The Manhattan Project

[Dougieboy1]

jealously reaching a dangerously high level!!!...... soon, i shall be able to drive mine...i hope!

Thanks for the other info on the forms

Regards

Andrew

[Dougieboy1]

me again being boring. I started reading the 'how to fill it in' form. Lots of references to data taken from the Cert of Conformity or IVA. Do i presume these are the questions you left blank?

txs

[Nine One One]

Just got some temporary plates made so can hopefully start running and fettling the car this bank holiday weekend.
Interestingly the guy making the plates commented that my V5 was of a style he'd never seen before - for a moment I thought he was going to claim it was a forgery!

They have all changed again - there will be no previous keeper details (not that this applies to you Andrew) recorded on them anymore due to ‘Data Protection’.

The fact most second hand cars will come with service receipts, with the previous owners details on them, seems to have escaped DVLA.

They are also meant to be harder to forge, like the 5 previous attempts that went before them!!!

[AndrewSlater]

me again being boring. I started reading the 'how to fill it in' form. Lots of references to data taken from the Cert of Conformity or IVA. Do i presume these are the questions you left blank?

txs

Hi Andrew, yes I left those sections blank. Anything that states 'Take this information from the Certificate of Conformity' I left blank, other than the question on number of seats - as I managed to count them.

With my temporary plates fitted ( althought the white/yellow ones are growing on me ) the first thing to do was take up the offer from the MOT centre to go back and use their emissions tester to set the CIS up properly.

We had adjusted the gear selection again hoping that this time we would have a full house of gears.

Previously at the test centre the CO value was a little high as were the HC values but I didn't have the tools with me to adjust the mixture at the time.

Knowing the engine was running a little rich I had leaned it off a smidge on my return.

We arrived at the MOT centre with at least three gears selectable and this time the bloke at the centre fitted the exhaust sensor, calibrated the machine and immediately announced I was running far too lean, which surprised me. He left me and my son to our own devices in charge of the machine but after 20 minutes or so of making adjustments and not really seeing any effects on the exhaust analyser we both concluded we didn't trust the readings and that the car was definitely now misadjusted.

So we experimented with the positioning of the exhaust sensor and found a position where the results were more repeatable and started the process again. Conscious that we were running on a nearly empty tank we cracked on again to get the CO level to around 2.0% ( spec is 1.5% to 2.5% ). This still gave a high HC value but at this point we ran out of time.

Off we went to find some fuel so we could then look for a few more gears and then go for a longer run.

At the petrol station we bumped into Bootsy and his new Range Rover - the omens were good!

[AndrewSlater]

It had now been a few weeks since the car was registered and we had started the process of running the car in and troubleshooting.

The first problem encountered was that the spark plugs were fouling after 20 or 30 miles.
At first I experimented with different plugs with different heat ratings but to no avail.

Despite the issue we managed to get to the Simply Porsche event at Beaulieu as it was within range of a spark plug change.
As we spluttered into the grounds I was hoping for a chance to quickly park up as the car was running only on four cylinders.

Luckily ( or unluckily ) we were singled out to park in the Arena parking, which meant having to slowly follow an official on foot for what seemed an eternity to our parking spot. Long before we got to our spot the engine died but fortunately the route was downhill so we could coast most of the way much to the amusement of the official . We were then shown our spot and I quickly started the car reversed it in as quickly as I could where it promptly coughed and died again. We quickly abandoned the car to distance ourselves from the woeful entrance much to the annoyance of the Macan owner in the next space ( who had to realign his car to even out the untidiness our presence had brought to the event ).



Over the next week or so I would go for a run and check / change the plugs on return and lean out the mixture a little.
After a number of iterations we got to the point whereby four of the plugs seemed OK but two consistently fouled - cyl 1 and 3.

I wasn't 100% sure whether this was oil fouling and would improve as the engine ran in (or when I changed to a thicker oil) or whether it was a symptom of overfueling.
At this point the car is still running the Millers running in oil which is thinner than the final oil I will be changing to, but as the car has only done 100 miles or so I don't want to change the oil until I know the engine has run in properly. The limit on the Millers oil is 500 miles so I can run on this for some time yet.
I rechecked compression on the engine and all looked good so decided to consider the overfuelling issue.

I removed the injectors from the inlets and rechecked the fuelling.



I found that the fuel lines to cyl 1 and 3 are running about 15-20% higher flow than the others, so have concluded that fuel distributor isn't correctly balanced / calibrated. Now when I rebuilt it many moons ago I was pretty confident it was just about in spec of around 10%.

