DeTomaso Mailing List: June 2001, Message #276

Looks like I forgot to send this one too, also in my drafts file:

Okay, I'll bite:

1. Does anyone believe you can transmit 20 horsepower by wrapping a single,
small V-belt around only 90 degrees of a pulley about 5 inches in diameter
at 5000 rpm?  Plus the HP going into the alternator with the same belt?

2. The electricity driving the pump has to come out of the alternator, I'm
assuming this car is expected to not operate as a "total loss" electrical
system, so that's going to take away from any net HP gain.

3.  What's the linkage between taking the "surge" tank out and adding an
electric pump?  Why not just take the "surge" tank out, plumb around where
it used to be, put the cap somewhere else (at the highest point, hopefully)?

----- Original Message -----
From: Richard Barkley <rbarkley@earthlink.net>
To: Multiple recipients of list <detomaso@realbig.com>
Sent: Saturday, May 26, 2001 11:25 AM
Subject: Re: What I learned about cooling systems this week


> David,
>
> Mike Trusty is planning on installing an electric pump. From my
> discussions with him he seems to have researched it fairly well. You
> should talk to him. Seems to me the main benefit is the added (20?) HP
> at high rpm.
>
> But before you run off and do it, there are some things you should know.
>
> The heat transfer rate of a given radiator on the water side of the
> equation is dependent on two things: #1) the temperature difference
> between the air and the water (most important) #2) the water to metal
> heat transfer rate.
>
> #2 can be effected by the velocity of the of the water in the tubes.
> Below a certain velocity, you get laminar flow in the tubes and a much
> reduced heat transfer rate. That's why we have a 2 pass flow radiator:
> the velocity is 4x what it would with a cross flow design. Beyond a
> certain velocity, there is probably not much to be gained here. My guess
> is that the stock water pump with the stock radiator is beyond that
> point. That may not be true (but probably is) for a replacement radiator.
>
> To first order #1 is the difference between the average temp of the
> radiator { (in temp - out temp)/2) } and the ambient temperature. Also,
> to first order and assuming #2 is not an issue, the temp drop across the
> radiator is inversely proportional to the flow rate. So double the flow
> rate gives 1/2 the temp drop. So you can measure how much increase in
> cooling performance you would get from increasing the flow rate.
>
> Lets run some numbers. Let's say you have a 20 deg temp difference (200
> in, 180 out) once things get heated up and the thermostat is wide open
> so you're getting full flow. BTW the 190 thermostat on my 4.6L Cobra
> engine is not supposed to be fully open until something like 220 deg.
> Anyway, the average is 190 deg. If we tripled the flow, The drop would
> be about 7 deg and the average would be 196.5 deg, a 6.5 deg
> improvement. So if it is 100 deg outside, there would be about a 7%
> improvement and your car would run about 7 deg cooler as you'd have to
> run at 207/187 to get the same cooling. I think there are cheaper ways
> to get a 7% (if it is as much as that, I have no idea what the relative
> flow rates are for the electric vs mechanical pumps). One, you could
> reduce your idle rpm about 7 - 10%. To first order idle power/heat is
> linearly proportional to engine rpm. Or you could by some better fans.
> The high output strait bladed SPALS are probably the best on the market.
> Unfortunately they run about $150 each.
>
> If you decide to go electric and it has significantly more flow
> capacity, you might consider using a cross flow radiator to minimize the
> resistance to that flow. I suspect their flow rates are fairly sensitive
> to the head pressure. Also, I'd be concerned about the electric pumps
> ability to supply the engine with enough flow rate at high power/rpm. If
> a standard pump is using 10 to 20 hp at high rpm unless it is terribly
> inefficient it is going to be supplying a lot more flow than the
> electric pump which may be on the order of 1 hp (I see ads at 4 to 6
> amps which at 12V is about 1 hp max).
>
> BTW, I'm running Wilkinson's Fluidyne radiator with my 4.6L. I told him
> I wanted my money back. It's got WAY TOO MUCH cooling capacity for that
> engine :). At idle (70 deg outside) and the hood up the engine temp
> won't go above about 200 deg with the electric fans OFF; just the
> convective air flow is enough! I decided to keep it 'cause it looks so
> cool. One reason the engine is so cool is that it idles (very smoothly)
> at about 600 rpm.
>
> Richard Barkley
>
> "David and Marilyn Bell" <mdbell@ev1.net> wrote:
>
> > <snip>
> > One way to take the swirl tank out of the high pressure side of the
cooling
> > system would be to plumb in one of those high capacity, inline electric
> > water pumps.  A good location might be in the main coolant tube between
the
> > front luggage compartment and the radiator (laydown).  That way the
inlet
> > side of the radiator would see the full pump pressure and the swirl tank
> > would only see low side pressure.  Of course you'd have to remove the
> > impeller from your mechanical water pump to make it work (you'd still
need
> > the pulley to drive the alternator).  You'd also have to have a well
built
> > radiator and heavy duty alternator.  Come to think of it, that extra
relay
> > on Rick Moseley's electrical upgrade kit could come in handy for
powering a
> > water pump.
> >
> > I'm considering that approach as a way to improve the idle speed coolant
> > flow and reduce idle temps.  The biggest problem is that it's an
expensive
> > alternative.  I know this has been around the list before - but has
anyone
> > out there tried an inline electric water pump on a Pantera?  Or for that
> > matter, any car?
> >
> > Dave Bell
> --
> Sincerely,
> Richard Barkley
>
> To Ski or not to Ski, that is the question!
>
> Richard Barkley
> (310) 373-6695 (home), 813-2432 (work)
> E-mail: richard.barkley@trw.com (work), rbarkley@earthlink.net (home)
> Mammoth Condo 2BR+loft/3Bath - Horizons 4 #186 (760) 934-6758
> Condo web address: http://home.earthlink.net/~rbarkley
>