PMO carb issue: 3.2 911 engine

[911hillclimber]

Hello from England!

I hillclimb (autocross) an old Lola sports racer with a 3.2 911 engine running stock except for PMO carbs and custom headers.
I have a strange problem with the tuning and could do with some help please.

The engine was fresh 2 years ago and ran really well when finished and raced for 2 years.
This year it has been going 'off' tune, so i have decided to have another go to refine it all.

Each bank of the engine has it's own header and silencer so you can literally feel what is happening on tick-over on each side of the engine by placing your hand by the exhaust.
Some basics:
All plugs have sparks (engine has a distributor with points)
Carbs are 1000 miles old
All valve clearances just checked and at 0.004 inch
All cylinders have just been leak down tested at 25 psi and all tight.(hot and cold)
Pressure set at 3 psi feed

I have balanced the tick over stops, each side pulling '6' on the meter.(and 1200 rpm)

1/2/3 bank all pulling '6' in balance and mixture set and work! Exhaust is hot and has an even beat to it by feel and ear.

4/5/6 is quite the opposite.

#4 and #5 flow '6' and #6 can only reach '4' even with the needle removed from the carb completely!
I cannot set the mixture on any choke of 4/5/6 bank it is either too rich or non at all.

I get a random back fire too.

The velocity of the exhaust from 4/5/6 is 2 x that of the clean running 1/2/3 and is much cooler.

Any ideas please?

PMO's are rare over here and i cannot find much help.

[Flat6pac]

Be sure you contact PMO, they'll set you up. They're a long ways away but they'll set you up.
Bruce

[1QuickS]

PMOs have a single throttle shaft vs. the two used in Webers. In early triple throat Webers there is the real possibility of the coupling slipping that connects the two shafts which results in an issue like you describe.

Setting your idle mixtures at 1200 RPM will expose the first progression hole which will affect progression mixture so first recommendation is to try to balance air flows at 900 RPM. The air flow balancing should be performed with throttle drop links disconnected to avoid cross-talk between the banks of cylinders.

The idle air correction screws will increase the STE meter reading but if you remove the screw the air through the opening will bypass the air flow meter and show no effect.

If all was good and there is an issue now I'd recommend you have the offending carb off the manifold so you can look to see that the throttle plates are all closing the same.

Tuning procedure for "Lean Best" and for balancing air flow at 3000 RPM are offered for Webers which is same for PMOs, the only real differences between the two are the main jet assembly and fuel floats are from two barrel IDA Webers, accelerator pump and idle/progression are per triple throat Weber. Link to tuning procedure: Performance Oriented

Pretty cool toy!

[gamin]

Bruce is correct. Contact Richard Parr at PMO. He is the guru for those carbs. Great guy. His phone no. is 310-394-0088. He has helped me a lot with my PMO set up.
Also, basic set up, each carb separately, you match the intake air to the high throat by adjusting the idle air bleed screws. Small screws down low with a lock nut. Linkage disconnected of course.

[gamin]

And 1QuickS is the weber guru. And he is right. Very similar.
Friend of mine will be sending his weber's to Mr. Abbott as soon as his name migrates to the top of his long waiting list. There's a reason for that.

[Jack Stands]

I don't know if you are cursed with Ethanol laced fuel in England, but if the car has been sitting for a while, the fuel may have jelled in the carbs and is not metering properly. I would also agree that Richard Paar @PMO or Paul Abbott would be your best sources to solve your issues. Both are magicians with Webers/PMOs.

[cabmandone]

No one has said it so I will. That looks like A LOT of fun.

As for carbs, have you checked each cylinder temp to see if one is colder than the others? My brother's car with PMO's was lacking some power and backfiring occasionally and we found a slightly plugged main jet IIRC. Found it by checking each cylinder exhaust temp with a IR thermometer. Also while performing lean/best that carb wouldn't change at all no matter what we did.

[mreid]

The plastic intake spacer on #6 melted and created a vacuum leak. Shoot some starter fluid around the intake gasket at #6 and see if the idle goes up.

[1QuickS]

I think mreid has it! If you use standard intake manifolds or PMO manifolds with circular ports in the bottom flanges as opposed to openings designed to match the shape of your ports on the 3.2 then you have an unsupported portion of the insulator (plastic) which will create an air leak. It does appear you have PMO manifolds, but are they of the correct interface is the question.

