Date:         Wed, 16 Jul 2003 15:43:07 -0400
Reply-To:     David Brodbeck <gull@GULL.US>
Sender:       Vanagon Mailing List <vanagon@gerry.vanagon.com>
From:         David Brodbeck <gull@GULL.US>
Subject:      Re: 2.1L coolant light problem
Comments: To: gary hradek <hradek@YAHOO.COM>
In-Reply-To:  <20030716193052.48547.qmail@web41310.mail.yahoo.com>
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Wed, 16 Jul 2003, gary hradek wrote:

> I have always observed that there is a  bit of a
> disconnect between what you see as an engine
> temperature(the light) and what the fan in the front
> is doing(sensor system with three speed fan).   I
> suspect the temperature sensor for the fan is poorly
> placed.

It's placed halfway through the radiator.  That's an odd but reasonable
spot.

I've noticed that the fan and the engine temp are often "out of phase",
because the fan is monitoring the radiator temp, not the engine temp.
What I mean is the engine temp will start to climb, and a minute or two
later the fan will switch on.  The engine temp will drop, and a minute or
two later the fan will switch off.  In a long stretch of stop and go
traffic it can almost look like the fan is running mostly at the low
points in the engine temperature.  I figure this is because of that long
coolant loop between the temperature you're looking at, and the one the
fan switch is seeing -- it takes a while for the cooler water to make it
back to the engine.  But the fan is doing its job, keeping the water
coming out of the radiator "cool enough" by assisting the airflow when
necessary.  This temperature is only *indirectly* related to the
temperature of the water leaving the cylinder heads.

Another thing to keep in mind is that the temp gauge on the dash is
bimetallic and has quite a bit of lag itself.  If I see a temperature rise
on my aftermarket gauge it's generally 30 seconds to a minute before I see
the dash gauge respond.


David Brodbeck, N8SRE
'82 Diesel Westfalia
'94 Honda Civic Si