<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML xmlns:o = "urn:schemas-microsoft-com:office:office"><HEAD>
<META content="text/html; charset=UTF-8" http-equiv=Content-Type>
<META name=GENERATOR content="MSHTML 11.00.9600.18639"></HEAD>
<BODY id=role_body style="FONT-SIZE: 10pt; FONT-FAMILY: Arial; COLOR: #000000"
bottomMargin=7 leftMargin=7 rightMargin=7 topMargin=7><FONT id=role_document
color=#000000 size=2 face=Arial>
<DIV>Hi list,</DIV>
<DIV> </DIV>
<DIV>Since the list seems quite quiet these days I thought I'd post something
that might be of interest. I replaced the 20+ year old aluminium alloy
radiator in my 69 Sprite a couple of months back with a similar aluminium alloy
radiator that I'd drilled a hole for and had a weld bung welded to (duh it was
supposed to be made with this!) for a radiator temp sensor. I made
appropriate wiring modifications including fitting a relay (so as not to draw
the full fan current across the sensor switch (no such relay exists for the
toggle switch!) for the sensor but leaving in place the decades old toggle
switch as a manual override. Having fitted a brand new switch and found it
did nothing I tested it in boiling water with a jury-rigged test set up and when
it still did nothing I returned it for a replacement. Fast forward to this
evening when I decided rather than sit in front of this PC I'd test both
the replacement switch and another one with a different (lower) temp switching
range. Here's the results:</DIV>
<DIV> </DIV>
<DIV>
<P class=MsoNormal style="MARGIN: 0cm 0cm 10pt">For the first test I used a
saucepan of water being heated on a domestic gas cooker hob with an 88-79 switch
wired to a 9V battery (of unknown condition) and a brand new Draper test light,
and proved the switch worked (more than can be said for its brand new
predecessor) at about boiling point, albeit it seemed to switch off
abruptly.<SPAN style="mso-spacerun: yes"> </SPAN>The second test only
using a 82-68 sensor was more encouraging with a clear on and slower off
although the on point seemed similar<o:p></o:p></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 10pt">The 88-79 switch was fitted to
my Sprite and the car run up to temperature and with a temperature gauge reading
of 115C the fan had not been switched on by the sensor so I used the manual
override switch until the gauge reading dropped below 90C.<SPAN
style="mso-spacerun: yes"> </SPAN>I then removed the 88-79 switch and
replaced it with the 82-68 switch and ran the engine whereupon the fan was
switched by the sensor at about 112C and was switched off at about 100C (it may
have been 98C and thus 14C between on and off as expected).<o:p></o:p></P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 10pt">With testing finished I
concluded that the coolant temp in the top of the rad is probably a lot lower
than the coolant temp in the cyl head (where the temp gauge sensor is
fitted).<SPAN style="mso-spacerun: yes"> </SPAN>It might have been an idea
to have fitted the sensor lower down in the rad.<SPAN
style="mso-spacerun: yes"> </SPAN>It’s possible both sensors switch on at
a temp much higher than their design states, based on my boiling in a saucepan
test and practical testing.<SPAN style="mso-spacerun: yes"> </SPAN>Not
least I’ll be buying a switch with an even lower switching range.</P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 10pt">Finally - I know my STACK water
temp gauge will also switch the fan (and I have a spare relay for that purpose)
but I wanted to get the temp sensor working first.</P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 10pt">Final, finally - anyone know of
a solurce to buy a neat Aluminium alloy M22x1.5 rad plug because the nylon ones
are rubbish and I don't want a steel or brass one.</P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 10pt">Regards</P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 10pt"> </P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 10pt">Daniel</P>
<P class=MsoNormal style="MARGIN: 0cm 0cm 10pt"> </P>
<P class=MsoNormal
style="MARGIN: 0cm 0cm 10pt"> </P></DIV></FONT></BODY></HTML>