:: Diary - January 2014 ::

:: Wednesday, January 1, 2014 ::

After a rake around in the garage, I have found all of the necessary components for making this smoke machine. A Cadbury’s Roses tin with a buckled lid. A Cadbury’s Heroes lid, with a buckled tin. Fortunately, the Heroes lid fits the Roses tin. I have also found 6 ft of old garden hose and two hose clamps. I went to Screwfix yesterday and handed over £2.99 for a box of 75 smoke matches - they are meant for testing chimneys and flues and gas appliances and stuff like that. We’re in business!

I only needed one last ingredient - the necessary confidence to place my life at risk in the interests of science, for you, my faithful readers. Well, that and saving a few quid. Today, me being Scottish and this being New Year and all, I am too pissed to care whether I blow myself up, or not. How hard can it be, right? And hospital food isn’t that bad.

So - step 1 is to make a hole in the lid of the tin, big enough for the hose to go through, but a tight fit. Rather than drill it, I decide to batter a pointy drift through from underneath, so that the metal is opened up into little “tangs” that stick upwards. Push the hose through so that there is an inch sticking into the tin, and then put a hose clamp around the “tangs” on the outside to hold it in place.

Experiment 1… Take a smoke match, light it and put it in the tin with the lid off. Hmmm, not much smoke.

Experiment 2… try again with 4 matches. Much better, you get a good cloud of smoke…

Experiment 3… (See, I might be pissed but I’m still not rushing in where angels fear to tread) - light 4 matches and put the lid on, and wait for the smoke to come out of the hose, maybe under combustion pressure. It doesn’t.

When I take the lid off, there’s plenty smoke - it releases a cloud that would have a whole tribe of Comanches mounting their Palaminos. I need air flow…

I make another hole in the lid, cut off 2 feet of hose and feed that through and apply a second hose clamp around the tangs. Then I fit an airline adapter into the other end of that bit of hose, and clamp that.

Now, I don’t really need a lot of air pressure yet, just a flow to see if that pushes the smoke along the hose (and pressurises the plenum enough, that the leak will show), but not blow the lid off the tin. So I set the regulator to 3 psi for now.

Experiment 4 - light 4 matches and put the lid on, count to 15 (which is about the burn time of the matches) and then connect the airline to the short hose - and a cloud of smoke comes out of the other end of the long hose. Eureka!

That’s as far as I got - the next test is to try it and see what pressure blows the lid off - rem I mean “activates the integral safety feature to prevent over-pressurisation”. Ideally I’d like to get a wee bit more - an idling engine generates about 21 inches (mercury) of vacuum, which is equivalent to about 10psi. The commercial smoke units only seem to run at about 5 psi though. I think that there are bound to be pressure losses anyway, through open valves, valve guides and breather systems.

Then I need to test it on the car at a bit below “lid-launching” pressure, and see if that’s enough smoke and air pressure to find a leak.

Total cost so far - £2.99.
No medical bills (so far)
No “False alarm” callout charge from local Fireman Sam
No tribe of Comanches looking for a fight.

Here it is, this marvel of engineering, ready to take its place in my garage beside the £2.84 ECU fault code reader, the wood-and-string camber gauge, and various other bits of cobbled-up crap (all of which do the job!)

Note the careful sealing of the hose joints, to maximise the amount of smoke reaching the engine… See, quality!

:: Sunday, January 5, 2014 ::

Bloody cars - who would have them?

I haven’t got back to the TVR - I’ve just been too busy. And too scared to use my smoke generator machine.

Meanwhile, the Porsche has been safely along at the farm since mid-November. I decided yesterday to nip along and give it a start, just to circulate oil etc. Unfortunately I find that the battery is as dead as a dodo. There isn’t even enough power to put the lights on a a glimmer.

I borrow a charger and stick it on.

Today, after a 24-hour charge, I find that it starts and runs as normal. That’s a relief! I also find that the boot light, which wasn’t working at all, has now decided to stay on permanently, even when the boot is shut. That explains the flat battery, then. I disconnect the switch then put my own trickle charger on it.

Then it’s off to the TVR Car Club meeting, in Dave’s Seat. Only 3 brave / mad souls have brought their TVRs out, but there’s still a good attendance, and the usual chat about TVRs and sticky toffee puddings. Phil has found the dreaded outrigger rot, and is trying to lift the body off the chassis, but despite following the advice of the No TVR S web site, it still won’t budge. Dave and I agree to “drop by” on the way home to have a look.

