CIS full of gum

[Hil911]

Hi again

For some reason my previous thread seems to have disappeared completely. Does this come from being Australian and ungodly?

Anyway, I have a serious problem in my fuel injection.

I bought a gasket kit for replacing the diaphragm from Precision Machinery. The car wouldn't run afterwards. I checked injector delivery and 4 to 6 were injecting 10 times as much as 1 to 3. When I came to take the regulator apart, all the screws were loose. The reason is clear in the photos. The glue on the plastic film has absorbed petrol, swollen, turned to goo and squeezed out.

Acetone softens the goo more than petrol but I don't know how I am going to clean the lines and injectors. Hopefully whatever was carried back to the tank will end up blocking the fuel filter.

Has anybody seen this before? Is there something weird in Australian fuel? Any suggestions apart from a new car? And I previously sealed the regulator with ITW Permatex Form a Gasket #3 and THAT had dissolved out as well!

Depressed,

Hil

[E Sully]

Here is a good thread on the fuel distributor. CIS fuel distributer cleanout w/pics
I believe the best initial method is to avoid any type of sealant when assembling a fuel distributor. There are very small pinhole orifices in the plate to bleed small amounts of fuel that must be left clear.
If it leaks with no sealant, then a very thin and carefully applied sealant can be tried, as recommended in the thread.

[Hil911]

Hi Ed

Thanks but I did that! The 2.7 CIS diaphragm is steel. Some are rubber and they will self seal but I don't think there is any way that the steel-steel-steel combination will keep petrol in. I originally applied the zero clearance sealant mentioned above, in minimal quantities. It is the same that can be used to join 911 crankcase halves. Anyway, despite the manufacturer's claim of being petrol resistant, it washed straight out and the steel-steel-steel combination bled petrol like crazy.

I've seen a thread that used Indian Head shellac but that is also marketed by Permatex and I wonder if it will also be soluble in Aussie petrol, like the ITW. I've chased down an aircraft fuel tank sealant that should do the job. So now I just have to clean everything out.

I can't believe that I am the only person that this has happened to.

If you don't hear from me again, I was probably killed by one of the solvents that I tried.

Hil

[T77911S]

you dont use sealant.

if all the rubber orings have been removed, get a can of the carb cleaner (not the spray) and soak it in that. no plastic or rubber, that stuff will eat it up.

now you have another problem other than removing the sealant. thats lining up the plunger barrel that has the 6 ports to the fuel heads 6 slits.

put a dab of grease on the Orings to get them to slide back in.

good luck.

[Targalid]

I use methyl ethyl ketone (MEK, 2-butanone) to remove adhesive goo. Unlike other organic solvents it does not usually soften plastics or paint. Try it on a small section and see how it works. It is usually found in the paint section of big box stores.

[The cap'n]

I thought MEK was removed from the market years ago. It's REALLY nasty stuff, and is absorbed through the skin. Special breather masks are recommended, and gloves that won't let it through. Goes for the liver, I think.

The Cap'n

[boyt911sc]

Originally Posted by The cap'n

I thought MEK was removed from the market years ago. It's REALLY nasty stuff, and is absorbed through the skin. Special breather masks are recommended, and gloves that won't let it through. Goes for the liver, I think.

The Cap'n

Got a pint of MEK from our local HD store recently. This is a highly toxic solvent and should be used only in a well ventilated work area. For those not familiar handling hazardous substances or chemicals, use caution when working with MEK. BTW, gasoline is also a toxic liquid.

Tony

[T77911S]

trust me, the carb cleaner will do it. it comes in a gallon can. i had a 5 gallon can of it back in the 90's, use to soak my carbs in it. it WILL take the paint off and it WILL ruin anything not metal. but ya never know, like all the good stuff somebody prebably drank it thinking they could get hi and now its watered down.

[porterdog]

Originally Posted by The cap'n

I thought MEK was removed from the market years ago.

In CA maybe.

[Hil911]

I have an aerosol of throttle body cleaner - which I think is basically carb cleaner. I'll try that.

I run a lab, so I can use fumehoods etc to protect my bodily well-being! Still, I can't bring anything really nasty in without paperwork.

