Coolant Flow for a 15Hp 2 Stroke

[EMS21]

A mate of mine has a Small 260cc 15Hp Mercury Outboard 2 stroke power head thats used on a small Experimental UAV Helicopter, The cooling system has been modified to run in a closed loop system. I believe the Radiator is big enough for the job but it does overheat to some degree as the load increase's.
If I compare the Radiator from a water cooled motocross bike that has similar or higher HP etc this Radiator has plenty of size, what concerns me is the water pump and its capacity to pump, this pump btw is the old rubber Impeller from the bottom of the Leg thats been adapted into a new housing and driven at a ratio of 1:1 of the crank.

The Radiator has a high capacity Brushless Fan mounted on it, my thoughts here are that the Fan Shroud needs to be moved away from the core as its to close in my opinion and possibly drawing Air from the center rather than all over the core .. current Draw is around 7 Amps.

As for the modified pump I did a quick and dirty check of it's capacity using an electric drill running at 2800 Rpm it was only around 7 LPM free flow, this engine has been setup to run at approx 3600 Rpm (max Torque) so I calculated that this pump could be pumping around 9 LPM at those Rpm's but I doubt it some how due to a small hose in the system plus other loss's, btw max Pressure is only around 7 Psi.

I read in another thread that water flow should have an ideal flow rate of 2 - 5 fps, if so then I believe that a higher water flow is needed if the Radiator hose connections are anything to go by.

I have not seen the complexity of the cooling system passages in the bottom of the Block where the leg would have origionally bolted on but from what I am told one of these passages is only around ~10mm in diameter and this may need to be increased to cater for the increased flow of a bigger pump.

At the moment we are waiting on some bearing's for a modified tail rotor to arrive... after its running again I will Data Log some temps at the Radiator in and out and at the thermostat housing

Radiator: Aluminium (new)
Brand / Model unknown (24F23 painted on the side)
Dim's : H 180mm x W 220mm x T 30mm
22 FPI, Tube Spacing 10mm , 2 Rows inline .

I would appreciate any help or info / formula's that would suggest a flow rate and Radiator size's to use for a 2 stroke ? or even a rule of thumb for water flow verse's Hp if one exist ?

TIA
Thomas

 

[kenre]

One small but important thing you may have overlooked is that a radiator only has a few degrees drop in temperature.

An outboard has a few billion litres of cooling water
availabe, so it is always recieving nice cool water, and the engine water pump, hoses and water passages are designed aound this fact. Cooler water means less is needed as it can absorb much more heat before boiling.

With your setup you are pumping water that has been heated considerably higher than lake/river water back into the engine, The volume and flow which would not be high enough to remove the nessesary heat.


Hope this helps.

Ken

 

[EMS21]

Yes I agree with you..
having worked with plenty of overheating Caterpillar/ Detroit Engines and the specs for cooling an engine seem to hold reasonably true for most of them with regards to heat rejection and temp drops over Heat exchangers and Radiators. I spose I am looking to apply some of these rules to cooling this 15Hp engine if thats a reasonable assumption ? so if the water flow & radiator size is correctly matched to the amount of heat thats needing to be rejected then I think a 6-8*C drop over the Radiator would be ok, without overheating nor overcooling the block.

Not overlooked but Granted there is an endless supply of cool ~25*C or so water to cool that outboard eng and that can absorb more BTU/Lb than say water at 80-85*C and trying to maintain a safe operating temp of approx 90*C ..

with a bit of help thats what I need to figure out.

Regards
Thomas

 

[patprimmer]

The position of the water pickup might act to increase water pressure at speed.

Rubber impeller pumps take a lot more power to drive than a conventional centrifugal pump.

You can get better efficiency from a closed system by increasing the coolant temperature with the use of pressure and by adding ethylene glycol to the system.

The glycol may be detrimental to the rubber impeller.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Warpspeed]

Pat is definitely on the right track with pressure.

What is not generally realised, is that most of the heat is liberated around the cylinder head and exhaust port, perhaps only a few square inches per cylinder, and the local temperatures can get rather higher than the bulk water temperature exiting the cylinder head.

Two things are needed, sufficiently high water flow velocity, and a high static water pressure to prevent steam bubbles from forming in the immediate hottest areas. Once steam bubbles form they can grow rapidly, driving the bulk cooling water away from the hot spot. That is what cracks cylinder heads.

Normally there is some static pressure controlled by the radiator cap, and some additional dynamic pressure created by the water pump. The engine driven pump pressurises the whole engine block against a flow restrictor in the cylinder head water outlet, usually at the thermostat.

A flea powered water pump is probably not going to generate sufficient dynamic pressure, especially in a high constant load application. I believe a much more robust water pump will be required to prevent cylinder head hot spots from forming, regardless of radiator size.

