Hydrodynamics

[aloha450x]

so i was looking at the radiators at the hondas in the garage. the new 450 the waterpump pulls from the bottom of both rads and the returns to the top of both rads. so in theroy it is only getting half the cooling.

now the crf250 the rads are connected at the top and bottom. the water gets pulled from the right bottom. the return is top left. since the rads are connected the return could flow straight into the right rad and thus only get half the cooling again. or the water could make it to the bottom of the left rad to the right and back to the water pump. again only getting half the cooling.. What if you closed off the top connection on the left and the bottom of the right. so water would have to come out of the motor to the top of the left rad. down to the bottom out to the top of the right rad and down to the bottom and back to the water pump. in my silly mind it works. and would be a way better system, however some smart mofo's built these things and im sure it has been thought about before.

[Pedroski]

Your way, you would have twice as much water flowing through one rad and then the other....thus you would have half the cooling efficiency. The way they are at the moment, the rads are only having to cool half the water each.

[KJ790]

Remember, since it is pulling from the bottom of both, and returning to the top of both then the fluid has to travel down one radiator before it gets picked up again. Since water is being pulled from both radiators, the rate at which the water goes through the radiator is half that of one that is only pulling from one radiator. It ends up being a wash for the most part. One way sends the water through the length of two radiators, the other way sends the water through one radiator at half the rate.

[adynes]

If you run the radiators in series, you will reduce the flow area for coolant, which will increase the resistance to flow, which will create more turbulence. Turbulence is good for heat transfer between the coolant and radiator fins. But, increased resistance will lower the coolant flow rate, which will raise the temperature at the engine outlet, which is bad. So, I think you would just have to do a test to determine if the good out-weighs the bad, but I'm going to guess it would be worse, since the engines coolant outlet is at the head, and that's the last place you want to get hot. You could actualy improve the heat transfer of the system, but at the same time have higher cylinder head temps, which is obviously bad for detonation and material wear reasons.

[aloha450x]

What if you added a boyesen high output pump as well?

[adynes]

What if you added a boyesen high output pump as well?

Sure couldn't hurt! Running the rads in series might actually be a good modification for a bike that spends a lot of time at lower speeds, like in tight trails, when the water pump is spinning slower and coolant velocity is low. You should get some of those adhesive temperature stickers, give it a try, and let us know what you find.

[aloha450x]

Will do! I'm gonna do a good study on this. I will buy a thermal temp guage. The lazer kind and record my findings. Full report soon.

[adynes]

My bike has a small hose connecting the top tanks of the two rads; I'm assuming yours does too. You'll probably want to leave that one there so the air bleeds out properly when refilling the system.

[highmarker]

There's is a correct flow rate for maximum transfer of heat, coolant flow too fast or too slow can be counter productive.

[rayivers]

With the parallel (450) connection, if you carried two tie-wraps or pieces of wire you might be able to tie off both hoses to one radiator quickly in case of damage, allowing you to make it home; could be a real advantage for woods and long-distance riders.

Ray

[adynes]

There's is a correct flow rate for maximum transfer of heat, coolant flow too fast or too slow can be counter productive.

I don't think increasing the coolant flow rate will ever be counter productive from a heat transfer point of view. There will be a curve of diminishing returns, and a higher flow pump will require more power to turn (hydraulic power is proportional to: Flow x delta-Pressure). If you ignore side effects like turbulence, rate of heat transfer is independant of coolant flow rate. What it will change, though, is the temperature difference from engine inlet to outlet, and radiator inlet to outlet. A faster coolant flow rate will lower the temperature delta, but the average temp of the system will be the same.

[adynes]

With the parallel (450) connection, if you carried two tie-wraps or pieces of wire you might be able to tie off both hoses to one radiator quickly in case of damage, allowing you to make it home; could be a real advantage for woods and long-distance riders.

Ray

Good point.