And because we are gentlemen of our word, we do, and spend an hour and a half totally removing a couple of bolts (rather than just taking the nuts off) and disconnecting a couple of wires and the fuel filler. That’s better, but it’s still not for lifting. we run a wallpaper scraper up between the chassis and the sill, to dislodge all the grit that collects up there, and acts like velcro to hold the body in place. That seems to free it up.

Time is moving on, though, and Dave and I are not wearing our “already covered in garage shit” clothes, so we suggest that we come back on Thursday and help with the actual lift.

:: Thursday, January 9, 2014 ::

It’s always more fun playing with somebody else’s car, than your own, don’t you think?

Dave and I meet up, dressed in our “covered in car shit” clothes, to go through to Phil’s to lift the body off the chassis. We do it the way I did mine (and Jim did his afterwards) - by jacking it up at the back, then the front, then the back again, a fraction at a time, until it;s lifted by about 4 inches. It’s easy when you know how (and if you don’t know how, then see the “how to” page from the menu on the left).

The chassis isn’t too bad - it’s only corroded through at the front of the outriggers, much less severe than mine. We fire up Phil’s angle grinder and Dave cuts off the two bits of corroded tube.

So Phil needs to buy some tubing, and get it welded in. He has a metal fabricator friend so we’ll leave that to him.

Back home just in time before my local chip shop shuts - he’s got stuff left over so he gives me two suppers for the price of one. Result! See, kindly deeds are always repaid - just not necessarily by the person who benefitted the first time.

I really need to get my own car sorted out.

:: Sunday, January 12, 2014 ::

Right - I’ve faffed about long enough - a 4 week “faff” is as much as I can muster. I know that I’m an amateur compared to those who manage to put things off for 6 months, until 3 days before they need the car, but I just don’t have the stamina to do bugger all for that long).

So today, let’s fire up (literally) the smoke machine and see if we can find any inlet leaks.

Step 1 - I remove the air inlet hoses (the ones from the air cleaner) at the throttles, and stretch a rubber glove over each throttle inlet, and replace the hose. That means that the smoke stays in the plenum, and doesn’t just fill up the air cleaner.

Then I remove the brake servo hose from the back of the plenum, and connect up my bit of garden hose from my biscuit tin.. erm I mean smoke machine. Start up the compressor, set the regulator as low as I can (about 10psi or so). Quick call to the hospital to pre-book an ambulance, and another to make sure Fireman Sam is on standby.

Right here we go. I put 4 smoke matches in the tin with the heads together, and light a fifth and put in in, and put the lid on. I leave it a few seconds to let the matches burn out, and then snap-connect the air line.

It looks like the engine is on fire! There’s smoke pouring out of the back, near the distributor. I try to blow it away to see where it’s coming from, but the garage is so cold that my breath condenses and only adds to the smokey confusion. Never thought of that!

OK let’s try a different approach. I turn the regulator down a bit (below the first reading on the scale) and light another 5 matches and put the lid on. It’s still hard to see, but it seems to be leaking at the plenum joint to the manifold, in the back left-hand corner. There also seems to be smoke rising from the plenum bolts…

OK let’s try a second different approach. I mix some soapy water into a sprayer bottle, then connect up the airline straight to the manifold, bypassing the biscuit tin. I spray around the area that seemed to be smoking, and the resulting mass of bubbles shows that, as I thought, it’s that rear joint that’s leaking.

So are three of the plenum bolts.

So - what’s the verdict on the smoke machine? Well, it works - but too well! There’s too much smoke to perform a proper diagnosis, and it doesn’t last very long before you have to re-light more matches. It needs some improvements:
1. A better seal around the lid - a lot of the smoke doesn’t reach the engine at all;
2. It needs lower air pressure, I think, so that the smoke just wisps out, instead of pouring out and masking where it’s coming from;
3. It needs a way of generating smoke over a longer period of time, so that you have time to test where the leak is.

I’ll have a think about those for a Mk 2 version for testing how effective my next repair is! In the meantime though, I am dead chuffed that my £2.99 machine got me a result! Modesty forbids me from claiming to be a genie-arse, but… no wait, hang on - that was Modesty on the phone, and she doesn’t forbid me at all, so - I’m a genius!

