I am thinking more about this graph of IAT2 over time as the Cadillac STS-V accelerates:

Let’s consider the stock, OEM Bosch intercooler pump.  This pump is rated for 8 gallons per minute (gpm) against no resistance, and likely does 5 gpm in the STS-V system.  It is overall a very good pump, and very efficient.

Now, a gpm is 1/60=0.083 gallons per second (gps).  So 5 gpm would be 0.42 gps.  The STS-V can accelerate 0-60 in under 5 seconds.  So on an average 0-60 run, the intercooler pump moves 5 sec x 0.42 gps = 2 gallons of fluid through the intercooler.

Now, keep in mind that the OEM system had 2.6 quarts of fluid, and my new HX added 2 quarts, so now my system has 4.6 quarts or just 1.15 gallons. A small part of the capacity is likely counted in the side fill loop, so we’ll say 1 gallon.

So on an average 0-60 run, the pump can move 2 gallons of fluid through the intercooler, and the system has 1 gallons in the main loop, so the coolant sees the inside of the intercooler twice in that 5 seconds.  Put another way, my current 4 quart capacity system has a circuit time of ~3 seconds — every ~3 seconds the system circulates the entire coolant capacity.

For the intercooler testing I was focused on just getting an acceleration spike and not best time.  But if we look at this graph of speed vs time vs IAT2,

What it appears to show is that the IAT2 temps are near constant for the first 3.27 seconds (looking at the exact figures in the data table), and rise dramatically after that.  Now it is dangerous to read too much into a single sample, but this seems to validate our figures that at 3.27 seconds the intercooler coolant starts a 2nd loop through the system.

From this I conclude that a system with a 2 gallon capacity of coolant would have fresh, cool fluid in the intercooler for the first 6 seconds.  Now 0-60 mph is not the end all and be all of performance.  To keep uncirculated coolant in the intercooler for a dyno run one might need 10 seconds of fluid or 3 gallons — as a thumbrule think 1 gal per 3 seconds.

Also, if you speed up the pump, you’ll need more system capacity to offset the faster circuit time.

What do you think?  Am I missing the boat?  Do you agree?  Hit the comments please!

This morning I took the front clip off my STS-V once again to replace the intercooler heat exchanger hoses with Spectre braided stainless sheathed hoses and then retest.

I did a first shake-down test run, then let the car sit and heat soak, then did another 20 min test capture.

The chart shows intake air temperature 2 (IAT2) after the supercharger over time during a 20 min test drive. The spikes on the chart are acceleration runs.
Today’s weather is overcast, with 72F air temperatures, similar to the historical data file for the OEM capture. What this suggests is that the S3TC heat exchanger in series with the OEM heat exchanger acts to lower IAT2s around 10F.  The end of the test was curtailed due to a traffic stackup; I stopped scanning and capturing data to save time on conversion.

Because today’s temps are similar to the historical air temps, I compared today’s run to the historical data without modification.

Overall the new S3TC Heat Exchanger results continue to be good.  The Spectre hoses are on and do appear to have better success at avoiding kinks or crimps than the rubber heater hoses did.

Here is a zoom-in on the acceleration run today, showing the trend for IAT2 vs RPM and Speed.

In this graph we see initially there is actually a decrease in IAT2, as the throttle opens, all the flow goes through Supercharger as opposed to the bypass, and the intercooler is at max cooling. As the supercharger continues to dump heat into the system, the manifold heats up more than the intercooler cooling system can sustain.