MFI troubleshooting - high idle, pops and bucks at part throttle

Car - 1972 911T 2.4 MFI, engine all stock

Background - Years ago (at least 3 owners ago) MFI was swapped to carbs, but parts boxed and retained. Previous owner had engine rebuilt (all stock spec) and MFI put back into service. Project stalled immediately thereafter. I bought it as a running car, but with only 2-3 hours on the engine, and no road testing (due to bad brakes).

Now that the car is road worthy again, I'm looking at some of the basics in order to correct running problems. Main problems are high idle speed (~1300) and popping and bucking at part throttle. Smooths out and pulls nicely at wider throttle opening. These problems only surface when the engine is fully warm. When cold or during warmup neither symptom is evident.

Keeping the old Check-Measure-Adjust doc in mind, the ignition and engine bits are good. New air filter, plugs, points, plug wires, good compression, fresh valve adjustment. Timing appears good, and the vacuum retard is working as expected.

So I need to look at getting the idle speed down, and perhaps richen(?) the part throttle on the pump.

So first I look at the idle. And it is my understanding that the idle is adjusted with the 6 air screws on the stacks, and then the synch is re-checked.

But there is my first failing. No amount of changing on the air screws appears to have any effect on the idle speed. The screws were all over the place when I started, after trying to drop the idle speed for a bit with no result, I stopped the engine, screwed all of them in to the stop, then backed each out 2 full turns to have a starting baseline.

Still idles at 1200-1300 when warm. All of them check fairly close on my synch tool, but still no idle change.

Any thoughts?

Do you have some idle after the main rack adjustment? You should. Then with the idle rack (below 1500 rpm or so) you can further tune the idle up or down. Then go to the fine tune with the idle air bleed screws. The recommended screw settings are just that, recommended. If you have CMA follow it to the letter. It works. I followed Grady Clay's advice and finally figured it out when realizing the main rack has substantial influence on basic idle.

Start with the basics, timing must be set correctly, and then the downfall of most is the basic mechanical linkages. I can guarantee they aren't right. Somewhere there are throttle air valves being held open. And probably the pump rod is the wrong length.

Yes and what Ed says.

Just a quick ask,
How old is the petrol in the tank ???

Thanks for the info so far.

Quote:

---

Originally Posted by NZVW

Just a quick ask,
How old is the petrol in the tank ???

---

It's fresh. Tank was drained when removed for the front pan replacement.

Quote:

---

Originally Posted by edmayo

Start with the basics, timing must be set correctly, and then the downfall of most is the basic mechanical linkages. I can guarantee they aren't right. Somewhere there are throttle air valves being held open. And probably the pump rod is the wrong length.

---

Timing is good at idle, both with vacuum connected (5° ATDC) and disconnected (~5° BTDC). I'm reluctant to check at 6000 rpm as the engine is still fresh and not broken in. Been trying to keep it below 4500 or so until I get some miles/hours on it.

I'm not concerned about the function of the thermostat or the warm air supply from the HE, as there is a marked change in behavior as the engine gets up to temp, that seems to indicate the thermostat is working.

I pulled the pump rod to measure it, and it is correct at 114mm. A little surprised to find that was correct.

Tomorrow I'll pull all the rest of the linkages, check and set the throttle stops.

When I used my car every week I never had any MFI problems. Now that I leave it stored for a couple of months at a time I am having some small cracks and pops when I first use it. I run a few tankfuls of fuel with stabil marine and take it for a couple of long drives and the problems go away. Stuff just gets gummed up. I now make sure to leave the tank full and with stabil when I put it away. Stuff gets gummed up.
Have you cleaned out the idle screens ports?
All this in addition to all of the above comments.

OK, today's topics.

Checked the pump timing, it was spot on. Not too surprised to find that, the pump was installed and set up while the engine was out and the timing marks were easy to see.

Took it for a drive to get it up to temp.

Got back and pulled all the throttle connecting rods off, reset all the throttle plates. The plates were not set too badly, most were pretty close. Started it up and fine adjusted them.

Started to check and re-install the connecting rods then, found that one of those on the right bank was a touch long and was probably holding #5 open a tad. Corrected that.

Next went to the 2 throttle pushrods, and found what I suspect was the root cause of the fast idle. The left side rod was too long and was lifting all 3 throttles on that side just barely off the stops. Re-adjusted it to stop that from happening, then rechecked synchronization. Still good.

Re-checked the pump rod for 114mm center-to-center length. It's right there, or as close as I can measure it. So I popped it back on as well.

Results of today's work? Mixed. The fast idle is definitely corrected. I don't 100% trust the in-dash tach, but it shows a touch under 1000 rpm, and it sounds right too. But the mixture is apparently a bit worse now. It pops and spits a fair bit just while idling, and the popping and bucking while driving is worse.

So my next steps are???

Should I attempt anything on the pump rod, or assume that it is correct and adjustments are needed elsewhere? If I leave the pump rod as is, what should my next step be?

I have the same experience with high idle after warm up. My tech says it could be the throttle bodies are sucking air past the throttle plates because of ware. Have yours been re built? Chris

Quote:

---

Originally Posted by raspy2point2

I have the same experience with high idle after warm up. My tech says it could be the throttle bodies are sucking air past the throttle plates because of ware. Have yours been re built? Chris

---

I don't think so, but don't know for sure. Indications are that my throttle shaft bushings are not leaking, that was one of the first things I looked at. But based on my experience today, I would warm your car up and then pop off all of the various pushrods and see where it idles with no throttle connections. And then reset all the throttle stops while you are in there. If one or more of those rods are not correct, a high idle can be the result.

But maybe your tech has already done that.

Pull your spark plugs see if they are black or white (rich or lean) Ideally you will have an AFR meter, but it can done without, just takes longer. If the plugs look lean, which I think they will be, go two clks richer on the main rack adjustment (counterclockwise) if car can be driven, accelerate slowly but with steady throttle opening in second gear going slowly up to 5000 RPM. Assuming it is better but still not right repeat two clks richer adjustment and repeat test. If you are going in the right direction the performance will be getting better not worse. Ultimately your goal is to be able repeat this acceleration test with the engine pulling smoothly up to 5000. Generally when I see these cars someone has gone way to rich. In that case I will keep leaning rack two clks at a time until I know it has gone too lean, then two clks at a time richen until that acceleration curve is smooth. Then that last adjustment is the idle adjustment, which without a meter you will simply have to do by ear, feel, and idle speed. I use an AFR meter of course, but I could get very close by this seat of the pants method.

By the way, generally 72-73 throttle bodies were redesigned from the previous ones that wore prematurely and so didn't wear out as fast.

Arne-

Is there play in the throttle shafts when you try to move them up and down?

Quote:

---

Originally Posted by Scott Clarke

Arne-

Is there play in the throttle shafts when you try to move them up and down?

---

Nope. As I noted earlier, that was one of the first things I looked at when I started this, due to the high idle speed, now fixed.

Arne -
Not sure if this will help u but lots of decent research on the bird from back in the day. I went through it when messing with our MFI and some was helpful.

https://www.pelicanparts.com/techart...MFI/TipMFI.htm

Ed Mayo had some great information. I would add, not to play with the 114mm rod.

Quote:

---

Originally Posted by Scott72

Arne -
Not sure if this will help u but lots of decent research on the bird from back in the day. I went through it when messing with our MFI and some was helpful.

https://www.pelicanparts.com/techart...MFI/TipMFI.htm

---

I had seen that page, and downloaded the PDFs from there. Definitely worth reading.

Quote:

---

Originally Posted by 2.5MFI

Ed Mayo had some great information. I would add, not to play with the 114mm rod.

---

Thanks, Mark. I gathered that I should leave the pump pushrod alone, and move on to other adjustments.

I am still a touch concerned with the popping at idle. Some of the info on the above linked Bird page seems to indicate that this popping at idle may be due to slight mis-adjustments in either throttle plates or air screws. Should I try to eliminate that before moving to the main rack adjustment?