The Bosch specs that I can find are as follows:-

Setting point 6cc/min: delivery 6.0-6.8 cc/min ( 0 - 13% )
Setting point 40cc/min: delivery 40.0-44.0 cc/min ( 0 - 10% )
Setting point 120cc/min: delivery 120.0-131.0 cc/min ( 0 - 9% )

This points to why I can't get a mixture setting that works for all cylinders - even at the limit of the Bosch spec it might be a bit of a challenge.

So the task at hand was to get my fuel distributor rebuilt and to get each of the cylinders much better matched.
I checked with a UK rebuild specialist but they wanted a 10-12 week turnaround so I decided to rebuild it myself again.

How difficult could it be!

[AndrewSlater]

I'm conscious that this thread is a little behind events, so will try to add a little brevity to catch up.

Before rebuilding my fuel distributor I had to first understand what had gone wrong with the build last time and what I would do different.

After a lot of reading around it became clear that one of the important pieces of information that is missing from the rebuild instructions supplied by different sources, is that they don't mention that each port is shimmed to take out manufacture tolerances.



In the picture above each of the ports has a shim, spring and end cap that sits in the housing between the metal diaphragm. ( the photo only shows one of the spring/shims/caps - obviously there are six in total)

Having stripped down my FD again I noticed that the shims varied in thickness from 1.08mm to 1.28mm - so it probably mattered whereabouts they went back together, and oblivious to this gem of information had simply refit them.
Measuring the FD body the depth of the cavities seemed to vary up to 0.2mm so my guess is that the shims were matched to take out the variance in the machining of the cast iron core.
Interestingly I measured all the spring rates of the springs and they were still closely matched to within 2 or 3 %.
I also measure the end caps and found some variance in these also but only about 0.05mm or so.

Having stripped down the FD I also noted that the recently fitted diaphragm was dented on a number of the ports, which was strange as although the ports were mismatched I wouldn't have expected any damage. The plot thickened but before I could rebuild I now need to buy at least another new diaphragm.

After a bit of research I found the following kits.



What I liked about these kits that I hadn't found in others was that they supplied new springs, end caps and also the small fuel filters for each of the injector lines.
The only thing missing in my view from the kits is a selection of shims.
So I ordered a couple of kits and waited a month for them to ship from Australia.

Which gave me plenty of time to think about how I would properly test the rebuilt FD.

[Bootsy]

Ha ha! If only!

[AndrewSlater]

I decided that I didn't want to have to test the rebuilt FD back on the car, but thought that an investment of time into a test rig would make life easier in the long run.

Here's a photo of the first attempt at replicating the setup in a test jig - the jig would later evolve to include a fuel accumulator, filter, and better fuel vessels.
Notice the high spec fuel pump and adjustable pressure regulator to allow me to mimic the WUR function and pressure gauge to verify system pressure.



Shout out to my son who helped along the way. ( The crap bits are my work the good bits his )

As part of the rebuild I measured and re-measured everything plenty of times.

The one thing I couldn't understand was that knowing where I had incorrectly fitted the shims the first time around, the maths didn't stack up with the two ports that were overfuelling - something strange was happening which was pointed to by the dented diapraghm.

Even rearranging my existing shims / caps and springs in the best possible combination the physical tolerances weren't that great.
In the end I pooled an extra sets of shims from a spare FD to give me the best possible match to the new springs and caps.
I then spent ages sanding down shims to get even close to the best accuracy I could given my resources. At the end everything matched to 0.01mm which was my measuring accuracy.
Now given that all the shims I had seem to fall in 0.1mm increments I was guessing this was better than factory accuracy.

Now as part of the rebuild I reground the FD halves to get a mirro finish on the surfaces using 1200 wet and dry, ultrasonically cleaned everything, crossed my fingers and rebuilt the FD.



Now I had to test it to prove it had the perfect specification

[AndrewSlater]

On the first test I was consciously checking all of the connections for a fuel leak when I spotted the system pressure gauge sitting at 10 bar - WTF!

I quickly switched the fuel pump off and double checked the plumbing - all seemed logical so I tried again - still 10 bar - grrr!

Turns out I'd forgotten to remove the system pressure regulator valve when I had ultrasonic cleaned the case halves, and it was now stuck in the barrel ( flash rust ).
Two days later and having tried compressed air, compressed grease, and a number of other unsuccessful attempts the piston was finally out, flash rust and all.