In the meantime, though, I need to take the plenum back off to get to that gasket. I remove the throttle cable and disconnect the ISCV, throttle pot and fuel temp sensor, as I explained last time. Then I remove the 6 plenum bolts, and remove the plenum chamber.

I was hoping that the gasket might come off in one piece, but it doesn’t.

Here’s the bottom of the plenum - you can see the bit, bottom left, where the gasket wasn’t sticking - that’s exactly where it was leaking.

So I order up a new gasket set.

Then, while I am in the mood, I go along to my barn and remove the drivers seat from the Porsche, so that I can take it to get the piping repaired. I take a selection of tools, but there’s no drama, I only need one allen key - 4 bolts, pop, pop, pop, pop, and a couple of electrical connectors, and the seat’s out. Why can’t all cars come apart like this?

When I take the seat out, there’s a flip-flop on top of the ECU. Is that a code for some ECU modification? Or has somebody just lost it? If you’ve ever seen anybody hopping on one flip-flop around a white Porsche Carrera, then I own that car now!

:: Thursday, January 16, 2014 ::

The new gasket set was delivered on Tuesday, and it looks a much better quality kit than the one I had - although I bought that about 10 years ago just to get exhaust manifold gaskets, and it's taken me all that time to use the rest up...

I also got the Porsche seat delivered to the upholstery man, so progress on that front at least.

Anyway - today I start thinking about repairing the leaky plenum gasket. I have laid the plenum bolts down in the order I removed them, so I know which thread they were in. I try each of the bolts in the threads in the top of the manifold, and they are as tight as hell - I think because of a combination of crud on the bolts, and on the female threads in the manifolds. I think the threads are so manky that the bolts are reaching their torque setting long before there is enough clamping force on the gasket - so it leaks.

So, first I check the bolts, by clamping each one in a vice, and checking the thread size and pitch with a pitch gauge and a new nut the right size, which is also tight. I run a die up each thread, with a little bit of oil for lubrication, and test again with the nut - now it spins up the thread easily.

Once I have cleaned all six bolt threads, I start on the threads in the manifold itself, which is aluminium so soft and easy to strip. I try a new bolt and it feels tight, binding in the thread. I find the right size of tap, and carefully clean the threads, again with plenty of oil.

The secret of cleaning threads, in case you didn't know, is to take your time and wind the tap or die down the thread a quarter turn at a time, winding it in and out all the time to let any debris clear into the flutes. Don't just horse it down in one direction, all you do is create a "bow wave" of shit at the front of the thread cutter, and increase the risk of stripping. The other thing to remember is that, as you wind it in, you’ll push the crud and oil up the flutes in the tap, by hydraulic pressure. Wipe around the tap when it’s fully in, to clean up any dirty-looking lubricating oil, otherwise it just gets sucked back down the thread by the “air-lock” you create as you wind the tap out.

So with all six female threads cleaned, I dip cotton buds in some grease and use them to pick any loose crud out of the thread holes. And a lot of crud there is! Once I have done that, I run the tap down the threads again, with more oil, and pick out more crud in the flutes.

Now when I try the plenum bolts in their threads, they go in nice and smoothly. That should help to get the right clamping pressure on the new gasket. I want to do one more thing though - I want to squirt some carb cleaner down each thread hole, just to wash out any last remaking bits of crap. Only problem is, I haven't got any carb cleaner left, I seem to have used it all…

I scrape the remains of the gasket off both mating surfaces, and leave it for final cleaning and assembly.

I don’t have time to do any more today though - the demands of having to earn enough money to be able to afford to run these things (and keep a couple of daughters in food and board, with what’s left over) means that I need to get on with some work. And some shopping. So for now, this is Porky the pig saying - th-th-th-that’s all folks!

:: Sunday, January 19, 2014 ::

Right, let’s get this show (and this car) on the road…

First step is to clean the remaining hylomar blue, off the gasket mating faces between the manifold and the plenum. It’s designed to be resistant to just about everything, but carb cleaner takes it off.

Then I plug the manifold inlets with cloths, and squirt some carb cleaner into each mounting bolt hole, to wash the last of the crud out.

This set has 3 separate gaskets, so you have to be careful to match them to the ports. First I stick them to the manifold with some RTV silicone, then spread some more on the top, and put the plenum on.

Then I put a smear of sealer around the top of each of the bolt holes in the plenum, and then a little oil on each of the bolt threads, before I put the bolts in, nip them up and then torque them to the right settings.