You can try resetting just the idle circuit adjustment, but remember, adjustment of the main rack affects the entire range, an idle adjustment only affects idle and slowly phases out around 2-3000 rpm. That is why you try to make main rack adjustments first, then finish with idle. Hard to describe feel and experience via cyber space. When you said you measured the pump rod at 114 I assume you meant center to center,,,,not overall. Measure overall, then measure o.d. diameter of socket, then subtract from overall, that will be center to center.

Thanks, Ed. Yes, the 114mm was measured center-to-center. I have the old Lee Rice articles that also gave that method for determining the length.

I don't think I want to go out of order on the adjustments, will do the main rack first and only go to the idle circuit once I'm happy with the main rack setting.

Just unsure on the popping at idle. Is that from mis-adjusted throttle plates, or pump adjustments? Obviously no one can make that determination over the internet...

Might try to further fine-tune the air screws to reduce popping before moving to the main rack adjustment.

Pulled the plugs this morning to verify where I was prior to going further. Two surprises found. The minor surprise is that the plugs are NGK, but not the BPR6ES (or iridium equivalent) I was expecting, but BPR7ES which is one step colder than I had guessed for this car. EDIT - Looks like BPR7ES may be correct for a 2.4T?

The bigger surprise is that all 6 plugs (gapped at .032") are black and sooty - dry fluffy soot, not wet or oily. Not what I was expecting at all. EDIT 2 - Info I have is that plug gap should be .024", so I will probably want to adjust that.

So despite my gut feel, it appears to be running rich, not lean.

Stopping here to wait for other opinions.

I use the BPR7ES for all my 'S' cars, they should work fine even for a 'T' but a BPR6ES would be a better choice for a 'T'. That was why I always start by looking at spark plugs, gives you a better idea of where to go. I would start now by leaning the idle adjustment only, we know the main rack is likely too rich already. The idle adjustment is the lower pump adjustment , reached thru the fan blade and the hole in the shroud. Push in and turn until you feel the tang drop in the slot, then turn clockwise to go leaner. Go about 4 clks at a time. If leaning the idle seems to help you may need to go leaner on the rack since that leans out the whole system otherwise the idle adjuster may run out of adjustment range. And yes, I use the .024" gap.

This afternoon did not go well. I may well be the poster child for why average home mechanics should not mess with MFI. :(

Despite my best intentions of keeping track of where I was and what changes I was making, I believe I am well and truly lost. I suspect that at least once, when I was slowly turning the idle adjuster waiting to feel the tang drop, it had already dropped and was randomly turning the adjuster. I now have no clue where I am, or how to get to some sort of baseline.

As of now, the car is hard to start, doesn't want to idle, smokes black when at what passes for idle, and pops a lot when held at higher revs (2k to 3k) with no load. It is no longer warm enough to make good adjustments, but isn't running well enough to be able to drive it to get it warm.

Definitely seems very rich at low RPM, no longer sure what it's like at higher speeds.

Despite the thinking that the main rack was rich, I'm still thinking that it may have been lean. (Who knows what it's like now.) I say this because while driving it to warm it up to temp this afternoon (before I screwed it all up), it ran great until the thermostat dropped it into full lean. Pulled strong and clean, no popping or bucking, just like I feel it should run. And then when it finally got to temp, it got bad. Popping, bucking, won't pull up a grade, way down on power. Still made me think the main rack setting was lean.

But now? Who knows where I'm at.

Any suggestions on how I recover from this and get to some sort of starting point?

Don't get discouraged . your description of the running when cold and then when the thermostat is warm certainly seems like a lean condition . I think any MFI car that has not been driven for a long time needs to be adjusted
richer for todays fuel . My '73 E had not been driven since 1980 . I had to turn the screw 10 clicks to get a decent AFR reading . It seems like the idle is too rich , so turn that screw counterclockwise .

It is all downhill to Eric Jones. It would make for a handy tech session car.

Haven't talked to Eric about it yet, since he is racing this weekend and I only messed it up today. He did comment that he hadn't worked on any MFI in a long time...

Quote:

---

Originally Posted by edmayo

Push in and turn until you feel the tang drop in the slot, then turn clockwise to go leaner. Go about 4 clks at a time. If leaning the idle seems to help you may need to go leaner on the rack since that leans out the whole system otherwise the idle adjuster may run out of adjustment range. And yes, I use the .024" gap.

---

Ed, Arne, The idle mixture adjustment your talking about will go "Richer Clockwise". Leaner is CCW. Now the main rack adjustment is the opposite, CCW for richer and CW for leaner.

Wait, what?

Oh, crud...

Thanks for pointing that out, Mark. I actually knew that the two adjusters go in opposite directions. I've got the Lee Rice MFI articles in PDF from the Pelican link. Read them multiple times, so I had the "opposite directions" bit down pat. But I missed that Ed crossed up the direction above, and I kept using that as my guide. Oops.

On the plus side, this gives me some hope. Basically, everything I've done has been going the wrong way, making things worse. That means there is some reasonable hope of getting back to a decent starting point. If I back out of everything I think I did, I'm going to be far closer than I am now.

Wish me luck.

Popping out of the intake or exhaust? Out the intake is likely lean, out the exhaust is likely rich. If you run too long on the rich side you might want to be ready to throw a new set of plugs in it otherwise you could be chasing your tail.

Well, better today than yesterday. After reversing what I did yesterday I'm more or less back where I started, only without the high idle speed.

But once I got back to that point, I haven't been able to get much improvement. Warm or cold, the idle is quite slow, rough and stumbling. But the other idle symptoms are mixed—exhaust is dark and somewhat sooty, yet it pops and spits in the intakes.

As for the main rack adjustment, I believe that is still lean. It actually runs great while the warmup thermostat is still enriching the mixture, pulls smoothly and cleanly, no popping, sounds great, etc. If it ran like that all the time I'd be a happy camper.

But as soon as the thermostat gets fully warm, it becomes bad. It pops, spits and bucks (intake, not exhaust), won't pull under load, just not a happy car.

Does any one (Mark?) know about how much enrichment the thermostat does on the main rack? It would seem that I need to go rich on the main rack to about that about. Then once that is running clean, try to figure out what to do with the idle circuit.

For now, I've called it a day. Tomorrow I'm planning to go back and double check all the throttle plate settings and pushrod lengths. I've done it once, but it is worth another look. I'd like to ignore the idle circuit for now and try to dial in the main rack. But since I need to enrich the main rack, that doesn't seem to help the perceived too-rich condition at idle. So I may need to do them in tandem.

Open to other ideas, too.

I have to ask. Have you studied this?

https://www.pbase.com/slidevalve911r...docs_bosch_mfi

Quote:

---

Originally Posted by 72targa

I have to ask. Have you studied this?

https://www.pbase.com/slidevalve911r...docs_bosch_mfi

---

Yes, I have. I downloaded a copy some time ago. I have referred to it, but some of the other guides seem more useful to home mechanics.

Sorry Arne, I knew that,,,,just didn't proof read my post well enough.

Quote:

---

Originally Posted by edmayo

Sorry Arne, I knew that,,,,just didn't proof read my post well enough.

---

No harm, Ed, just lost a little time. And I've got plenty of that.

Printed this on a card to keep with the tools:

Attachment 487096

I'm not trying to encourage more rabbit holes, but sometimes the thermostat has to shimmed differently. The thermostat consists of 25(?) wafer pairs and then various shims to adjust the cold running characteristics. It could be possible you have to go into that, but for now I'd try to just work on the warm running adjustments. By the way the thermostat housing has to get very warm,,,,I assume yours does?

Quote:

---

Originally Posted by Arne

No harm, Ed, just lost a little time. And I've got plenty of that.

Printed this on a card to keep with the tools:

Attachment 487096

---

Unless it's a 69 one year only pump :)

Your throttle bodies are worn. This summer with fully warmed car your idle will run closer to 2k

Quote:

---

Originally Posted by beh911

Unless it's a 69 one year only pump :)

---

It's the same for all the MFI pumps.

