Saturday, May 30, 2009

Thursday, May 21, 2009

Head rebuild and new rod bearings.

With confirmation of the blown head-gasket and the likelihood that the head is warped means that it has to be removed. Given the audible rod-knock and the oil leak at the rear main seal it appears the easiest solution is to remove the entire engine to fix everything at once.

This leaves two options, rebuild or replace with used. After checking a number of sources, I wasn't comfortable that any of the used engines currently available were going to be any better than a partial rebuild.

After discussing my options with a number of reputable engine builders, I think I'm going to completely rebuild the head to OE spec, replace the rod bearings with the pistons and crank still in the block. A full rebuild is tempting, but the additional cost of machining the block and rings/bearings is prohibitive ($2k). Most builders think the mains are bulletproof and BMW rings should be good but will need a good look while things are apart.

Some early decisions include using ARP head bolts and rod bolts, installing a VAC Oil Pan Baffle, and upgrading the valve keepers.

Monday, May 18, 2009

Bummer -- Blown Head Gasket

It looks like I lost the head gasket at the PPIR event a month ago. It slightly overheated and I thought I pulled off the track in time but it appears that I did not. The car shows no symptoms on the street but on the track after extended sessions (20+ min.) it starts to slowly overheat and after pulling off the track the cooling system is fully presurized and spit about a half gallon of coolant. I drove a couple of days at HPR with it blown but it was cold enough (30s-40s) that it was also asymptomatic. However in the 85 degree heat of Pueblo it failed pretty quickly.

My options are:

1. Replace just the head gasket.
2. Replace head gasket and rebuild top-end
3. Rebuild entire motor
4. Swap to used motor

Option One is a bad choice for a track car as the head is likely warped and will need to be surfaced to provide reliable sealing for more than a couple of months. Not gonna do it.

Two is pretty attractive. It's relatively inexpensive, will provide a long-term solution and can be done in a home garage. There is really only problem with this. It's not enough. In addition to the headgasket, I also have a Exxon Valdez style oil leak from the rear main seal. So to get to that I need to pull the engine or drop the trans.

Rebuilding is also attractive -- it seems better than a used motor of unknown history but it is really expensive. This on only really an option if I do the work myself. Labor on a full top and bottom rebuild is well over $5000 and can approach $10k. It appear that this can be done at home for about $3500k in parts and tools

The final option, a motor swap, is pretty attractive, 3.2L S52USs seem to sell for $2500-$3000. This is a little more displacement than my current 3.0L S50US and by using all my OBD-1 accessories it is a simple swap.

The real decision here is: who does the work? I have the technical ability and the tools to do it myself. I just don't think I have the time. I spoke to a local shop and they reccomended the swap as it is the most labor effiecent option. In fact they think the full motor swap is cheaper than a top-end.

So any way you look at it, this is a $2500 -- $5500 unbudgeted expense. Shit.

Tuesday, May 05, 2009

HPR Analysis

Here is a track map and speed/distance chart of a 30 Min session at HPR in the M3. I'm actually quite proud of this session as I was focused on carrying more speed into corners and clear improvments are shown in the data.

This was my first session on Nittos and the average lateral G for the session was .2 G higher than a similer session on Kuhmo MXs a month earlier. It was cold and blustery -- 45-50 degrees F and overcast.


Segment 1

Turn 1 is a fast, flat 60 mph left and the end of the second longest straight. The apex is slightly obstructed by the pit wall on entry. On the first laps I was slowing to 57.4 mph, over the next 7 laps that increased to 61.5 Mph. Minimum speed is directly related to exit speed, so this is worth 3/10th in this segment but will also help in the next segment.

Segment 2

Turn two is a 180 degree sweeper and there is a 3-2 downshift under braking and a 2-3 up shift 100 ft after track-out. I've tried the whole thing in third and the data shows that the shift is worth 1/10 in the ~10 second segment.

Segment 3

Turn three is a little kink leading onto the straight and the most important corner on the track.

This is a corner that is easy to overdrive, it is tempting to carry a bunch of speed in, but the really fast way is to be really smooth in and get in the throttle before the apex. There was a 5/10 improvement from lap 2 through lap 7 and an increase in minimum speed from 60.4 mph to 65.8 mph. Unfortunately there is not a total correlation between the highest minimum speed (at the apex of 3) and the highest maximum speed (at the exit of the segment). There is correlation between the highest minimum speed and and the shortest time for the segment.

Segment 4

Segment 4 is the end of the back straight, a slightly uphill right hand 90 with a slightly blind apex. As you can see from the speed/distance chart I am braking later and slowing less with each lap. It's really all about not slowing down too much.

Segment 7

This is a high speed, steeply uphill sweeper that falls away near the top. To is easily one of the most entertaining corners on the track.

What is shown in the middle of the segment is less and less braking at entry. Notice how the speed conserved early is carried all the way to the end of the segment.

Faster in is faster out here -- the more speed you can carry in, the more speed you retain climbing the hill. There is a 5/10th sec and almost 3 mph difference between the best and worst.

Segment 9

Another entertaining corner, downhill, off-camber, with risky run-off. Again the speed/distance chart shows slowing less and less each lap.

In summary, the data shows a consistent improvement, taking little bites, and the reward is better laps.

From Lap 2 (2:13.57) to Lap 8 (2:12.45) there is a 1.1 sec reduction in lap time. The Traqmate claims the Theoretical Best lap (combining all the best segments into a single lap is 2:10.05, So I have plenty of room to improve consistency.




Here is a friction circle for the same session. Notice how accelleration is grouped -- the strongest (in second gear) is shown by a dark line near the bottom of the chart, third gear acceleration is shown by the darker line above.

This indicates that this is a HP limited car, since it can accelerate near the max and turn near the max at the same time. It also shows that this is a clockwise track because the left side is darker than the right.