|So sad to see this happen. Rod saw an
air/water unit like this on a USA. internet site, recognized straight
away the advantages for his new blower set up & sussed out having it
fabricated in Sydney. Weeeeeeeeeell, the results of what he was charged
for are shown in the top two pictures, beside the unit we fabricated for
him. The guy sounded totally confident & competent, so Rod just went
ahead & ordered it. The sad thing is that the shop took money for this
botch up, gave it to him knowing it leaked. If you haven't seen someones
work, make sure you do before committing!
It almost looks like they ladelled the alloy filler on, like a person
who can't solder! One of the worst "professional" jobs I've ever seen, &
it's not as though it's for a low dollar car either.
The air/water unit we made was 66mm longer & 8 rows of tubes high
instead of 6 rows, giving 48% more cooling area straight off. Plus we
shaped the inlet to trap 90% of the charge air instead of approx. 50% of
the other. There is a surprising amount of cooling on offer here because
of the surface area provided by the two core matrix. Because of the tube
length, a fair water speed through the core is needed. If this was a
tougher engine combo, then we would've routed the water differently
through the cores, approx. 20% more cooling is available, but because
the engine is a sensible street combo, this unit will remove most of the
heat out of the charge air anyway, so a waste of our time & his money.
ARE's in house computer programme gives us the volume & speed of the
charge air, the heat kw. that needs to be pulled out of this air, how
long it's in the core for (only milliseconds in this case), speed of the
water through the core, how long it's in the core & how many Kw's heat
the water can transport out. Then it's a matter of adjusting the water
flow, sizing & configuring the front radiator, sizing the hoses ( that's
another programme we have developed) & it will work pretty bloody good!
Actually we're not this smart - as of the end of '03 our programme
has an accuracy of +/- 6% for most calculations, up to +/-15% for some &
total accuracy for some. By the end of '04 I want these figures @ 3% &