Thermostat: On the end of the thermostat rod, after the 25 pair of thermo disc's, are various sized shims (from .10mm to 2.0mm) that make the adjustment for how rich the fuel will be at cold starts. One .30mm shim can make a 10 to 15% difference in fuel delivery. When below 32 degrees this will move the main rack for 35% to 45% more fuel for the cold start. Adding shims will lean out the cold start fuel, removing shims will richen up the cold start fuel. Once the thermostats 25 disc's warm up and expand to the max, about 105 to 110 degrees F, it has reached the warmed up condition and stays there till you shut off the engine and it cools way down.

Quote:

---

Originally Posted by beh911

Your throttle bodies are worn. This summer with fully warmed car your idle will run closer to 2k

---

Thanks for sharing your thoughts on this. As noted in post #10 of this thread, that was the first thing I suspected and checked. There are no apparent vacuum leaks there.

Quote:

---

Originally Posted by edmayo

I'm not trying to encourage more rabbit holes, but sometimes the thermostat has to shimmed differently. The thermostat consists of 25(?) wafer pairs and then various shims to adjust the cold running characteristics. It could be possible you have to go into that, but for now I'd try to just work on the warm running adjustments. By the way the thermostat housing has to get very warm,,,,I assume yours does?

---

Yes, the thermostat does get hot. It is plumbed to the HE, although it is currently connected with a big rubber hose rather than the double paper hose. But the evidence is that the thermostat is working fine. While the enrichment is present, the part load behavior is very good. Once the thermostat gets warm and enrichment ends is when it starts running badly. So the thermostat function is not in question.

Quote:

---

Originally Posted by 2.5MFI

Thermostat: On the end of the thermostat rod, after the 25 pair of thermo disc's, are various sized shims (from .10mm to 2.0mm) that make the adjustment for how rich the fuel will be at cold starts. One .30mm shim can make a 10 to 15% difference in fuel delivery. When below 32 degrees this will move the main rack for 35% to 45% more fuel for the cold start. Adding shims will lean out the cold start fuel, removing shims will richen up the cold start fuel. Once the thermostats 25 disc's warm up and expand to the max, about 105 to 110 degrees F, it has reached the warmed up condition and stays there till you shut off the engine and it cools way down.

---

Thanks for this explanation, Mark. So there is really no way to know just how much the thermostat is moving the main rack without disassembly and analysis. I don't think I'll go there now.

The key still seems to be that part load behavior is very good while the 'stat is doing its thing. So that still points to the main rack being lean when at normal temp.

As noted above, I'm going to recheck base settings of the throttle plates today, before making any other changes. I've got this feeling that I may have them all too tightly closed, and that may be the cause of the very low idle, as well as the perceived richness at idle.

Quote:

---

Originally Posted by 2.5MFI

It's the same for all the MFI pumps.

---

Indeed, my bad. It was the number of adjustment clicks that is different in 69. Thanks

Arne, the air-bypass screws should be able to adjust to the idle speed you want. It is possible that their passages are coked up, but in addition to removing the adjustment screw and cleaning that passage way, there is also a counterbore that comes in from above the throttle plates that can only be accessed after removing the throttle stacks. Another diagnostic tool you might try is a infra-red heat gun. While the engine is running aim the gun at each exhaust pipe as it exits the head and compare exhaust temps. They will be in the 200-275 range. What you are looking for is a cylinder, or cylinders, noticeably different, like maybe 125 when all the rest are higher. That would be a cylinder not working well, maybe a clogged injector. If you find such you can swap it to another cylinder and see if the problem goes with the injector. Given the history of these parts that is a real possibility.

Good tips. I'm going to look into cleaning the bypass passages, as the air screws don't seem to change things much.

Also noted this warning in the C-M-A document. Giving this a bit of thought, as while I did add about 6 gallons of 87 the other day, I don't know what fuel the seller put in prior to shipping the car to me. I wonder if this is really a significant possibility, especially in combination with the slightly colder plugs. I think I'll pick up a set of BPR6ES-6 later today.

Attachment 487241

Somewhere I read- "I lean left". Being a leftie myself that works fore me.

Remember the fantastic NGK spark plugs heat range numbers are reversed compared to what many are used to.

When tuning ours we also used one of the below which helped map the whole mess. Really need to weld in a bung for the sensor though as the tailpipe readers don't work well.

https://www.innovatemotorsports.com/products/lm2.php

Not a lot of progress today. I installed a fresh set of plugs (just because), and used some B12 injection cleaner to make sure the bypass passages were clear. Verified the passages by shining a light down the air screw holes and looking for the spot of light inside when I open the butterflies. All 6 passages are clear.

Still don't seem to see a lot of difference in idle speed when adjusting the air screws, though. Perhaps I'm simply not opening them far enough, but the idle speed is still barely above stalling. A bit of black smokiness to the exhaust, too. So still rich at idle.

Warmed it up, then went to richen the main rack again. Promptly dropped the Allen head plug, and it -- with unerring accuracy -- fell straight through the hole in the shroud used for idle adjustments. Tried using my inspection scope and magnet on a stick to get it back out, ended up having to remove the left heater duct so I could get in there to retrieve it.

So with the time I had today, I only had time for one adjustment to the main rack. Might be a little better once off the t'stat. But still not even close to right. Hope to have time to run through a couple of tweaks and checks tomorrow.

Quote:

---

Originally Posted by Scott72

When tuning ours we also used one of the below which helped map the whole mess. Really need to weld in a bung for the sensor though as the tailpipe readers don't work well.

https://www.innovatemotorsports.com/products/lm2.php

---

If I thought I was going to be doing this very often, I'd get one, Scott. But my plan all along has been to get it close by trial and error, and then take it to my local shop for fine tuning with AFR. Then once it is set, NEVER MESS WITH IT AGAIN!!

I will be replacing the heat exchangers at some point, I will probably have a bung welded in the new ones at that time, just for future ease of use.

Quote:

---

Originally Posted by 2.5MFI

Ed, Arne, The idle mixture adjustment your talking about will go "Richer Clockwise". Leaner is CCW. Now the main rack adjustment is the opposite, CCW for richer and CW for leaner.

---

Mark, you are a lot closer to Eugene than Ed is. Do you do house calls? ;):D

Sure would be good to have someone who is familiar with this stuff peek at it, if only to tell me I'm on the right path (or not).

Arne, If I wasn't so friggen busy and behind schedule, I'd be glad to take a fun drive over the Cascades down to Eugene to help figure out the problem.

Quote:

---

Originally Posted by 2.5MFI

Arne, If I wasn't so friggen busy and behind schedule, I'd be glad to take a fun drive over the Cascades down to Eugene to help figure out the problem.

---

While it would be great if you did, I wasn't serious about that. I hope to get it handled eventually. I'll have to drive it over the hill for a visit someday, when it's better.

Should have some time to work on it this afternoon. Y'all wish me luck.

Further MFI adjustments on hold as of now. I made another adjustment to the main rack (richer) and then leaned the idle a bit as well, then took it for a drive. And things felt pretty good, for a while.

Then it suddenly went bad. Popping, cutting out, backfires, no power, and even died once. Struggled a bit but got it back home.

The thing is, it came on so sudden and so hard, I don't think it's the injection. I suspect ignition.

This was not like with what I've felt (or thought I've felt) with it coming off the thermostat, this was like "I'm going to have to call AAA to have it towed home" sudden failure.

Which doesn't mean I haven't been experiencing this all along, to some degree.

So now to look into the ignition. Still seems to be temperature related, so the likeliest culprits would seem to be the coil and CDI boxes.

Swapping a different coil won't be hard, but how can you test or bypass CDI? I tried a search here, and the forum seems to ignore 'cdi' as a keyword. Any suggestions on testing or bypassing the CDI?

My car showed similar problems after the CDI box heated up- it was so bad that a sheriff stopped where I was pulled over, saying he had reports of gunfire... Changed the CDI and the problem was solved. Try swapping in a known good CDI. If that's the problem, there are shops that rebuild them with new electronics but maintain the original appearance.

