mg-mmm
[Top] [All Lists]

Re: Water Pump Impeller & Cooling generally

To: Bill707@aol.com, "Kahler, Jack" <MGJack@aol.com>
Subject: Re: Water Pump Impeller & Cooling generally
From: Patrick Mullen <pmullen@direct.ca>
Date: Thu, 10 Feb 2000 17:56:06 -0800
Cc: NA-MMM <mg-mmm@autox.team.net>
References: <63.13d38fe.25cb1761@aol.com>
Reply-to: Patrick Mullen <pmullen@direct.ca>
Sender: owner-mg-mmm@autox.team.net
Jack, Bill:

Many thanx indeed for the beautiful impeller which has just been
delivered by Canada Post to our house - I didn't even need to collect it
from our friendly "Letter Carrier" just over the border. I appreciate
the gift very much.

I hope to test it in the NB when I have it running, possibly versus an
original worn one AND against the one in a Barry Walker jobber pump I
bought 20 years ago. The challenge will be to design an experiment which
shows up the differences.
I have done a lot of thinking about cooling MMM engines but have had no
practical experience (at least not since I blew a cylinder head gasket
in my water pump-less M-type in a traffic jam for the 1951 Farnborough
Air Show).
 
How do I test it under real life conditions when the ambient temperature
will probably change with the time of day while I change the impeller?
What if the three impellers are of slightly different blade widths? (I
wd prefer to keep the same pump body for all 3 impellers for the
experiment).

What variables wd be meaningful to measure?

- Obviously temperature in the header tank is one. This wd probably only
be meaningful under racing conditions or under arduous conditions such
as driving foot-on-floor on a midsummer afternoon up several miles of a
good gradient Interstate Hwy such as the Siskiyou Pass on I5 between
Oregon and California.  
- I like Pip Bucknell's suggestion of putting a restraint in the hose
from the cyl head to the header tank (see 1998 Y'book p30). Pip
suggested this to pressurise slightly the water between the pump and
the  h.tank hose; this in order to reduce the bubbles on the metal
surfaces in the block & head and hence presumably to increase the heat
transfer. By measuring this pressure we wd have some idea of the
efficiencies of the various impellers.
Any other ideas?

While on the subject of cooling water:
 (a) What about the late Geoff Coles's article in the 1971 Y'book p28
where he points out that many of the prewar racers blocked off the water
flow thru the block mainly or fully, and diverted it to flow into the
core/frost plug in the rear of the cyl head? This is of interest to
everyone now that we no longer have leaded fuel. I imagine that the
potassium based fuel additive available in Anzacland and even a cyl head
with hard inserts for the exhaust valve seat wd perform better with the
coolest possible water available to the exhaust valve seats - it also
keeps the bores warmer and hence less likely to wear due to overcooling
I imagine. Has anyone tried this?
 (b) How about using a (failsafe?) thermostat as Pip's restricter? This
wd avoid overcooling. Didn't the KN version of the N engine have this?
Where was it placed?

Comments and advice please, Pip?, Bob Rich?, all gurus? in OZ-UK-NZ-NA
et al.

Cheers.....Pat


Bill707@aol.com wrote:
> 
> In a message dated 2/2/00 16:10:21, MG Jack writes:
> 
> << Bill...Will send Pat an impeller tomorrow at no charge. Good test case!
> Jack >>
> 
> Pat... we're granting you a "comp" impeller due to your tireless work on the
> tech articles etc... I havn't seen the ones that Jack has made... they were
> done on his auto-router, whereas mine were done much more laboriously on a
> CNC mill. I specified that the hole be 15/16", or some such, but slightly
> under .5" to allow you to ream to your exact shaft diameter. Also you'll have
> to provide for your own fixing screw.. I believe I used a 10/32 set screw.
> Good luck!
> Bill
> 
> -----------------
> Forwarded Message:
> 
> Subj:   Re: P/N gears and water impellers
> Date:   Wednesday, February 2, 2000 16:10:21
> From:   MG Jack
> To:     Bill 707
> 
> Bill...Will send Pat an impeller tomorrow at no charge. Good test case!
> 
> Jack

<Prev in Thread] Current Thread [Next in Thread>