Bulk water temperature is not always a good guide to what is going on inside the cylinder head at critical locations, so be careful.

 

[EHudson]

The Merc's original thermostat was a 140 degree F unit. Outboards are limited to this low coolant temperature to keep deposits from building up in the cooling passages. So figure a normal temperature rise of at least 70 degrees F.

You will need to proportion your flow based on your expected temperature drop across the radiator. The floppy vane pump of the Mercury tolerates sand, dirt, a bit of dry running (sometimes) and primes easily but does take more drive power than an loose fitting impeller because of the friction of the tight fitting parts.

Also bear in mind that the Mercury prop did not load the engine down to 3600 rpm and thus would be running at close to 6000 rpm at full load with a corresponding increase in pump speed.

 

[EMS21]

The load on this Engine has been geared for approx 3600 Rpm it can easily increase in Rpm if a higher throttle setting is desired its not overloaded.
I pulled the thermostat out just to check to see what it was opening at, it has "120" stamped in it, 120F perhaps ? this one opens at 55*C or 131*F ... but I can change that latter.

looking at a Moto X engine the cooling flow is not that complex IMHO, basicaly in the bottom and out the Top of the head kinda deal,this Merc has a bunch of passages running all over it, is this just to try to pre-heat the cool water so as to obtain a more even heat distribution through out the Block ?

It will be another week or so before I get to remove this Engine and check out all those passages, but what if a more traditional approach was employed like the MotoX engine ?
There is a small restriction ~8mm on the thermostat housing cover, these are all the things I believe should be enlarged to handle a Higher flow rate for a closed loop system.... and without seeing inside yet I'll bet there are more ?
When I see all the small passages and trying to use a Flea powered pump I'm not surprised that this Engine cant handle the heat load.

How does this sound ...
Looking at a similar sized Hp Engine (slightly bigger) the Radiator on this bike is around the same size (frontal Area / thickness) as the one on the Helicopter now, and the Moto X (YZ125) Radiator hose has an ID of around 14mm and with flow rates of 2-5 fps or 0.6 - 1.52 mtr/sec, so for the sake of this I will use 1.52mtr/sec (average velocity of the Fluid) so to get a laminar flow this works out to be ~14LPM, is this a reasonable flow rate for this engine ?

For right now it would be nice to control the over heating issue and then try to make the engine Run at

a safe controlled temp of around 85*C ... any idea what is the operating temp at full load ?


Thanks
Thomas

 

[EHudson]

I think a bottom - in, top - out circulation system should work fine. Be mindful that one of the requirements of an outboard cooling system is that it must drain itself to prevent freeze up in cold weather. This may be why some of the passages are not as straight forward as you think they should be.

14L per minute should be about right:

I'd guess that peak torque would be about 20% higher than at max. h.p. So if max hp at 6000 is 15, the torque should be about 13.1 ft-lb. Let's guess peak torque is 15.8 ft-lb.

Then at 3600 with 15.8 ft -lbs of torque, the hp output would be 10.8.

Assuming roughly a third of the energy goes into the crank, a third out the exhaust, a third of the energy, again about 10.8 hp, goes into the engine coolant.

This means we need to get rid of about 458 BTUs per minute with 14L, or 30.8 lbm of water. The temperature change will be about 15 degrees (F); a reasonable drop in temperature for a radiator, I think.

 

[EMS21]

There's a bit of a network of passages on the RHS of the block (opposite side to the Carby), there are 2 plates that I think are for the Exhaust and the other has water flowing around it, I'm not sure of the function but perhaps its to help get a more even operating Temp throughout the system.

Your Calculations agree with mine like you say assuming 1/3 for the cooling system etc 458BTU/min is correct.
Btw what formula did you use to calculate the heat rise of 15*F with a water flow of 14 LPM ?

My thoughts are to modify these passages in order to get better flow within the system also to get the Bottom in Top Out idea working with minimal restriction's working better, so a pressurised system of ~15Psi should keep coolant against any possible localised Hot Spots mostly around the Exhaust port I'd suspect.

I'm not worried about Freezing in the Block for the moment
and if it should freeze here in Jakarta I hate to think about all the other poor soles on the planet who will be long gone.

btw I have also ordered a small 1-25 Lpm flow meter with a pulsed output and I'll use this to to Data Log the Flow, Temp, Eng Rpm and see whats going on in the real world.

Thanks for your time and input
Thomas

 

[EHudson]

14L per minute and 458 BTUs/minute are numbers we agree on.

I changed the 14L to 30.8 lbm of water using a calculator to change liters to 3.8 gallons and 8.336 lbm/gal to get a weight of 30.8 lbm of water, (I happen to prefer to work in God's units). Since a BTU will raise a pound of water 1 degree F, 458 BTUs raise 30.8 lbm of water about 15 degrees. (The minutes cancel)