I tend to agree that you getting ignition failure, are you using the correct Bosch black short coil for C.D. ignition? no newer silver Bosch coils or generic Pep Boys specials. As to the C.D. box,,,we cheat, we plug in a known good one, less time pontificating means less charge to customer. There are however two fuel system faults that can come on quickly. One is the return circuit check valve in the outlet of the fuel filter housing (hopefully that wasn't lost when all that stuff was removed and replaced) or a partial clogged fuel tank strainer (filter).

Definitely seems temperature related. I let the car cool down a couple hours, went out and started it and it ran reasonably well again. Didn't take it for a drive that time, I'll give that a try probably tomorrow. If it drives OK until something (CDI box?) gets warm, then I'll go the next step. I have a friend who has a '73.5 T, I'm assuming that car's CDI is the same 3-pin as mine. She will gladly let me borrow the box, I'm sure.

Yes, Ed, the coil is a Bosch black with brown top. Still has the German 'high voltage' sticker on it. Which means it is probably also 48 years old, and therefore suspect as well.

If I can confirm an ignition problem, no further tuning on the MFI will happen until I get the ignition issue addressed.

When the CDI is good it should hum/buzz with the ignition on, but the engine not running. So the first time you go out when the engine is cold check for the buzz and then once it warms up and runs poorly check for the buzz.

Had the same issue with our 69. Was trying to tune the injection and the CDI box started to go. Definitely put a strain on the backyard mechanics skills as it caused a few days of head scratching.

With this newly described behavior, Ed's suggestion of checking for fuel flow issues is a simple task.

Remove the fuel level sender and shine a light down on the strainer sock area around the outlet bung. If it is all silted up, or looks like it could get that way after some sloshing around and pump running, you should consider draining the tank and cleaning it.

Definitely not CDI box or coil. No change when those parts were swapped. Putting both cars back to original, then looking at fuel delivery.

Will check inside tank, also wondering about the fuel pump. Do they tend to have problems with keeping up when they get older?

http://www.early911sregistry.org/for...4&d=1554485763

Tank has virtually no sludge or grit or silt in it at all, and the strainer sock looks new. Pretty confident there's nothing blocking the tank outlet.

Would the return check valve start misbehaving only when things get good and warmed up? That doesn't seem likely, to me. The issue still seems temperature related. It doesn't start cutting out right away, has to get warmed up. If I start with it cold, 2-3 miles driving before things start cutting out and popping.

Wondering about the fuel pump itself. I've had other vehicles where failing pumps were first noticed when they get warm.

Inside the tank:

Attachment 487572

Easy enough to look at the check valve, unscrew the banjo from the filter console (has the return arrow stamped on housing) and look in the of the banjo and you will see a spring loaded, phenolic washer. Should easily spring back when pushed in. To check the pump pressure remove the fuel feed line at the fuel filter housing (has cast in arrow pointing in towards filter) hook up a (min.) 60 psi pressure gauge and turn on pump. Should make about 30 psi dead head pressure. To test for check valve remove inlet line into fuel filter housing (the fitting aimed towards rear of car) and install your pressure gauge using a tee fitting so you can run car with gauge installed. With car running you should be seeing 10 to 12 psi. Try to read gauge when car acts up. An AFR gauge would really help here. Need to know mixture strength when hot and acting up. A volume test of the pump can be done from that fuel feed line you tested pump pressure at, aim hose into a small coffee can, should easily fill that up in less that 60 seconds.

So today I traced and checked the fuel delivery path from the tank to the MFI pump. As noted with the picture above, the tank is very clean inside with a new strainer screen in place. I found that all fuel hoses for the entire path (both supply and return) are new, done when the engine was out for rebuild. All clamps are proper "not-worm-gear" type, and are tight - so the pump is not sucking air from a loose connection anywhere.

Quote:

---

Originally Posted by edmayo

One is the return circuit check valve in the outlet of the fuel filter housing (hopefully that wasn't lost when all that stuff was removed and replaced)...

---

Quote:

---

Originally Posted by edmayo

Easy enough to look at the check valve, unscrew the banjo from the filter console (has the return arrow stamped on housing) and look in the of the banjo and you will see a spring loaded, phenolic washer. Should easily spring back when pushed in.

---

I ended up pulling both banjo fittings and the plain barb fitting for the return-to-tank line from the top of the filter console. First the banjo for the return from the MFI pump. No check valve present. So I pulled the other banjo (inlet from fuel pump) and the plain barb fitting (return to tank) just in case it got into the wrong port on reassembly. Nada. So your fear about it being lost during the rebuild, rust repair and engine compartment painting was spot on. I'm assuming that this could conceivably be the root cause of my fuel supply problem.

I'm not seeing any indication that the check valve is available as a separate part anywhere. Am I going to need to source a new filter console to get one?

Late addition. Is the check valve supposed to be inside the plain barb for the final return? Part number 901-110-915-01? Not inside the banjo? Gonna have to go give that another look.

Even later addition - The check valve is in fact inside the barb fitting, not either of the banjos. Mine is present, but is pretty much toast. It was being held part way open by chunks of the phenolic material, and sticks wide open when pushed. Doesn't go back to the "closed" position by itself, either. So I need to replace it. No purpose in going any farther until that is handled.

We last bought those a couple years ago, readily available then, expensive, but available. I always carry a spare in my travel kit. Sorry, didn't mean to mislead by saying 'banjo' I knew what I meant !!

Still expensive, but still available. Should be here later this week. No further work on this part of the car until then.

I am currently experiencing similar problems with my 1972T.

I pulled the car out of hibernation in 2017. (been parked since 1986)
I had all the lines/filters replaced, tank drained etc. lots of oil in the exhaust. Car didnt want to run.

Then I had the plugs and points replaced and a new fuel pump.
The car would run but once warm it would stall and wouldnt start again.

I sent the MFI pump to MFI Werks and the car worked better but still have the problem of it sputtering once it becomes warm.
It will run when it is cold, but soon as it gets up to temp it will not run. It sputters and stalls. i was lucky to make it home.

Would it be a matter of just adusting the MFI? Or is there something else going wrong here?

Quote:

---

Originally Posted by tcsracing1

I am currently experiencing similar problems with my 1972T.

I pulled the car out of hibernation in 2017. (been parked since 1986)
I had all the lines/filters replaced, tank drained etc. lots of oil in the exhaust. Car didnt want to run.

Then I had the plugs and points replaced and a new fuel pump.
The car would run but once warm it would stall and wouldnt start again.

I sent the MFI pump to MFI Werks and the car worked better but still have the problem of it sputtering once it becomes warm.
It will run when it is cold, but soon as it gets up to temp it will not run. It sputters and stalls. i was lucky to make it home.

Would it be a matter of just adusting the MFI? Or is there something else going wrong here?

---

Sounds like the exact same symptoms my car has. Runs OK when cold, and very badly - bucking, sputtering and dying - when warm.

My first advice (based on recent experience) is "do not touch any MFI adjustments". Chances are there is something more basic at fault.

I don't know yet if the bad fuel return check valve is the cause of my problem, but I have high hopes. I've heard from others now who had failed check valves, they experienced similar symptoms. Mine will be here later this week.

In the meanwhile, it might be worth taking a look at the one on your '72. See if it is sticky, or stuck partially open like mine was.

Attachment 487882

Quote:

---

Originally Posted by Arne

Sounds like the exact same symptoms my car has. Runs OK when cold, and very badly - bucking, sputtering and dying - when warm.

My first advice (based on recent experience) is "do not touch any MFI adjustments". Chances are there is something more basic at fault.

I don't know yet if the bad fuel return check valve is the cause of my problem, but I have high hopes. I've heard from others now who had failed check valves, they experienced similar symptoms. Mine will be here later this week.

In the meanwhile, it might be worth taking a look at the one on your '72. See if is is sticky, or stuck partially open like mine was.