[911hillclimber]

Wow!
Thank you all for all your help, very appreciated as I feel I can't see the 'wood for the trees' at the moment.

Some years ago I had the issue with the insulators and gaskets esp on #6 cylinder. found a hir-line gap about 50mm long by using thCarb cleaner spray on the running engine.
Emailed Richard Parr, and indeed what a great support. Had the right insulators and gaskets off him (arrived in 48 hours!).
That got the engine running and tuned to 247 bhp on the rolling road, 3 years ago.

Only this year has the engine gone to backfiring and at my last event it was not right at all, so decided to investigate this week.

The engine today is like this:

Bank 123 is running smooth, no pops or bangs, exhaust is hot to the touch exiting the silencer on that side, and the velocity of the exhaust is smooth. It feels 'right' if you know what I mean. The carb is disconnected and this assessment is at 1200 rpm.
All chokes are to '6' on the sync meter.

Bank 456 is quite the opposite. Again disconnected, 1200 rpm and balanced to 123 for air speed using a weber sync meter.
#4 and #5 chokes are running at '6' but #6 will only pull 4 at best @1200rpm.

In England we have a small tuning device called 'Colour Tune'. It is a substitute for the spark plug and has a spark element and a glass base. With the engine running you can actually see the spark jumping and the colour (hence mixture ratio) of the combustion.
Yellow is far too rich, petrol blue is just right, black with spark is no combustion.

#4 will only tune to yellow or black with spark, no blue. A few years ago blue was easy to get.
#5 is the same as 4!
#6 is black

With 4 and 5 in the 'yellow' the 456 bank will back fire and you can see this change in the colour tune.
If I blip the 456 carb to say 3000rpm the engine slowing down, carb released to tick over stop there is a short time when petrol blue appears so for a sort time that rich mix thins-out to the right colour. This makes sense I hope.

So;
What to do next?

I will follow your thoughts and try 900 rpm. The engine has never liked such low tick over, it has always been set to 1200, but will try to get it as low to 900 as I can.

Another trick is now to swop the carbs over so the good side (123) is on the 456 bank.
If the carb is at fault then the problem will be on the 123 bank and 456 will be great.

I have had the 456 carb apart yesterday to air-line the #6 choke and prod MIG wire down all the drillings and holes I can find.
After re-assy that changed nothing.

I will email Richard and refer him to this thread for comment.

The PMO's are a great carb, and I know these have run well in the past, but cannot find the sweet spot on the 456 side. Must be something simple?

Again, thankyou for your help so far!

[Lapkritis]

I wouldn't rule out ignition troubles. Check plug wires, plugs, engine grounding/coil/timing/distributor/cap/rotor. A weak spark will be more prone to ignite as cylinder pressure increases and far more prone to misfire at lower cylinder pressures/overly rich mixture. The plug wires appear routed over exhaust header which may lead to faults over time without adequate heat shielding.

Good luck - hope for an easy sorting.

[911hillclimber]

Thank you and I follow your thoughts.

Just swopped the 123 carb to 456 bank and vica versa.

The 123 bank now runs bad, the 456 back sweet as a dream.

Thus, the bloody issue is in the carb.

I'll gut the blighter to the bone and find the gremlin and hopefully get tuning again on sunday.

going to the drag races in Wales tomorrow, my roots are with hot rods!

Now;

Are there any little tricks I should know in cleaning the carbs except to be sparing with the carb cleaner?

[1QuickS]

I have been using Colortunes since 1979 and use one in each cylinder of my test engine for mixture adjustment on Webers, et all. Can't live without them for multi-carb tuning

Check your idle air bleeds are clear of debris. These are screw-in, stainless steel jets located in the top of the main throttle body over each idle jet. If blocked, your mixture will be super rich. Black on the Colortune indicates mixture so rich that there is not enough air for combustion, consistent with your too-rich issue.

Also, PMO mixture screws are of brass and can have their tips broken off if tightened too much. When opened from the too-tight situation the tip shears off. This would obviously create a lean issue but pull your mixture screws and check them. On Webers the mixture screw orifice is 1.00mm diameter. The reason I mention this is if the tips broke off and then "magically" cleared themselves from the orifice you would then have a rich mixture.