T77911S - I'm afraid that you will have a hard time convincing me that no sealant is required! The pressure on the diaphragm does not seem adequate for a steel-steel seal. All the more so as when my sealant washed out (the screws were still tight after, so it had negligible thickness) petrol seeped steadily out of the join. I've already rebuilt the unit completely twice - first time with sealant and the second time with he gaskets, so the 'O' rings and I are getting to know each other well! I have to say that the first build with sealant worked perfectly and the injectors all delivered evenly (tested with a stop watch and measuring cylinder) until it began to leak.

The Precision Machinery diaphragm and 'O' ring kit is good, it's just the gaskets that caused me trouble. Actually I kind of wonder about the whole idea. The nozzles in the distributor are machined flat to the surface and therefore sit against the diaphragm at rest. Adding a gasket means that the diaphragm will have to distort more to close the nozzles, which may shorten the life of the diaphragm if nothing else.

Hil

[E Sully]

Originally Posted by Hil911

I have an aerosol of throttle body cleaner - which I think is basically carb cleaner. I'll try that.

I run a lab, so I can use fumehoods etc to protect my bodily well-being! Still, I can't bring anything really nasty in without paperwork.

T77911S - I'm afraid that you will have a hard time convincing me that no sealant is required! The pressure on the diaphragm does not seem adequate for a steel-steel seal. All the more so as when my sealant washed out (the screws were still tight after, so it had negligible thickness) petrol seeped steadily out of the join. I've already rebuilt the unit completely twice - first time with sealant and the second time with he gaskets, so the 'O' rings and I are getting to know each other well! I have to say that the first build with sealant worked perfectly and the injectors all delivered evenly (tested with a stop watch and measuring cylinder) until it began to leak.

The Precision Machinery diaphragm and 'O' ring kit is good, it's just the gaskets that caused me trouble. Actually I kind of wonder about the whole idea. The nozzles in the distributor are machined flat to the surface and therefore sit against the diaphragm at rest. Adding a gasket means that the diaphragm will have to distort more to close the nozzles, which may shorten the life of the diaphragm if nothing else.

Hil

I have taken apart 3 early distributors, and there is no sign of sealant being used.
The gaskets seem to be a new product. I have to agree with you that using them, when previously there was nothing, would seem to change the calibration. Any way to contact the manufacturer for information?

[Hil911]

I can't say anything about the manufacturer calibration but just another word on sealant. I was sent to a posting where a guy had rebuilt a 928 fuel distributor. He used Indian Head shellac, diluted and painted on with a little brush. I basically did the same. Honestly, when I split mine again, you would never have noticed the sealant. It closes up basically metal to metal.

Hil

[bfunke]

Hill - I agree with your approach and using Indian Head Shellac.

[Hil911]

I thought that I should update this thread.

I rebuilt the distributor with no gaskets and used Loctite 515 sparingly to seal the diaphragm in. I put an extra bead at the outside where a bit of squidging wouldn't do any harm by blocking ports.

I then screwed the distributor together firmly and wiped off the excess Loctite that came out. With regard to blocking ports, I have a hope that where the sealant is not locked in a tiny crack, it will remain liquid (because it hardens anaerobically) and therefore will not block ports.

So far the distributor is not leaking fuel and appears to be functioning correctly - all injectors were delivering approximately equally. On that subject, I would like to repeat the line in the Porsche manual where it states that the injectors serve NO metering function. I started out by thinking that irregular delivery was partly an injector problem.

One other thing to add: some of these anaerobic sealants take quite a while to harden up. I put my fuel distributor aside for a week in a warm cupboard before I fitted it and pumped petrol in.

Hil

[Hil911]

My car is still not running right.

I decided to see how well the injection was balanced at a relatively low flow rate. I used a target rate of 20 mL/min, which I think works out as about 2 U.S. gallons per hour.