Attachment 487882

---

thanks, i will take a look!
Would be nice to finally drive the car!

Check valve arrived today. Now installed, I think I may be finally making some headway.

Except...

I now (or probably still) have another fuel delivery issue, and it's a bit odd. Was apparently doing this prior to the check valve replacement, but the general lack of reliable fuel pressure made it hard to notice.

Here's the deal, it runs pretty well (given that the mixture settings are all screwed up) when on the flat or going down hill. But go up much of a grade and it starts to run out of fuel. The steeper the grade, the faster the starvation happens.

I can't think of anything or anyplace in the fuel system that could do this, other than in the tank. Yet the tank was really clean inside when I popped out the sender to peek about.

The tank does have a big dent in the bottom from some bozo and probably a floor jack sometime in the distant past. Not dented near the outlet.

Am I missing something here? Is there anyplace other than the tank I should be looking?

If not, I'll pull the sender again, put the front end in the air and look inside. Not sure what I'll find, but I can't think of where else to look.

Getting more confused as I go.

Putting the front wheels up on a pair of 8" (20cm) ramps is more than enough to trigger the problem. With the front that much higher than the rear, the car will not start. Put it back on the flat, it struggles a bit (as though it has to prime the fuel system) and then fires. Fuel pump sounds different briefly before it starts.

While up in the air and not able to start, looking in the tank shows nothing at all different than before. Tank looks really clear and nice. Nothing blocking the tank strainer.

Are there any check valves in the factory fuel pump?

Just simply remove the feed line from the discharge of the pump which goes up thru the frame section right where it enters the inlet fitting of the fuel filter console. Install a pressure gauge there and see if you're getting 30 psi or so of pressure, then run that hose into a small coffee size can and the pump should fill that in less than a minute. If it does all that then jack up front of car and repeat. If it fails that then fill a five gallon can with gas and with several feet of extra fuel hose run the car at an angle from the five gallon can. This will prove whether it is a pump problem or a fuel feed problem involving the car's gas tank and lines. And by the way make sure that the fuel hose from the bottom of the filter is going to the passenger side of the MFI pump, and the driver side of the MFI pump is going to the inlet fitting of the console. ( the fitting facing rear of car)

Hi Arne

Been watching this thread from the start and perhaps a couple of things. I don't see anywhere where it says that you have actually measured the fuel pressure with a gauge, both at idle and at a higher rpm. Also, I have seen pumps over the years put out enough pressure but not enough volume, again, both at idle and higher rpm. Ed referred to volume in an earlier post but saw nothing else about that. Going any further without really knowing and confirming both would drive me crazy. It is about basics, it sound like you are just not getting enough fuel.

Thanks, Rick

Quote:

---

Originally Posted by edmayo

Just simply remove the feed line from the discharge of the pump which goes up thru the frame section right where it enters the inlet fitting of the fuel filter console. Install a pressure gauge there and see if you're getting 30 psi or so of pressure, then run that hose into a small coffee size can and the pump should fill that in less than a minute. If it does all that then jack up front of car and repeat. If it fails that then fill a five gallon can with gas and with several feet of extra fuel hose run the car at an angle from the five gallon can. This will prove whether it is a pump problem or a fuel feed problem involving the car's gas tank and lines. And by the way make sure that the fuel hose from the bottom of the filter is going to the passenger side of the MFI pump, and the driver side of the MFI pump is going to the inlet fitting of the console. ( the fitting facing rear of car)

---

Well, I'm back, and no less frustrated than before.

I started going through some of the tests here with a borrowed pressure gauge. Pressure at the inlet of the filter console would jump to ~40 PSI when the pump was switched on, then the pump's relief valve would open and it dropped to a steady 28 PSI and stayed there. Did not matter the attitude of the car, same result with the nose in the air.

Verified that the lines to and from the MFI pump are correct, bottom of filter to the right side of the pump via large banjo fitting.

Started to look into volume testing by running the pump into a bucket, but noticed that the fuel pump was starting to leak again. (I had replaced the large o-ring that seals the motor case to the pump body before I got into this process.)

As a stopgap to deal with the leak, I bought a 2-port pump conversion kit from Zims to get the leaking pump back off the car. (Will decide later if I want to try to rebuild my original or buy one of the new Chinese pumps from Porsche Classic.) This new pump eliminates the bypass valve and return hose from the pump, and uses the check valve in the filter console to regulate pressure. (My check valve is new and factory Porsche part.) So I can no longer check pressure at the inlet side of the filter console. I will need to check it between the bottom of the filter and the MFI pump inlet. But I had to return the borrowed gauge, so I have not tested pressure since the pump replacement.

But I'm going to guess that with this new pump in place I shouldn't have a shortage of pressure. And...

No change. I'm not really certain at this point if I am really losing pressure when I start up a grade, or if it is just that going uphill just puts more load on and it starves out. But the result is that trying to pull up a modest grade at normal RPMs for town driving runs it out of fuel. Sputters, pops, and will even sometimes die.

I feel I must be missing something very basic. I will get the gauge back tomorrow for a bit more testing with this new pump. I will (assuming the pressure tests out OK) go back to the volume test into a can, and probably also try to run the engine out of a gas can to eliminate all existing tank and line possibilities.

But I'm really starting to wonder if there is something else afoot. Could the MFI pump itself have issues that would manifest themselves in this way?

I'm curious to see what you get when running out of a gas can vs. the tank. Even though that in tank filter looked perfect I don't think you ever took it out and inspected it did you?

No Ed, I never took it out and inspected it myself. It looks new when seen through the sender hole, and when looking at the connections coming out the rear of the tank I saw obvious new copper washers. But it could still be a problem.

After playing with gas cans, gallon paint buckets, wheel ramps, and a couple lengths of fuel hose, I've ruled out fuel delivery as my problem. Doesn't matter what or how I get fuel to the pump, there was no change.

In fact, today when it started sputtering going uphill, I found that if I could get the revs up above 4000 it would pull again. Between 2000-3500 when the engine is warm and under load it just flat doesn't want to run. So it must be getting fuel, it just isn't metering it properly at some speeds and loads.

Pulled the plugs for a look today, the side electrodes are now showing gray-tan color, not black and sooty as before.

The baffling thing is that it runs much better when cold, pulling up hills nicely. Then after some driving (5 minutes?) and it gets warm, it changes to really ugly. The change is rather sudden, maybe over 30 seconds or so.

This would seem to be related to the thermostat on the injection pump. My thinking has been the mixture is pretty good when cold, then when the 'stat warms and leans the mixture it gets really bad.

So I took off the 'stat to check it out.

Quote:

---

Originally Posted by 2.5MFI

Thermostat: On the end of the thermostat rod, after the 25 pair of thermo disc's, are various sized shims (from .10mm to 2.0mm) that make the adjustment for how rich the fuel will be at cold starts. One .30mm shim can make a 10 to 15% difference in fuel delivery. When below 32 degrees this will move the main rack for 35% to 45% more fuel for the cold start. Adding shims will lean out the cold start fuel, removing shims will richen up the cold start fuel. Once the thermostats 25 disc's warm up and expand to the max, about 105 to 110 degrees F, it has reached the warmed up condition and stays there till you shut off the engine and it cools way down.

---

Mine does still have all 25 pairs of discs, and a single 2.0mm shim. As best I can determine using my heat gun, this gives me about 3mm of movement at the pushrod.

Since it seems to run better when cold (pushrod 3mm shorter than warm), I tried shimming the thermostat out 3mm in hopes that once warm it would run like it often does cold. That failed miserably. Ran so badly cold that I aborted the test drive and limped home. Oddly, pulling the thermostat out 3mm made it run cold just about the same as how it runs normally when warm. That seems very wrong to me, as either direction the pushrod goes seems to make it run much the same (badly).

Totally frustrated now, and confused. I'm going to take a couple days off and head out of town with my wife. Feel free to offer more suggestions and advice while I'm away.

