Calculate water flow to control temperature 1

[Didier0001]

Hi,

I'm new here, let me introduce myself; I have a degree in racing engine design, and trying to develop a competitive engine to expand my knowledge on designing..

I came up to the point where I want to optimise the cooling circuit to reduce pumping losses from the water pump.

It has been a couple of years now since I had my thermodynamics lectures, I remember the processes but kinda forgot how to calculate the heat exchange processes.. (shame on me!)

So my question for you guys is if you can push me in the right direction?

The calculations I will have to do is the heat conducted from the combustion chamber through the X mm of aluminium towards the water jacket.

And the heat extracted by the waterflow in that waterjacket and the crossectional area of this passage (flow rate).

Anyone who can (is willing to) push me in the right direction? It would be much appreciated!!

Regards

 

[RossABQ]

I think you may be over-thinking it. Heat rejected by the combustion chamber (or whatever part) raises the temperature of the water. You need to fix the temperature rise. You will have an entering water temperature at the point in question, which will vary depending where in the engine the part is. Heat transferred = 500 x gpm x delta-T. (if you want to correct for ethylene or propylene glycol it isn't exactly 500) Solve for gpm, then determine the passage size based on desired delta-P.

 

[LionelHutz]

Most engines have quite a complex shape for the coolant passages. Then, each area of these coolant passages could be at a slightly different temp and have slightly different amounts of heat input (from the combustion chambers). I'd think you would have to create a full 3-D thermal model to do any reasonable calculation.

It'd likely be simpler to calculate the heat output of the engine by measuring coolant flow and temperature change. Then determine how much you can slow the pump down without overheating. In other words, experiment with the cooling.

 

[patprimmer]

One of my cars has zero losses to drive the water pump or any cooling system component ;-)

Oh it was made in 1912 and is thermo cycle no fan. It does have an exceptionally big radiator for it's power output.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[patprimmer]

Seriously, Lionel nailed it.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[patprimmer]

Oh

Also, water passages are normally oversized with the flow at different points being controlled mostly by the sizes of the various holes in the head gasket.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[Didier0001]

Hi everyone,

As I already told, I'm developing the engine, meaning, I am doing it in CAD.
All the components are pretty much designed, just want to know how to calculate all the areas of the cooling circuit.

@ Lionel

How do I make the thermal model then? Is there a software package that I can use to analyse the heat?
Or do I have to calculate everything?


@ patprimmer

Which car is this then? Never heared of something like that, all engines have losses don't they? Otherwise people would have stopped looking in to efficiency?


Hope someone can help me now. I am currently reading about the processes in "Internal Combustion Engine Fundamentals, John B Heywood" anyone familiar with this book?

 

[RossABQ]

Thermosiphon cooling was pretty common, Henry Ford must have hated pumps. Model T's had no water pump and no fuel pump. But then they overheated a lot and couldn't go up steep hills, either...

 

[Batista230]

If you doing it on CAD then you should be able to see what size it is
What bore are you working with?

 

[Superbeaver]

Hello
I think it will be quite difficult to optimize water flow from scratch. The problem is to know how much heat can each of your engine components can take. An then it would be a compromise between the size of the coolant chambers and the parts. For example, if you start a cylinder head without coolant chambers around valves, you will probably not cool them enough, but if you design big coolant chambers it will reduce the space available which can force you to reduce the size of your valves and induction system.
If you really aim at performance, I think you should design your parts for it and then see what space is remaining for coolant, and hope that it will still be makable.
One other question you have to ask yourself is at which temperature you want to run your engine. Knowing that at low temperature, you might get more power but you will need a bigger cooling system (radiator and pump) and at high temperature, you might have some reliability issues.

At the end, I think that if you really want to optimize your coolant circuit, you will need an iterative approach, because there are a lot of different factors. And you can't simulate them all before testing, unless you are a big company with a big "calculation/simulation" department (and even with that we still need this iterative work).

 

[EHudson]

RossABQ sent you on the right track.

Using the old 1/3, 1/3, 1/3 rule figure a third of the energy goes out the exhaust, a third into the cooling and a third is useful output. So, estimate the cooling load is roughly equal to the engine output. For every hp you need get rid of ~2545 BTU. Figure on a 15F temperature drop across the engine and you can determine how much water you need to introduce for steady state cooling.

Keep the velocity down below 20fps in the small passages and above 2fps in the larger areas and you'll have a good start. Refine with CFD (colorful fluid dynamics) if possible to improve uniformity. Failing that, start with small holes for passageways between castings, such as the block to head interface, and drill them out to balance the cylinder to cylinder temps using thermocouples.

 

[yoshimitsuspeed]

I think EHudsen as well as many others have very good points. My experience is more rooted in real world application than in the theory or math.
One thing I know is I have seen 160 hp engines taken over 600 without overheating problems caused by the engine. You will probably need a bigger radiator and may need bigger lines outside the motor.
There is also no reason to have extra material in the motor just adding weight so I would look at it more from the perspective of making as strong as it needs to be and remove all excess material for oil and water cooling.
I would also look at some engines similar to what you are designing and try to see how they did it.

 

[Didier0001]

Colorful Fluid Dynamics? Don't you mean Computational Fluid Dynamics?

Thank you for all your help, I was able to calculate everything I needed. Engine is 95% designed, so all that needs to happen now is make it, and test it.. Let's see how close the calculations has brougth me to what I want..

You guys really had some helpfull information!

 

[patprimmer]

Maybe his comment was tongue in cheek as Computational Fluid Dynamics displays a lot of bright contrasting colour.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[DanEE]

Quite a lot of activity going on in patents on new approaches to engine cooling. Might start with a patent search.

btw Google has a new patent search engine available in beta mode, easiest to use that I've found yet.

Search on "engine electric water pump" and you will find quite a bit on what others are doing with variable flow rate pumps for cooling.

Perhaps there have been some papers published in tech journals of the current generation Prius engine.

It now uses an electrically driven water pump unit to further reduce drive and pump losses. There are now no belt driven accessories at all on the engine.