[TR] Water pumps.

[Michael Marr]

This is a myth.  The rate of heat transfer in a heat exchanger is determined by the temperature difference across the heat exchange interface; the area of the heat exchanger; and the heat transfer coefficient, U, which is a constant unique to the heat exchanger.  The flow rate of the fluid through the heat exchanger does not affect how much heat is transferred.  What is affected is the temperature change to the fluid.  The heat either absorbed by the fluid or, as in the case of a radiator, given up by the fluid, is equivalent the the fluid's mass flow, times it's specific heat, times it's temperature change.  For water, the specific heat is 1, so the equation becomes:

Heat transferred in Btu/hr = mass flow in lbs per hour x deltaT

A gallon of water weighs about 8.33 lbs, so mass flow in lbs/hr = gpm x 8.33 x 60, which is close enough to gpm x 500 as to make no difference.  These are US gallons, Dr Fine!

What this means is that, for a radiator of fixed geometry, operating at a constant ambient temperature, changing the flow rate of the fluid cooled by the radiator will simply change the deltaT of the fluid, because the amount of heat exchanged is constant.  It will not result in less, or more, heat being transferred.  If you want to remove more heat from the engine, you need to put in a radiator with greater heat transfer area, pure and simple. 

On another note, the number of blades on a pump impeller will not change the flow rate of the liquid through the pump, with all other aspects of the pump's geometry remaining the same.  Theoretically, a pump with a single blade, operating in the same casing, will pump the same volume as one with a six or eight bladed impeller. The only reason pump impellers have multiple blades is for reasons of mechanical and dynamic balance.  

A rule of thumb for engines is that 33% of the energy in the fuel is converted to mechanical power, 33% is lost as heat to the radiator, and 33% is lost in the exhaust.  These are very approximate!  This means that, for a TR four cylinder engine, making 100 HP, around 250,000 btuh is dissipated in the radiator.  At a 10 degree coolant deltaT, this means that the pump must pump about 50 gpm at maximum speed.

Here endeth today's lesson, brethren

Mike



Michael Marr
Albion Technical Services
Mobile:  630-202-0065

Sent from my iPad

> On May 9, 2017, at 23:18, Paul Tegler <ptegler at verizon.net> wrote:
> 
> moving water too fast through a cooling system can be just as bad as not moving it fast enough
> (no time to xfr the heat)
> always a good bet to up the cooling capacity (fans etc)
> 
> Paul Tegler ptegler at verizon.net www.teglerizer.com
>> On 5/9/2017 9:08 PM, Michael Porter wrote:
>>> On 5/9/2017 4:52 PM, TERRY SMITH wrote:
>>> 
>>> Just curious.  Is this an engine that sits a lot, with a long time since last disassembly and boiling?  Wouldn't engines with corrosion in the water channels may need additional coaxing to cool enough?
>>> 
>>> 
>> Corrosion is an interesting question.  In a cast-iron engine, it's desirable.  Little-known fact:  the highest rate of heat exchange of known metals is off rusty cast-iron.  That likely has something to do with the large amount of surface area created by rust. Where corrosion becomes a problem is when microscopic flakes break off and combine with used-up anti-corrosion precipitates from the coolant. This forms an insulating sludge that can cause hot spots and nucleate boiling if the system pressure isn't high enough. Probably everyone tearing down a Triumph engine with lots of mileage has encountered the stuff behind the rearmost cylinder, where flow isn't high enough to keep the particulates from settling out.
>> 
>> In short, rust is okay.  The by-products of that corrosion are not.
>> 
>> 
>> Cheers.
>> 
> 
> 
> ** triumphs at autox.team.net **
> 
> Donate: http://www.team.net/donate.html
> Archive: http://www.team.net/archive
> Forums: http://www.team.net/forums
> Unsubscribe/Manage: http://autox.team.net/mailman/options/triumphs/mmarr@albiontechnical.com