That is new information, that it will run somewhat better if you can get it above 4,000 RPM. That could mean that the pump main adjustment is at least part of the problem. Since you don't have an AFR meter you'll just have to make adjustments and record what they are. Normally everyone just wants to jump in and screw with the rack adjustments, and now you've seen there are many other checks to make first, such as that bad check valve. By the way, reducing shims richens the thermostat. We're going to do this seat of the pants method. Richen the main rack adjustment 4 clks counterclockwise ( I proof read this time) record the change, and repeat your test. Four clks is a big step. If we're going in the right direction it ought to be an improvement. If so richen again 4 clks. By this time it will be obvious this is either working or it isn't.

Got back home yesterday, this morning I did the 4 clicks richer and took it for a drive. Once warmed up, the results were horrible. Had to struggle to get it back up the hill to the house.

So I back off that change, and then went 4 clicks leaner instead. Still undrivable. Very little difference. An 8 click swing, no real difference. Which is consistent with how this has gone through the whole process. Even significant changes don't seem to make much difference.

At this point I've thrown in the towel. I called Mark at MFI Werks, and will ship him the pump and all 6 injectors for inspection, cleaning and calibration.

It is definitely possible that a proper calibration of the pump may not fix my issue. There may yet be something else at play that I haven't found. But the MFI is so far out whack now that I need to get it to a baseline for further testing.

I (and possibly Mark) will update this thread as news develops. Probably nothing before late next week at best.

Also noticed during initial disassembly that the coldstart squirter (nozzle) for #3 is broken and was bypassed. Placed a Wanted ad for it -

http://www.early911sregistry.org/for...with-T-fitting

http://www.early911sregistry.org/for...5&d=1556332903

I'll be curious to see if a stuck rack was your problem.

Curious myself to see what is really going on. The term "stuck rack" is used a lot, but what is really happening is one or more plungers are getting stuck or frozen in they're cylinders. When a plunger is stuck/frozen it prevents the rack from moving. If a plunger is frozen, that cylinder will not fire and the rack will not respond to any space cam changes. The problem Arne is having does not seem to be a frozen plunger from all the symptoms he has described. But I have been surprised before.......

i am still struggling with my car..... going to bring it back to my Indy and try some things.
You can see why some people swap to carbs!

But im not giving up!

I've seen too many messed up carb tunings to buy that!

Mark called and said my pump was delivered to his shop this afternoon, so perhaps we'll have an idea what's wrong before too long.

On the other hand, I'm still aware that the pump may not be the root of my problem. But I need to eliminate that as a possibility before I go any farther.

The good thing on this is I'm hoping that a freshly calibrated pump will be close enough to correct when I get it back on the car that I can leave it alone while tracking down the real problem (assuming it still exists). Or maybe it will run great and I can move on to other things!

Talked to Mark this morning, my pump and injectors are on their way back home - inspected, serviced, cleaned and calibrated.

As for "THE PROBLEM", when Mark first looked inside he he was fairly sure he found the issue, but after bench testing he's not as certain. Perhaps he'll post here to explain better than I can. But what he first saw was that the shutoff solenoid was adjusted such that he felt it could be accidentally shutting off the fuel at times once the thermostat pushed the rack into the fully warm position. Unfortunately the initial bench test did not duplicate the issue. But it is still possible that when on the car this was the issue, as the pump has been bounced around by UPS on its way to Mark, and that could have changed things a tiny bit.

In any case, that has been corrected, and everything checks out great now, ready to bolt back on the car.

Mark also suggested - once we got to chatting about the shutoff solenoid - that perhaps I could have an electrical fault that is triggering the solenoid when it should not be triggered. I have checked the microswitch, which appears to be working correctly, but that is not all of the circuit. Therefore, when I first test it after reassembly, I will NOT connect the solenoid. If all runs well (other than popping exhaust) with it disconnected, I can try reconnecting it to see if the problem reappears.

I'm actually somewhat optimistic for the first time in quite a while. Because a shutoff solenoid problem of some type is the first theory that truly covers most of the symptoms. Here's hoping...

Got my fingers crossed for you! The solenoid is NOT something that one normally fiddles with ! But, like others, I have seen some strange stuff. Had one a few months ago that would run pukey rich no matter what adjustments were made. Turns out some one had removed half the thermostat shim stack,,,arrgghhhh :confused:

Half the shim stack? Sounds like a movie title - Rich and Richer...

Quote:

---

Originally Posted by Arne

Half the shim stack? Sounds like a movie title - Rich and Richer...

---

Arne, close....

https://youtu.be/hG46_SvgtRI

Sigh... So close.

The pump and injectors were delivered this afternoon, I have got it all reassembled.

Now I have a new issue. It starts right up, but will only stay running while the key is in the Start position. As soon as I release the key and it falls back to Run, the engine dies as if I have shut it off. Normally, I'd assume this was a bad ignition switch, but when tracking this in the wiring diagrams it looks like the coil/CDI don't get power on a separate wire from the switch. And all other electrical bits work fine. Fuel pump runs, headlights come on with the key, turn signals and gauges, wipers, everything.

I'm not seeing separate feeds from the switch to power the coil/CDI from Start and Run, so I'm thinking there must be a relay involved somewhere.

Any likely connections that I could have disturbed in the back while working?

Since this all worked before, and all you did was R & R the pump, the ignition switch must have been on the way out. It is a straight forward circuit for the ignition, power from 15 of the ign. switch to fuse 7 which is bussed to fuse 8, from fuse 8 to term. B of the CD box. Pull the plug from the CD box, put the red lead of a voltmeter to the middle term. of the removed plug, and the black lead to ground. Turn ign. to crank,,,you should see 12 volts, then return ign. switch to 'run' position, you should still see 12 volts. If not the electrical portion of ign. switch has gone south.

When the MFI pump and injectors are removed for service and then reinstalled it takes a hole lot of starter motor cranking to bleed out all the air in the fuel lines before the engine will finally run and idle. Maybe what's happening is the cold start system is firing the engine during cranking like it should (fuel squirting in the intakes) then when you back-off the key to run it dies because fuel lines are still pushing air.

Quote:

---

Originally Posted by 2.5MFI

When the MFI pump and injectors are removed for service and then reinstalled it takes a hole lot of starter motor cranking to bleed out all the air in the fuel lines before the engine will finally run and idle. Maybe what's happening is the cold start system is firing the engine during cranking like it should (fuel squirting in the intakes) then when you back-off the key to run it dies because fuel lines are still pushing air.

---

This is apparently what is happening. If I disconnect the cold start solenoid it doesn't fire at all. Reconnect it and it fires until I release the key, then dies.

I knew I'd need to crank a bit to purge the air out, but man alive, it is taking a really long time. I've cranked it (with cold start disabled) for several minutes now, no real signs of life yet. Taking a break for a bit to make sure I don't overheat the starter motor. Will keep at it.

I would keep the cold start circuit hooked up. It will speed up the RPM's during the cold start and that will bleed out the air faster. If you have a battery charger (50 amps or more) hooked up to the battery that will help speed up RPM's also and keep the battery a little more charged.

After a few more minutes cranking, it finally purged all the air and fired normally.

Took it out to get it fully warm, and THE PROBLEM is still there. Perhaps a bit less severe than before, but still there. When fully warm, pulling under a load between 2000-3500 RPM feels like running out of fuel. But it is apparently getting fuel all the time, as it runs better if I can keep the RPM above 4000. Still runs better cold than warm, although that has changed as well, it "feels" and smells a bit richer than before when starting cold.

I feel I've done all I can with fuel and fuel delivery. The mixture with the calibrated pump is close enough now to leave until later. The plugs after the test drive are no longer black and sooty, but instead have a nice gray-tan color.

So I do not plan to touch the MFI settings until I can track this down, and then will fine tune it properly using an AFR meter.

In the meanwhile, I'm going to go back and recheck some of the other basics, most of which I've checked at least once, some several times. Going to take a long look at plug wires and other ignition bits. Also plan to re-check valve lash.

Other thoughts and advice are welcome.

