Date:         Sun, 25 Apr 2010 10:53:37 -0700
Reply-To:     Alistair Bell <albell@SHAW.CA>
Sender:       Vanagon Mailing List <vanagon@gerry.vanagon.com>
From:         Alistair Bell <albell@SHAW.CA>
Subject:      Re: PWM controller ,
              was Water Cooler System Design Flaw Workaround?
Comments: To: Tom Hargrave <thargrav@HIWAAY.NET>
In-Reply-To:  <04eb01cae49f$33852fe0$9a8f8fa0$@net>
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed

Sure Tom, those parameters are good and all, but is it really that
much less complicated than an old school  controller with pots for
slope, offset etc?

or, lets all go modern and run it off a custom made control strategy
built in LabView, a laptop and a D->A interface.

:)

I really need to stop lazing around talking smack ,

cheers

ab




On 25-Apr-10, at 10:46 AM, Tom Hargrave wrote:

Yes, we are talking PID now and I can probably design something that
would
do the job. I suspect you could too since you brought up PID.

But I wonder if the solution can be done a little less "high tech"
and all
12V powered?

For example, there are a set of known parameters for pulling a long
hill:

1. You would want to know that the transmission was not in high gear.
2. You would want to know that the engine was above a certain RPM,
let's say
above 3500 RPM for arguments sake?
3. You would want to know that the engine was making power - throttle
open &
not coasting down a ling hill in lower gear to conserve breaks.
4. You would want to know that the vanagon met these two parameters
for a
specific period of time - maybe 15 seconds?

Once these four tests were met you would switch on the fan or an
auxiliary
fan to pull all the heat you could from the radiator.

Once test 1, 2 or 3 were no longer true for 15 seconds, you would
switch the
auxiliary control off. If this were a separate auxiliary fan it would
turn
off but if it were the main fan the fan would still be managed under the
vanagon's system control.

Tom