Date:         Fri, 26 Feb 2010 09:55:25 -0700
Reply-To:     Keith Hughes <keithahughes@Q.COM>
Sender:       Vanagon Mailing List <vanagon@gerry.vanagon.com>
From:         Keith Hughes <keithahughes@Q.COM>
Subject:      Re: external temperature gauge
In-Reply-To:  <BAY0-PAMC2-F6c5haUY00018c13@bay0-pamc2-f6.Bay0.hotmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

> Date:    Fri, 26 Feb 2010 07:23:16 -0500
> From:    Mike S<mikes@FLATSURFACE.COM>
> Subject: Re: external temperature gauge
>
> At 12:12 AM 2/26/2010, Tom Hargrave wrote...
>
>> Correction - 100' would be 3 - 5 ohms, still a non-issue with a 100K
>> thermistor or 1K RTD.
>>
> It would significantly effect an RTD. A Pt1000/.384, would change at
> least 2.3F. Since the inherent inaccuracy of the lowest spec RTD (Class
> B) is about the same, you'd have a thermometer only accurate to ~5F.
>
> The more common Pt100 would be off over 20F!

I doubt seriously that you'll find a platinum RTD (even a thin film
version) in an aftermarket automotive gauge application, and Pt1000's
are not very common, and not in 2-wire configurations IME.  2-wire RTD's
tend to be Nickel with an alpha=0.00617 vs 0.00385.  A 3-ohm error at
100°C for a Ni1000 probe would be more like 0.4C or 0.7F, which will
almost certainly be much greater accuracy than the gauge itself.  If you
have a 3-wire or 4-wire RTD, lead length is basically irrelevant
(irrespective of material or base resistance) as long as the extensions
for all leads are the same length and gauge.

Keith Hughes
'86 Westy Tiico (Marvin)