Check cam timing,,,,been fooled by that once. Car had come from dealer after new tensioners installed, (this was around 1975 or so ) On the theory of go back to the last thing done I checked cam timing,,,,sure enough it was wrong. Nothing you can do will make it run right ! This has become part of my to do list when the problem won't act normally. Only takes a half hour to check, and can save hours of grief.

I’m sorry for your frustration but this thread has been interesting reading.

Cam timing seems unlikely, Ed, since it runs so well cold. But I'll take a peek at it just to be sure. Any gotchas I need to be aware of with the Carrera pressurized tensioners that are on it?

Just thinking of things tonight. I once had a problem on a 240Z with spark breaking up at higher RPM - felt a lot like fuel starvation, but wasn't. Turned out it was an ignition condenser that was way bad. I don't recall seeing any condensers on this car, did the CDI eqipped cars not use condensers?

No condensers used on CD ignition. Seems to me in previous postings you had done some swapping with known good pieces from another car. The Carrera tensioners won't be problem in checking cam timing. I was going to ask if it had solid tensioners,,,I've seen those get so loose the chains whipped around radically changing the timing with engine speed. Measure the rotor resistance HOT just for grins.

Yes, swapped a known good CDI box and brown-top coil in, made no difference. So I'm comfortable that those two items are OK.

Cap and rev-limiting rotor are both Bosch, look to be fairly recent, presumably from when the engine was rebuilt. Points too. But plan to revisit all of that soon, along with valve lash and now cam timing.

Plug wires are suspect. Now that I have taken a fresh, longer look at them, I suspect they are rather old. The ends are Bremi brand, don't know whether that's a good or bad thing. Anyone have any favorite brand of wires they like for a driver? Not worried about original brand or look, just function.

Just measure total resistance of plug wire from end to end, about 3,000 ohms tops, that will make sure there are no open circuit ones. For example, on the rotor, you can't say "they look to be recent" you will have to verify one way or another.This will be a matter of slowly eliminating possibilities until you find the problem,,,,and when you do find it I can almost guarantee it will be simple. Those are the ones that drive us nuts!

Quote:

---

Originally Posted by edmayo

Just measure total resistance of plug wire from end to end, about 3,000 ohms tops, that will make sure there are no open circuit ones. For example, on the rotor, you can't say "they look to be recent" you will have to verify one way or another.This will be a matter of slowly eliminating possibilities until you find the problem,,,,and when you do find it I can almost guarantee it will be simple. Those are the ones that drive us nuts!

---

Its amazing how that works out on many things...sometimes its something simple. I hope so for Arne

Going through the basic non-MFI checks as time permits. Unfortunately, my family believes that I need to do other things at times, rather than spending all my time working on the Porsche.

Ohm'd all the plug leads, as well as the coil lead. Each plug wire was between 960-1020 Ω. The coil wire was just over 2000 Ω. I can't recall whether resistance is cumulative or if only the highest value rules. If you add them up, the effective resistance is ~3000 Ω. If you don't add them, they are only ~2000 Ω.

Also checking valve adjustment. I did 4, 5 & 6 today (I do them one side at a time, more turning of the crank but less mess and back and forth). 5 of the 6 were just the tiniest touch loose. Nothing I'd normally get worried about.

Going to be on the road for a couple days, so I won't get to checking 1, 2 & 3 until probably Sunday.

Check the cam timing while you're in there. You're fine on resistance. Be sure to check the rotor resistance. Since the problem occurs with temperature heat the rotor a little and check it. To duplicate the conditions you describe.

What's the simplest way to verify cam timing? Dial indicator on a valve to check where it opens or closes?

Arne sorry you're having so much trouble getting it to run right. Keep digging, you'll find the problem soon.

911 cam timing is done by measuring the intake valve lift at T.D.C. You use #1 intake valve for the 1-2-3 side and #4 intake for the 4-5-6 side. You'll need a 'Z' block and metric dial indicator. These are available from many sources, Stomski Tools is a good one. When I suggested that you do this since it only take a few minutes I guess I assumed you had the tools and had done this before. But none the less this is one spec that must be tested and either corrected or checked off as okay.

Metric dial indicator I can borrow locally. Will look into the 'Z' block, but probably won't get to this for a while. But really, pulling the upper valve covers is not hard.

I’ve got a pair of blocks and digital gauges. It was super convenient to use both. I can bring them Saturday if you need them.

John, I'd love to borrow them, but won't be in town Saturday, leaving tomorrow for a couple days. I'll text you later to see if we can match schedules.

Found a possible oddity in the ignition (CDI) wiring. Separate thread here - CDI wiring question

OK, here's what has transpired today.

Rotor ohms out at 4.85k Ω. I think it should be ~5k, so I'm going to call it OK. Checked, cleaned and dressed the points, they looked great. Re-installed and gapped them.

Finished checking valve lash. Found several that were the slightest amount loose, none that were too tight. Corrected them all, but since there were none that were way out of spec, I did not expect any change there.

Have not verified cam timing yet, as I haven't been able to meet up with @jt912 to borrows his blocks and dial indicator. Will do that before long, but as with the valve lash, I don't expect that to be the cause of this issue.

Corrected the CDI ground connections as noted in this other thread - CDI wiring question. As with the above, I did not expect any change from this.

Took the car out for a drive, and as I expected there was no improvement. Once it got good and warm, it began to stumble, sputter, and pop. Occasionally it would simply die, and generally was hard to restart.

When it happens, it still feels like it is running out of fuel, and the tach does not bounce or otherwise indicate that it has lost signal. However...

On the way back home, during two separate instances when it died and refused to restart, I checked for fuel supply by pulling the hose from the bottom of the fuel filter. Plenty of fuel pressure there, but the engine just cranked away like crazy with nary a cough.

That would tend to confirm that fuel delivery is not the problem. From that point to the MFI pump is just that 50cm length of fuel hose, which in my case is new, and not kinked.

Since the tach continues to show RPM while the engine is not firing and I'm under compression, it would seem to me that the points are still breaking (signal sent to the tach), but for some reason the coil is not firing.

Remember that earlier I had swapped in a known good CDI box and coil from a '73.5 T (CIS) that runs perfectly, with no change, my car still experienced the issue once it got warm with the known good parts installed.

The problem first shows up as a bad stutter, and rapidly progresses to complete no fire. Because of this, I don't think the problem is the plug wires, as having all 6 short out at once seems unlikely. The coil wire is still suspect, even though its resistance is OK.

I'm also wondering about the CDI wiring harness. I noticed that the harness itself was noticably quite warm to the touch, even near the distributor. The CDI box generates a fair amount of heat as well. What if the CDI harness shorts or breaks the circuit when warm?

I'm thinking of bypassing the CDI system (and it's wiring), doing a temporary re-wire to a pure points/coil ignition. Will the Bosch brown-top coil tolerate normal non-CDI ignition?

To resort to a conventional ignition you will also have to install a condenser in the circuit, and no, that coil's windings are not the correct ratio for a non-CD system. I still want to check stuff like rotor resistance when the problem occurs, when the rotor is hot. I would install a temporary tee in the return side from the filter, where you measured the check valve pressure of 12 or so psi. Run a fuel line into the car so you can read the fuel pressure WHEN the problem starts. I'd also run a timing light into the car so you can see if it is ignition or fuel that quits with time and temperature. As I recall you had once run it from a temporary fuel can source so you have ruled out fuel tank and filter/line problems at least back to the fuel pump. Correct? Have you also checked for ground circuit from CD box to the ground stud on the fender behind the fuel filter? I don't think I've really checked the wiring to the coil for temperature so can't say whats normal, but heat tells me there is a ground issue some place. Did you once say you have used ohm meter to check resistance of the transmission ground strap to the body stud?

Yeah, I figured I couldn't use the current coil if I did that. Oh well...

Will re-measure the resistance of the rotor after some judicious use of my heat gun tomorrow. Pretty certain I've checked the ground behind the CDI box, but will confirm, and check the ground strap on the transmission as well.

