees. Now, if you have a vacuum leak or a bad vacuum pod, this will not happen.

Example: Lets say you are leaking vacuum at a TB and you do not know this. Per the tune up instructions, you are suppose to set the timing at 5 degrees retard, right? OK, you adjust the timing with the vacuum hose connected to 5 degrees atdc. Now if you pull the vacuum hose off, it should read 5 degrees btdc. But in reality with a significant leak, the timing might be at TDC or even less. This was how I knew that I had to have a vacuum leak. Also, the higher idle will produce less vacuum and also reduce the amount the pod retards.

Now, lets say you over compensated the advance on the idle timing so that it was at 5 degrees BTDC. You will now be too advanced on the rest of the curve. Now throw in the worn mechanical springs in the dizzy, and you are advancing way too fast. Dizzys are not rocket science, you just need to have a reputable shop rebuild it, if a rebuild is in order. I think this is just good maintenance every 60k miles. I send mine to Advance Distributors and have been very satisfied with his work and price. I know there are people that will charge you $300-400 for a simple rebuild, but again, its not a specialty like MFI pumps.