So I'm pretty certain i've got to the bottom of why my car failed.
I didn't know whether to post this for fear of causing another shit fest but as I'm as much to blame as anyone else for this. I thought I may as well.
Because I only rebuilt the engine using more or less the same components as the original builder I made some assumptions that I shouldn't have.
The main one being that because I used the same rods, same pistons, same cams and same head gasket thickness that I wouldn't need to check the Piston to Valve Clearance. Turns out I should have but there are other contributing factors too.
First off here's a picture of the pistons when i first built the short motor. As you can see all clean with no carbon present.
Here's one of the pistons as it now stands.
As you can see there are witness marks where the valves have hit the pistons.
However as you can also see that these marks are also covered by carbon which means that the Piston to valve contact isn't recent. Had it been recent they would have been clean and not have any carbon on them.
So the PTV contact was historic, but when and more importantly how? Everything was built to ensure this didn't happen: high pressure double valve springs, and a thicker than stock head gasket stock is 0.6mm the previous builders used a 1.1mm gasket so i did too. Also a conservative rpm limit of 7800 was set up and although I'd hit the limiter on the strip a couple of times the engine had never suffered an over rev.
So that led me to look at the tuning.
Before we go any further I want to say that I'm not looking to appoint blame (like i said I'm at fault here too) but I did want to find out why my engine failed the way it did because it shouldn't have done. Obviously I also want to prevent it from happening again too.
Anyway, Since I built the car it has been run on two different tunes by two different tuners.
The running in was done on the original tune by the original people who built the car and covered approximately a very steady 1500kms. (Please no one name them it's been done to death and we all know who and that's really not the issue any more.)
The second tune was started by a different tuner but got aborted due to coolant leaking out of one of the exhaust ports on the cylinder heads due to a cracked casting. So the original tune was put back onto the
ECU.
The engine was pulled and two new cylinder heads were fitted. All the same components were re used.
After refitting the heads a tuning session was carried out where it made 643 at the wheels on Maximums dyno. The car has done around 8,000 kms since i swapped the heads.
A few people disputed the power figure achieved at MMS so I took it to Tokyo Motorpsorts Hub dyno for a comparison run.
First run made just over 600 at the wheels, 605 I think it was. Then two subesquent runs it made 575. At the end of the third run just after coastdown the car just stalled and wouldn't restart.
So after stripping down the engine it was found an inlet valve on Cylinder number two had dropped.
Struggling to reason why it would be a build issue I decided to look into the tuning side of things.
So I investigated the AVCS timing on the inlet cams.
Here' the timing from the original tune:
and Here's the same table from the last tune I had done that made 643
First tune runs 48 crank degrees AVCS advance (24 cam degrees) on the inlets.
This is the map i used to run the car in with.
Second Map runs 38 crank degrees AVCS advance (19 cam degrees)
The car has used this map for aprox 8,000kms.
So armed with this information I decided to actually physically measure the piston to valve clearances.
So using a degree wheel I measured where the cams would be when advance is applied and marked cam pulleys and timing belt accordingly.
Each tooth on the cam pulley is 15 crank degrees or 7.5 cam degrees.
So to get as close as I could to the 24 cam degrees (48 crank degrees) of the original tune I had to advance the cam three cam teeth which is 22.5 degrees. 24 degrees is midway between the next tooth. I marked the timing belt 24 degrees but the nearest tooth was 22.5 degrees.
So with everything marked up properly I put some playdough on the valve pockets and assembled the engine and timed it up with the correct advance applied.
.
Once assembled I did one full rotation of the crank. Then removed the heads and could physically see the valve clearance by how much the play dough had been squashed.
Anyway here's the results.
original tune at 24 degrees cam advance (nearest tooth was 22.5 cam degrees).
The impressions aren't round as I think the playdough is a bit too soft and it settled during the time i took to get the head off and takes pics after the valves made the impressions.
Then I took the heads off, put some more play dough on reassembled it all and measured it again only this time it was measured for the second tune which was set to 38 crank degrees (19 cam degrees). The nearest I could get this was one tooth less advance which works out at 30 crank degrees or 15 cam degrees advance.
As you can see there's a fair bit of difference in clearance between the two.
Just to note that all the measurements were taken with the original head that had the cracked coolant gallery.
So my conclusion is that there was a little bit too much cam advance applied and the piston to valve clearance wasn't enough to cope with it. I should have checked it but didn't. Now I'm assuming that a valve got slightly bent during contact with a piston and has just fatigued over time which ties in to what the valve manufacturers have said in email communication I have had.
Obviously both block and heads were refaced for the build so that has to be factored in but according to Galloways who did the machining there wasn't a significant amount machined off any of the faces.
Also just comparing the Coworth Pistons to the Wiseco ones I've just ordered. It would appear the valve reliefs are better placed on the Wiseco's
Anyway I believe I have found the cause of failure and will re measure everything again when it comes to build time after I have the sleeves honed for the new pistons.