Having carefully cleaned out the piston and the bore it was all refitted and tested.
The system pressure was a steady 4.5 bar which is at the low end of the spec, but should have given consistent results.
The test jig gave me pretty poor matching on the injectors outputs with differences of up to 50% on some runs.

Time to strip it again.
On disassembly the diaphragm was significantly dented on all ports, so I'm guessing it didn't enjoy the 10 bar system pressure - oh well another diaphragm for the bin, lucky I had bought two rebuild kits.

Rebuilt again the FD showed better matching and was in spec in most of the measurement ranges. However whenever I ran the low flow test (idle) of 6cc/min I could get different results each time.
I added a fuel accumulator and filter to the setup to better represent the car setup but still got varying results.
I even bought some new injectors to see if it was an injector problem but still got varying results.

At this point I thought it was best to put the FD back on the car to see if I got more consistent results.

On the car the system pressure sat at 4.5bar indicating I would need to reshim the regulator to get to closer to the ideal 4.9 bar.
First I would run some flow tests to see if the results were different on the car.

Once I started to run tests the system pressure would climb, on high flow rate tests it would climb to 5+ bar, on the low flow rate tests it was reading 7.5 bar (bugger another diaphragm damaged!)

What was going on?

[jb]

Perhaps you should talk to Kmi Injection http://www.kmipetrolinjection.co.uk
They are very knowledgable and friendly and I feel sure they will give you good advice.
Maybe get them to check it out for you.

[AndrewSlater]

So after careful consideration, I decided the only fault could be that the fuel return line was restricted.
After disconnecting the other end in the engine bay and trying to blow through the pipe in a scientific way it was clearly it was severely restricted.
Further investigation showed that the fuel return line was trapped between the gearbox and bodywork.

So it looks like when I refit the engine last September to fix the clutch arm issue, I had inadvertently trapped the line causing a restriction.
This is turn had caused the system pressure to run too high and had again deformed the diaphragm in the FD.
So when the FD was first fitted and tested it was in spec, only it had become damaged later on.

Oh well mystery solved.
I'm glad I didn't give up and have the FD professionally rebuilt as clearly that wouldn't have spotted the car issue, and the newly rebuilt FD would have been damaged again.

Oh well - light at the end of the tunnel .

Once more the FD was rebuilt with a new diaphragm.
This time I reshimmed the fuel regulator to hit 4.9 bar.

At high flow rates the injectors are matched to within 3%, and are well within spec at all rates with the exception of the 6cc/min rate whereby I still don't get consistent results but would say it is just borderline around the 13% limit.

Now I just needed to see if the car ran again (hopefully with a consistent mixture across all cylinders). - What could go wrong!

[AndrewSlater]

Perhaps you should talk to Kmi Injection http://www.kmipetrolinjection.co.uk
They are very knowledgable and friendly and I feel sure they will give you good advice.
Maybe get them to check it out for you.

Thanks for the advice jb. I had contacted KMI but they were running a 12week + turnaround.
I'm glad I didn't get the FD rebuilt as in the end the car had a fault that was damaging each FD rebuild.

I've gotten quite good at rebuilding them, they are quite simple things when you understand them and quite satisfying when it finally goes right.

Hoping to get the car running properly for the Goodwood Breakfast club meeting this Sunday - so touch and go at the moment.

[AndrewSlater]

Now that the return line restriction has been fixed and the FD rebuilt, the plug fouling seems to have gone.
I've gone back to a colder plug and put 80 miles or so on the car. Previously by 20 miles two had blackened to the point where the car ran rough.

I reset the mixture to approx. 3% CO as a baseline.
With the new setup the plugs all look pretty consistent after 80 miles.


Part of the testing included making it to the Goodwood Breakfast Club.

Even got a picture of the car on the Goodwood website!
https://www.goodwood.com/grr/event-cove ... ar-sunday/



The car drives pretty well especially at high speed / high revs, but is a little hesitant around 2.5K rpm.

The next step is to go around the loop of timing, dwell, mixture extra to see if I can dial it in a little better.
All good fun!

[jjeffries]

Thanks very much for documenting this; I have an 82 SC and am a CIS appreciator so all relevant and fascinating content for me.
Would the blocked return line have showed itself in fuel pressure tests (as shown in Bentley, CIS manuals, etc)?
How are you measuring CO? I put an AEM gauge in mine, but now keep a cover on it because I don't need to be looking at it all the time. That in turn makes me think a decent sniffer would be better...keep it in the garage and use it on any car.
Your car is stunning.
Best, John

[johnM]

What a great result Andrew, top marks for perseverance, many would have given up.

Sounds like you are just about there.