June 2008 - Dyno Test

Took my car to Marlin Motors who have a MAHA 1000BHP 4WD rolling road. Met Martin, who's a decent bloke. Had a natter then strapped it in. Main problem is that their very powerful blower doesn't reach my radiators and importantly my charge cooler radiator - So my inlet temperatures wouldn't come down.

The other problem is that its an automatic, so you have to play around a little to get it to stay in the same gear during the run - but Martin managed it without too much trouble.

So what does it do mister?

Dyno gave 230BHP at the wheels, 370BHP calculated at the flywheel. Considering that its a 5.7 running 9Lbs of boost I was expecting closer to 250 at the wheels - 400 fly. Given how it drives on the road and the amount of trick gearbox parts I've broken, I thought it would be putting down more torque(not power). The datalogs showed that the inlet temperature hit 50'c - which is pretty high.


I have entered the exact data collected during the run (specifically turbo inlet temp, AFR, RPM) into Eric Fahlgren's Turbo calculator and then adjusted the engine efficiency value until I gave the same calculated BHP as the dyno figure.


Using that as a baseline, I then adjusted the intercooler efficiency to achieve a more reasonable 30'c inlet temperature (10'c delta over the ambient) - Erics calculator shows I could potentially see close to 400BHP which seems a fair figure from a 5.7 running 9lbs of boost. So I now have an excuse to start a new project, namely a front mount intercooler. Interestingly If you change the boost to 0 psi, it gives a lousy 250BHP - from a 5.7! Datalogs

If you look at the torque curve above, you will notice that there is a "sag" in the mid range - which is where you would expect the car to be make the most torque. The datalogs show that full throttle was at 4800 and full boost at 5300. At 3500RPM, we only had 2.7 psi of boost - whereas on the road you would get full boost earlier, subject to traction. I suspect this is a feature of trying to nurse an automatic gearbox not to change down when doing a power run.


Dyno Datalog:

Purple is throttle position, not 100% till about 4800 RPM. Green is boost:


Comparison with a road datalog, notice that the boost pressure is much straighter from earlier - which should fill in the torque dip:


Chargecooler Radiator Testing

Given my high inlet temperatures and the unwillingness of the chargewater to reduce in temperature, I wondered if my chargecooler radiators were working effectively. So I performed some experiments with them to determine if water is flowing over the entire core and the effect of adding fans.


First I joined two Jaguar XJR chargecooler cores together and added a small fan - I plumbed it into my test bench filled with hot water. What was apparent straight away is that the fan wasn't pulling enough air through the core (as I had attached it directly to the core). So next I made a cardboard duct spaced one inch away from the core - the difference was very significant. I measured the ambient temperature and timed how long it took to get the water down from around 45'c down to ambient.


These are intercooler cores designed for inside the manifold - the water performs a double pass inside each core. I know that normally, you are cooling hot air with cold water - but they should be just as efficient at doing the same thing in reverse. I just needed a baseline to compare against my actual radiator core.


Next I removed one of my actual radiators and created a dual-fan duct and repeated the experiment.

The first thing it did prove that water flows throughout the entire core as whole matrix became hot as soon as the pump came on. Then when I turned the fans on the temperature dropped rapidly, far quicker than the Jag intercooler cores - which I hoped for.


This graph compare the results from both cores. You can see that there is a rapid drop in temperature because there is a large delta between ambient and the water, but as the temperature comes down - the delta is smaller and it takes increasingly longer to cool the water.

In normal driving I don't think the water temperature exceeds 25'c, so we can see it takes about 2 minutes to return the water back to ambient.


On Road

Whilst the radiator was ducted with the fans, I attached it to the front of the car and went for a drive - it made no difference at-all. I conclude that the air flow to radiator cools the water as best as it can. When on boost, essentially there was a 20'c temperature delta over ambient. When idling the car, the turbo inlet shoots up to around 45-50'c - the chargecooler keeps the manifold temperature around 32.c

Next steps

So my next steps are to play around more with keeping the turbo inlet temperatures down, with insulation and relocating the airfilters.

July 2008

I have been testing and playing around with the air filter layout. From datalogging, I can see the air arriving at the inlet of the turbos is significantly over ambient. Through some testing I have established that this is due to both the location of the air filter and the fact that the induction pipes aren't insulated. Surprisingly, even with the inner wings removed and the filters attached just behind the front of the car, I was still getting a 10 delta temperature increase. Clearly, there is masses of warm air swirling around the inner wing area.

The only way to get cool air into the filter was to attach it to the front bumper.

/Front bumper air filter pic/


Taking that a step further, I adapted a std land-rover air box - which also seemed to work quite well.

/landy air filter/

Again through datalogging, the inlet temperature is still high until you get onto boost which has sufficient air flow to cool things down. So overall, an improvement - but nothing ground breaking.