Then it’s just a matter of reconnecting the fuel temp sensor, the throttle sensor and the idle speed control valve, and then replacing the vacuum hoses for the map sensor and the crankcase ventilation system. Next I refit the throttle cable and mechanism.

I connect up my smoke tester to the brake servo connector, and clamp it tight, and put a rubber glove over each throttle intake. This time, there’s no sign of any escaping smoke, it seems fine. I pull off the map sensor hose and smoke pours out, so it is working!

It’s looking better. I start it up, but it won’t run unless I screw the idle screw in about 3 turns - it’s been running on leaked air for ages! Once it’s warmed up a wee bit, I adjust the idle speed to 850 rpm, and it sounds ok. Better than that, when I rev it, it returns straight to the right idle speed.

I decide to have a wee drive. It seems to run fine, there’s no hesitation when accelerating, and when I’m coming home, it pulls up the hill to my house in third gear at idle, no problem.

I think (I hope!) I’ve sorted it! Another result for the incompetent amateur.

:: Thursday, January 23, 2014 ::

I need to make some final adjustments to the engine. First, I was reading the spec sheets, and they say that although the 2.8 engine's idle speed should be 850 rpm, the 2.9 should be set at 900 rpm plus or minus 50 - so I've set it at the lower end of the range.

First thing, the ignition timing. After I warm up the engine a bit, I loosen the distributor clamp bolt, and connect up my timing light. It's one of the old basic ones, I think I bought it in 1976 or something, and one of the early xenon ones. It replaced an older neon one - anyone remember them? They were bleeding useless - the light was so faint that you had to do your timing in the pitch dark, so that the light showed up at all - either that or you needed a darkroom curtain over you and the entire front of the car, just so that you could pick out the tiny glimmering timing mark that you could see. Xenon lights were a technological marvel in those days, aye…

Then I disconnect the throttle pot and idle speed valve, which my old article in Car Mechanics magazine says, puts it into service mode (ie no ignition maps apply). There’s some disagreement on whether or not that works, but it’s worked for me in the past so I’ll run with it. When I start the engine and point the light, the timing is a wee bit too far advanced, so I set it back to 12 degrees BTDC, and reconnect the throttle pot and tighten up the distributor.

Then I set the idle speed to 900 rpm, which seems a bit better. It revs ok and doesn't run on at high idle when you lift off.

It seems ok, but yet... I don't know, maybe I'm just a pessimist, but there's something not right. I try another smoke test, but everything seems fine.

I've got the car booked back in to the engine tuner next Thursday, so I'll find out then, eh?

:: Thursday, January 30, 2014 ::

Up bright and early to take the car to the engine tuner. First I have to refit the left hand inner wheel arch, which doesn’t take long. I’ve practiced it often enough…

Then we’re off! This is the first time I’ve driven the car since I reset the engine idle speed and the ignition timing, and it’s absolutely… well, crap to be honest. It’s low on power, one huge flat spot, and backfiring when you move away, as if the ignition is retarded, or poor fuel - or injectors connected wrongly. But it was ok last time, so the only difference is timing, which is now (by the book) correct.

I get to the tuner, and the first task is another smoke test with his eleventy squillipounds machine. I’m pacing up and down like an expectant dad while he starts the test. To my surprise, it’s airtight, just like my biscuit tin and garden hose tester said it is!

So it’s on to test 2 - he connects up the diagnostic tester and we start it up. He starts with a “cylinder power comparison” test that shows that although no 3 cylinder is a wee bit down, the rest are all equal - and they wouldn’t be if the injectors were connected out of sequence. So I got that right as well!

So then it’s on to ignition timing - he checks what it’s set at, and it’s 12 degrees, as I set it. After a lot of faffing around aka experimentation, allied with our discovery that the distributor is moving round while the engine is running, we reach the conclusion that it runs much better at about 17 degrees BTDC - which, when I check with his light, is exactly where it started from when I initially aligned the distributor by eye, before I reset it to spec last Thursday. Very mysterious.

He thinks that the timing chain tensioner might be worn, so setting the timing by the crankshaft mark is leaving it retarded. He’s confident that he’s sorted it though.

He also sets up the throttle pot (or rather, makes sure that it’s already set to spec) and then goes on to check the lambdas. One of them is erratic, and the other is hardly working at all. So I need to change those.

However, a wee test drive shows that it’s running well, with no pinking, so I’m happy. Oh, and a steady idle speed, which is where we came in 2 months ago…