Electric water pumps

[fastdruid]

Anyone fitted one?

Any idea what the standard pump flow rate is?

Druid

[yorkie_chris]

As far as I can see whatever you gained in power you'd lose in weight. You could shave the stock impeller down for reduced power loss.

[fastdruid]

To my mind what it gives more than power is consistent cooling and removal of the vulnerable waterpump.

How much does the stock pump weigh, as an example one of the electric ones I was looking at was 350g (500-980l/h).

Druid

[embee]

If it's any help, some guidelines for coolant flows

- usually aim for around 7 degC delta-T across the engine/rad at max power (best kept below 10 degC anyway)
- take heat to coolant as being equal to brake power (engine not wheel), so for a 650 let's call that 55kW in round figs
- typical SHC for 50/50 ethylene glycol/water is around 3.62 kJ/kg.K (http://www.engineeringtoolbox.com/ethylene-glycol-d_146.html)
- typical density of 50/50 mix is 1.04 kg/L at working temp
- calculated flow requirement is thus

55kW = Q kg/sec x 3.62 kJ/kg.K x 7 degC (you can say degC and K are the same thing for this purpose)

Q = 2.2 kg/sec , call it 2.1 L/sec or 125 L/min.

(As a rule of thumb you can say between 2-2.5 L/min per kW brake power).

That's a ballpark anyway.

On the open road the air side cooling of the rad isn't usually a problem at high vehicle speeds (more a problem when stationary).

I did some tests on a hearty 300kW V8 regarding coolant flow reduction potential, and you could actually reduce the flow quite a lot before critical metal temperatures (between exhaust valves usually the greatest heat flux) reached the material limits, so the empirical guidelines are fairly conservative, as proven over the years.

[fastdruid]

Interesting figures but seem an order of magnitude too high, a pump for a hayabusa for example[1] is rated at 5 gallons/min or 22.7L/min, the hayabusa is 128Kw which would give by your maths a flow rate of 256-320L/min.

Druid

[1] http://www.mbcustoms.co.uk/80720/info.php?p=7

[embee]

Not sure how that would survive.

Think about it a different way, 128kW is nearly 50 standard 2.7kW kettles. A kettle will boil a litre of water from cold (let's say 10 to 100C so delta-T = 90 degC) in about 140sec, that fits with my experience of making tea which is huge! :D

2.7kW = 2700J/sec
SHC for water is 4200 J/kg.K
so 2.7kW gives 2700/4200 degC per sec per litre of water, or around 0.6 degC/sec.

You really do only want a delta-T across the engine of about 7 degC (OK that could be pushed to 10 if you wanted), and a standard kettle will do that to a litre of water in about 11 sec.
50 kettles would do it to one litre in about 0.2 sec, or around 5 litres per sec or 300 litres/min. Or put it the way I used in my first post, around 2 L/sec per kW.
So that's the sort of ballpark you're looking at, give or take a bit. Sorry, that's the way it is.

This book (http://books.google.co.uk/books?id=eL6TBaSnd78C&pg=PA130&lpg=PA130&dq=engine+coolant+pump+flow+rates&source=bl&ots=h9Y40QYMFz&sig=bBB6w0S-_L4UTOtjS22-JUjwYOo&hl=en&ei=fStSS9cvgvjTBKaN9JgG&sa=X&oi=book_result&ct=result&resnum=7&ved=0CB0Q6AEwBjgy#v=onepage&q=engine%20coolant%20pump%20flow%20rates&f=false) (see p.130)says 2.5 - 3.0 L/min per kW for petrol engines, but in my experience that's on the high side.

23 L/min from a Hayabusa coolant pump at 128kW would give a delta-T of somewhere in the region of 80 degC, just wouldn't work. But hey, what do I know? ;)

I'm open to being proven wrong.

[Lozzo]

Anyone fitted one?

Any idea what the standard pump flow rate is?

Druid

They're fine until the fuse blows, the bike seizes and throws you over the bars as you're entering Gerards at Mallory at around 95mph - just ask my mate Danny.

Personally, I wouldn't bother because the stock water pump is well up to the job and dead reliable

[fastdruid]

Not sure how that would survive.
<snip>
23 L/min from a Hayabusa coolant pump at 128kW would give a delta-T of somewhere in the region of 80 degC, just wouldn't work. But hey, what do I know? ;)

I'm open to being proven wrong.

I've no experience of these things and I quite frankly cannot be bothered to revisit my thermodynamics to check the maths but every one who sells an electric water pump gives figures far lower than you suggest are needed.

eg from: http://www.mawsolutions.com/html/ewp_comparison_0.html

Maximum Flowrate | 80 litres/minute @ 13.5v | 115 litres/minute @ 13v
Suits Engine Sizes | Normal tune up to 3 litres | V8’s 4WD’s etc or over 400hp

From another page http://www.daviescraig.com.au/Electric_Water_Pumps__EWP-content.aspx
What is the flow of an old fashioned mechanical belt driven water pump?

Davies, Craig P/L conducted tests which showed that each capacity engine had different coolant flow rates i.e. V8 was more than a 4-Cyl. One common V8 has a flow of 24 l/min at idle (600 rpm) and so about 240 l/min at 6000 rpm. That is 240 kilograms of coolant being rammed through narrow radiator tubes every minute. It takes a lot of power to do that.

However it was shown by the CSIRO that in a standard engine, flow above 80L/min increases heat loss by very little. Once the flow has reached a rate where all of the heat being produced by an engine is being dissipated in the radiator there is no benefit from pumping faster.


Druid