The edge of the throttle plates are, by design, to block the first progression hole at idle. This would leave all fuel for idle coming from the mixture screw. If you are setting idle mixtures at an elevated RPM then the first progression hole will begin to be uncovered thusly allowing fuel delivery from that and from the mixture screw. Better to be sure to get first progression hole blocked at idle (lower idle RPM) during mixture adjustment than running at higher RPM. This is not an absolute requirement, just a suggestion.

It is troubling that you cannot get equal flow on all three bores. Like I said, the throttle shaft is continuous on the PMO so it sounds like a mechanical issue of throttle plate alignment. Your insulator plates are in good condition since your issue follows the carb and insulator plates are at the heads.

[911hillclimber]

Thank you Paul.

I stripped the carb this afternoon.

Took every jet out, all 3 blanking pugs and every drilling screw/plug out. Washed with clean petrol and flushed all gallies out with brake cleaner aerosol and then air-line at 80 psi, and then repeated it.

Found nothing.

The carb is almost as good as when Richard's company made it in 2007. The carbs are about 1000 miles old.

all the needles are sharp, every jet hole bar the top accelerator injectors have been repeatedly poked through with 0.8 MIG welding wire and nothing seemed to come out as 'dislodged'.

I could not get the stainless idle jets out of the top however, but have poked through them, brake cleaner etc and air line. The cleaning fluid came out of the top side jet positions and also down below to the chamber where the 4 small holes are close to the throttle plates (butterflies if you are old and English!)

I have re-assembled the carb and it is back on the engine.
Had enough of petrol and brake cleaner fumes for the day, so I will try again to balance and tune on Sunday while the Grand Prix is on...

I too share your thought why #6 will not flow the air to the same flow as all the other 5. I didn't have that problem before.

The throttle plates all close down to the carb bores evenly with the idle screw well free of the spindle. I 'tested' the 3 of them by holding then to the light. the small crescent gaps might be 0.2mm at a max.

See what sunday brings....

I thought Colour Tune was a particularly English thing but such a good product. Quite hard to find now in England, injectors don't have screws!

[Gordo2]

Originally Posted by 911hillclimber

I too share your thought why #6 will not flow the air to the same flow as all the other 5. I didn't have that problem before.

Good move on the thorough cleaning - had to be something in there causing restricted flow. My guess, most likely in your idle jets.

I had an idle jet plugged and the engine was way under powered. Visiting my brother (Cabmando), he suggested I check exhaust temps (cheap laser temp gauge pointed at each header pipe) - found it right off the bat. The culprit carb's exhaust port was ~ 180°F while others were ~ 325°F.

On the #6 cylender not drawing the same as others - oddly enough I have the exact same situation with my PMO's. I've investigated, but haven't figured out why. Meanwhile I'm able to balance them out with the idle air adjuster screws - I adjust #1-5 to match the #6 cylinder, with #6 cylinder's idle air adjuster fully open, the rest closed or nearly closed...

Incase you aren't already aware, you are receiving top notch advice from Mr. Paul Abbott - a true carb guru... Check his site for some great info on tuning and such http://www.performanceoriented.com

Good luck, hope the cleaning did the trick.

Gordo

BTW - agree with Cabmando, that vehicle looks like a blast.

[911hillclimber]

Indeed I understand Paul's status!

Odd thing is that the carb on 123 will adjust to '6' flow number on the synchroniser, #4 cylinder having the greatest flow (idle needle shut) at '6' but just the #6 refuses to pull '6' and will do '4' at best with the needle completely out bar 4 threads.

As I delve into this issue I'm understanding far more about the carbs and Paul's point about setting at a lower rpm, say 900 if I can to keep fuel from coming in from the lowest of the 4 holes.
.
setting the tick over adjust screws to 1200 does expose that hole quite a lot so adding to the fuelling.

I also have a 73T coupe with a stock 3.2 engine in it, bit of an old skool hotrod.

That engine is 130K miles in, never been apart, gets thrashed and runs like a dream...

Maybe throttle bodies are good after all!