I pulled all the injectors and shoved them into jars, apart from the one I wanted to measure that I stuck into a graduated glass cylinder. After turning on the ignition, I wedged open the CIS vane with a plastic block to try to get the same flow every time. I realise that although I disconnected the cold start valve, I did nothing about the WUR. But as the vane was wedged that probably doesn't matter as the WUR works by biasing the vane's travel. The flow rates seem to increase through the testing on average. Here are my relative rates:

Cylinder Flow/min
1 21.4
2 23.3
3 22.4
4 24.4/23.6 with repeat
5 23.2/22.9 with repeat
6 24.3

Does that seem like a typical variation for a light load? Maximum variation is (24.3-21.4)/24.3 = ~12%

[RarlyL8]

Bosch doesn't sell rebuild components to the general public so whatever you purchased is pot luck. The seal is a thin film of pre-dried shellac. Tolerances are very tight so there is no room for slop if perfect function is desired. Every component and every line needs to be flushed to remove all residue or this gift will keep giving for months to come. The fuel tank should be flushed as well and the FD and WUR rebuilt. Not what you want to hear I know but this solvent soluble gasket is now everywhere in the system.

[Hil911]

Thanks Crotchety for those pearls of pessimistic wisdom!

Actually I have done most of that: I put all new lines in, new injectors, new accumulator, cleaned and rebuilt the distributor. I didn't clean out the tank. I rely on the fuel filter to catch that crap. The WUR was rebuilt by the previous owner and I cleaned my mess off the mesh in the inlet. Of course the WUR has no influence on my unequal flow issue.

Anyway, I did a retest. I got six jars and weighed them empty - so I tested the 6 cylinders simultaneous instead of one after the other. I then ran the system for 4 minutes at the same flow rate - about 20 mL per minute. I got similar rates to what I got when I did them one after the other .

So I juggled injectors around to try to switch the highest and lowest flows. I did manage to reduce the variation. So apparently the injectors may not meter (according to Bosch) but they do influence the flow. At this low flow rate, I now have 10% variation. The biggest problem is the port on the distributor that feeds cylinder 3, which is 7% down on the average. I changed injectors several times but it stayed low.

The distributor is non-adjustable so a low flow is due to a weak spring or a stiffer bit of diaphragm. It could be a partial blockage but it was all really clean. If I take it apart again there is no guarantee that anything will change.

Hopefully the situation is better at higher flows.

[RarlyL8]

Flow is metered continuous to all cylinders (the C in CIS). The fuel head is adjustable but I wouldn't do that as it masks a problem.

[E Sully]

RarelyL8, there are fuel ditributor rebuild kits on ebay for the Bosch fuel distributors. bosch fuel distributor rebuild porsche | eBay
Here is a thread started by Walt Fricke on his balancing efforts on a later style fuel distributor. Earlier fuel distributors are tough, as the only way to adjust is to change the shims under the springs inside the fuel distributor.
CIS FD balancing flow to injectors
When you applied the sealant, it is critical not to get any squeeze out into the chambers, and also the primary restriction port for the control pressure. There is a small hole in the thin steel plate for this orifice, circled in red.

As for not cleaning the fuel tank and relying on the fuel filter, if the filter screen in the bottom of the tank is clogged, the fuel pump will not get enough flow from the tank.

[Hil911]

Yup - continuous it is!

As E Sully noted, Crotchety, the fuel head is fixed on these early ones.

I tested max fuel flow and there is no problem. These tests were done at about 20 millilitres per minute, which is a very low flow. I notice that most people have tested at very high flows and earlier tests that I did at high flows looked more even. The reason that I did low flows is because the engine is at its worst on light transition. Scream off the lights and it is fine, accelerate gently in traffic and it bunny-hops.

I was pretty careful with the ports on the rebuild - I think that it's my fourth go! The 515 that I used is runny when aerobic and so should go through even the small hole in the diaphragm. I also used it sparingly as it should squeeze to about nothing.

I think that I have another diaphragm that I could use. I assume that when Bosch built it originally, they would have used six matched springs and shims. I can't see them doing trial builds and tests! You'd have to assume that the diaphragm had homogeneous properties.

Could I get springs and shims if I needed to? Or would I end up trying to match from old parts?

I'll have a look at the thread that you linked to and see if anything jumps out.

Thanks for both your inputs.