Not sure if my timing light leads are long enough to reach inside the car while driving, but if I remove the engine grill and go through the rear quarter window it might reach. Ditto for the fuel line for pressure test while it fails.

You are recommending I check the pressure with a tee in the return line that goes from the injection pump to the filter console? Ahead of the return check valve in the fuel return path?

good to see you have not given up!
i have same issue and been putting my car off..... but i will be back to it soon.

Quote:

---

Originally Posted by Arne

Yeah, I figured I couldn't use the current coil if I did that. Oh well...

Will re-measure the resistance of the rotor after some judicious use of my heat gun tomorrow. Pretty certain I've checked the ground behind the CDI box, but will confirm, and check the ground strap on the transmission as well.

Not sure if my timing light leads are long enough to reach inside the car while driving, but if I remove the engine grill and go through the rear quarter window it might reach. Ditto for the fuel line for pressure test while it fails.

You are recommending I check the pressure with a tee in the return line that goes from the injection pump to the filter console? Ahead of the return check valve in the fuel return path?

---

Yes, we need to know fuel pressure when it quits on you, same for timing light, should reach thru quarter glass to back seat, may need helper to hold light trigger in back seat while you drive. We need to know what is failing, If there was access to a chassis dyno that would be even better!

Timing light shows no loss of spark when this happens, not even when it dies completely. Moving back to fuel, which was where I started.

I don't yet have fuel pressure gauges back in my possession, but again, today when it died completely and wouldn't restart, there was lots of pressure at the outlet on the bottom of the fuel filter can. I will arange to borrow the gauge again, but I really do not think this is a delivery issue.

Back to wondering about the fuel shutoff solenoid. While the symptoms are unchanged whether the wire is connected or not, I still wonder if it is somehow triggering spontaneously and shutting down the fuel when something gets warm. Not sure how to test that, though. I suspect I can't simply remove it from the outside and drive it.

Update - the car is now parked about 5-6 blocks downhill from home. Will not start. Cranks fast (dual Miata batteries), timing light still shows spark, but no fire. There is plenty (too much?) fuel pressure at the inlet port of the injection pump on the right side (looking forward). But no fire, and no smell of gas at the exhaust.

While there is lots of pressure at the MFI pump inlet (fuel sprays everywhere when the hose is pulled, even when the pump is not running), there is nothing at the return line from the left side of the pump to the filter console. Seems wrong to me, if the fuel pressure exceeds what is needed by the pump, shouldn't it flow through back to the filter console and then back to the tank?

Going to let it cool off for a couple hours and hope that it will then start up (as it always has when cold).

At this point something has been overlooked. I would drain tank and remove the filter sock and fuel return fitting. You should not have pressure when removing those lines. I would visually verify clean sock by holding up to light, visually seeing the tube inside. I would verify return with compressed air from engine bay inside of tank. I want absolutely verify open circuits from start to finish. I'd also want verify clean ground studs. (You said engine bay was painted).It will be something simple.

Got the car started and back home later this afternoon. While frustrating, the experience was still valuable as I was able to verify that I have spark, and I can ignore the ignition from here out.

Called Mark (MFI Werks) to discuss what I'm seeing, and he - of course - agrees with Ed that the supply line to the MFI pump should not be under pressure like that. He also explained how the fuel supply and return at the MFI pump and filter console is supposed to work, which helps me to better understand what I should be looking for. Will post again when I know more.

Arne, You most likely have double checked this but, fuel filter console fuel lines corrected in early service manual. Should be like this....
Attachment 491023

Quote:

---

Originally Posted by 2.5MFI

Arne, You most likely have double checked this but, fuel filter console fuel lines corrected in early service manual. Should be like this....

---

Oh. My. Word.

No Mark, my manuals all showed the inlets from the two pumps reversed. Oh, crud. Really?

I was checking the output from the fuel pump at the console when I saw this. Was pondering the difference in volume from the line from the fuel pump compared to the output from the bottom of the filter. Lots of fuel from the pump, not much more than a dribble from the bottom nipple on the filter canister. Didn't make sense, since I knew the filter cartridge was new.

Then I saw the post above.

I had "verified" early in the process that the various hoses were connected "correctly", but I was apparently basing that on a faulty diagram. Those two hoses have been reversed the whole time I've owned the car.

I corrected them according to the picture above, then re-checked the output at the bottom of the canister, which now flowed fuel beautifully.

Hooked everything back up and took it for a drive. Wonderful! For the first time, the car will pull nicely up a grade, in any gear.

Not certain why it ran OK when cold when the fuel flow was that low. Or why the flow test results were sometimes inconsistent. But at this point, it runs good cold or warm. I could stop here and I'd be thrilled.

But I will get the AFR and do some final tuning. The way it is running now, I don't expect to have to go more than one, maybe two clicks at most.

Thanks to all here, especially to Ed and Mark for their patience and help!

Fantastic !

Once again, Ed was right...it’s was something simple.

Kudos to Ed and Mark for their expertise in resolving this.

I’m very happy for Arne, but in a way I’m sad to see this entertaining thread come to an end.

Alan
N.J.

Just think, Mark and I and all the other professional mechanics get to do this stuff every day :eek: Ya just gotta step back, stop over thinking, and get back to basics. Glad you got it Arne, that's the best part of this job, when it all comes together!

Great to hear!

An applause for Ed & Mark

This thread really showed some of this forums core strengths!

i will check my car now! Hopefully this is all i need!

Superb help from the Old Sweats here. Reminds me of the sub heading of Agatha Christie’s book with Hercule Poirot “ The ABC Murders. if after you’ve eliminated everything that’s impossible , then whatever’s left , no matter how impossible must be the possible.”

Quote:

---

Originally Posted by tcsracing1

i will check my car now! Hopefully this is all i need!

---

Hope this is your issue, as it is a really easy fix!

Just got back from a half-hour drive over the back roads and hills south of here. Car runs great. Will do the final tuning with AFR and call it done.

Congrats Arne!

A final note. Mark's bench settings for the MFI pump were extremely close. Both part load and idle were borderline lean, so I went 1 click richer on the main rack (part load), and that took care of both. Not planning to mess with this stuff again.

Attachment 491147

WAIT! Your drove in the rain?

Quote:

---

Originally Posted by jt912

WAIT! Your drove in the rain?

---

Indeed. Like you, I live in Western Oregon. The car won't melt immediately, and it gets a nice warm, dry garage to air dry in.

What a great thread. Thank you! Very impressed by your persistence Arne and also by everyone’s contributions and expertise. Amazing.

A major disturbance, but one hek of a nice casting..

Attachment 491150

Wrong diagram in question below.
Glad you got it sorted out.

Attachment 491221

And what is wrong about that diagram?

Quote:

---

Originally Posted by edmayo

And what is wrong about that diagram?

---

That the inlets are the opposite of the photo Mark sent.

Not if you follow the cast directional arrows on the housing.

Hi Ed,
The illustration shows inlet from electrical pump at 9 o'clock
the return line to tank at 6 o'clock
the inlet from the MFI pump at 3 o'clock

Mark shows inlet from electric pump at 3 o'clock
inlet from MFI pump at 9 o'clock

Which is correct?

Quote:

---

Originally Posted by edmayo

Not if you follow the cast directional arrows on the housing.

---

The thing is, both the inlet from the fuel pump and the return from the injection pump have the exact same inbound arrows cast. But they are NOT interchangeable! When the console is mounted in the car, the one facing toward the front of the car is the fuel pump inlet, and the one facing the rear is the return from the MFI pump.

Exactly,,,,"when mounted in the car" which is of course how you would do it.

Quote:

---

Originally Posted by edmayo

Exactly,,,,"when mounted in the car" which is of course how you would do it.

---

For clarity:

Attachment 491248

Compare to:

Attachment 491255

looks like this is not my issue.... back the garage for me!

Quote:

---

Originally Posted by tcsracing1

looks like this is not my issue.... back the garage for me!

---

Sorry to hear that. Put 150+ miles on mine today on a PCA drive.

Attachment 494240