[1QuickS]

From my flow bench and test engine testing of remanufactured carbs with new butterflies, spindles and journals and from CAD drawings generated from locational data of all components using my theodolite (very accurate optical transit) and my digital read-outs on my mill table I have characterized butterfly clearance with throttle bores on Solex 40PI, Zenith 40TIN, 40 & 46 PMO & Weber 40 & 46IDA3C carbs. The result is that first progression holes are routinely blocked by the edge of the throttle plate when at idle (IDTP Webers are the exception) and the radial clearance from the throttle plate to the bore is consistently 0.0025" or 0.064mm. This clearance is measured perpendicular to the spindle axis. Your clearance would be greater since you have been adjusting at 1200 RPM.

I have also found that the best way to visually observe butterfly alignment within the throttle bores is to remove the main venturis and sight through the bores from the top toward the bottom. Do this against a white cloud for the best light source. If you leave the main venturis in place then your observations will be a bit dodgy since your line of sight is not parallel to the bore and cannot see the clearance. A disparity in closing of the valves will be instantly obvious.

The point is this: I am worried about the throttle plate alignment within the #6 bore, something is not quite right. Even a VERY small amount of misalignment in one of the butterflies will seriously disrupt the parity of air flow through the carb at idle. The fresher the carb then the more sensitive this alignment becomes. Worn throttle plates bleed air at a higher rate and the resulting balancing air flows are less sensitive to a couple of thousanths of difference.

If #6 is mechanically closed more than #4 & #5 then it will generate more vacuum below the closed throttle plates and suck in more fuel than is combustible. Since 1, 2 & 3 are running well and you manually set the air flows using the spindle stop screws you have an unbalanced situation with the 4, 5, 6 bank, they are acting like an engine brake so 1, 2 & 3 want the engine to run faster than it can due to the braking action of 4, 5 & 6.

This is a best guess and I am surely interested in what you can report. Pull the 4, 5, 6 carb & pull the main venturis & have a good look at butterfly alignments and Let us know. It is possible that the pinch screws have loosened enough to let the valve slip in the slot in the spindle which would then make #6 close first relative to 4 & 5.

Otherwise I'm runnin' on empty here.

[Bob Kontak]

Originally Posted by 1QuickS

If #6 is mechanically closed more than #4 & #5 then it will generate more vacuum below the closed throttle plates and suck in more fuel than is combustible.

Best thread in a while. 1QuickS, thanks for your input.

I was thinking air sucker at first blush. Best I just read and shut up.

[Gordo2]

Originally Posted by 1QuickS

... Otherwise I'm runnin' on empty here.

I mentioned the #6 issue to Mr. Parr last year - he seemed to think it had something to do with the cylinder.

I also re-inspected my throttle plates last weekend when I had them off - no visible difference between 4,5 and 6. Additionally, all looked equally closed with the same amount of opening as I twisted the idle speed screw.

The good news is they seem to balance/sync up just fine at 3k RPM and up.

I will probably remove them again sometime this summer - will dig deeper when I do. I've never looked closely at the gaskets, I keep them in place / don't remove them when I remove the carbs (they stick to the manifold or bottom of the carb) - I guess it could be a gasket leak.

Stumped... good luck.

Gordo

[911hillclimber]

Excellent stuff gents!
When I had the carb apart yesterday, I had the tall venturi's out and the 38mm choke sleeves too. That was when I took a look at the discs and their clearances held up to the sky, just as Paul says.

The light gap and peripheral 'length' or crescent looked the same to me.
When I was an apprentice a very long time ago I earned the nickname 'micrometer-eye' and for good reason, so I'm running on instinct here, hard to measure anything this small to that shape and geometry.

The carb is back on and dry so I will pop it back off and look again at #6.

When the engine was working well last year it would go to the red-line easy and smoothly and sounded "great" according to track spectators and marshals, so this lot can work well.

At tick-over it seems that these carbs truly are instruments and fine, maybe less so at full-chat!

I have left the gaskets as they are, can't get new one in the UK, nobody stocks PMO spares.

The other reason for #6 not adjusting to '6' flow could be the cylinder, but the rough running followed the carb and the 'good' carb adjusted to '6' dead easy and the 456 bank was instantly converted to good running right from start up. I think the base engine is ok.

Shame I'm having this problem, but it is very interesting.

You learn something every day.

Thank you so much to everyone so far into this little adventure!

Looks like we are off on a Porsche club GB trip on Sunday as well as today, so tune-up on Monday, wife's instructions.