Edit: I guess you could potentially do the same thing with a series system by rerouting a hose or two, as long as you carried a screw driver and had enough hose length

[KJ790]

If you run the radiators in series then the water has to get pumped from the head to the top of one radiator, then again from the bottom of one radiator to the top of the other. If you run them as a parallel circuit then the water just has to be pumped to the top of one radiator and gravity does the rest. The slower the water moves through a radiator the more heat will transfer. running hot water slowly through one radiator transfers more heat than running it through two radiators twice as fast. I don't know why you want to run it through the radiator faster, you will just lose cooling capacity and increase the power loss due to the waterpump.

[adynes]

If you run the radiators in series then the water has to get pumped from the head to the top of one radiator, then again from the bottom of one radiator to the top of the other. If you run them as a parallel circuit then the water just has to be pumped to the top of one radiator and gravity does the rest. The slower the water moves through a radiator the more heat will transfer. running hot water slowly through one radiator transfers more heat than running it through two radiators twice as fast. I don't know why you want to run it through the radiator faster, you will just lose cooling capacity and increase the power loss due to the waterpump.

No, this is a common misconception.

With a higher coolant flow rate, its true that each unit of coolant will conduct less heat to the rads because it spends less time there. But, there are more units of coolant flowing through per unit time, so it ends up equal.

[KJ790]

No, this is a common misconception.

With a higher coolant flow rate, its true that each unit of coolant will conduct less heat to the rads because it spends less time there. But, there are more units of coolant flowing through per unit time, so it ends up equal.

There are not more units of coolant flowing through the radiators per unit of time. Running in parallel the waterpump pumps the same amount of water through the radiators, half goes through one radiator, half goes through the other. The water travels through each radiator at half the speed, so it is a wash in the end. The same amount of water goes through the waterpump per unit of time.

It's been a while since I took a heat transfer course, but if I remember correctly, the larger the difference in temperatures (water to ambient air) in a heat exchanger, the faster the heat transfer rate. The rate of flow through the heat exchanger doesn't really have an effect (assuming that the heat exchanger length is proportional to the velocity of the fluid through it so that fluid is in the heat exchanger the same amount of time).

If you run them in series, you have hot water going in one radiator, when it gets to the bottom of that radiator it will be cooled a fair amount b/c the temperature difference was large between the water and the outside air. Then this warm water goes into the second radiator, since all of the water in this radiator is cooler than in the first radiator (closer to ambient temperature), the heat transfer ability of this radiator is less than the first one. You will end up with one radiator running hotter than the other, and thus one radiator transferring more heat than the other.

The cooling capacity is the same either way, but you are adding unnecessary pumping losses to the system, which will take power from the engine to overcome, thus costing you a bit of power, or making the engine work harder to put out the same amount of power (thus building more heat). In the end there is no benefit to running the radiators in series.

If you really wanted to get rid of some heat, figure out a design for a counter flow heat exchanger instead of a cross flow, they are the most efficient type of heat exchanger. Isn't really feasible with a bike do to dimensional restrictions. What makes you think that the bike needs to be cooled more anyways? If it is not overheating then there is no issue. These new age four strokes make more power the hotter they run. I've seen this on a dyno many times, the hotter you can get the engine to run without melting parts, the more power it puts out. This is why James Stewart blocked off half of his left radiator with tape all season last year.

[adynes]

No, this is a common misconception.

With a higher coolant flow rate, its true that each unit of coolant will conduct less heat to the rads because it spends less time there. But, there are more units of coolant flowing through per unit time, so it ends up equal.

There are not more units of coolant flowing through the radiators per unit of time.

I was discussing high vs. low coolant flow rate.

I was replying to the part where you said:

The slower the water moves through a radiator the more heat will transfer. running hot water slowly through one radiator transfers more heat than running it through two radiators twice as fast. I don't know why you want to run it through the radiator faster, you will just lose cooling capacity and increase the power loss due to the waterpump.

This is not correct. It goes against everything I've been taught/read. Also, there are many cooling discussions on the internet with engineers stating the same things I have posted.

[KJ790]

I was discussing high vs. low coolant flow rate.

I was replying to the part where you said:

The slower the water moves through a radiator the more heat will transfer. running hot water slowly through one radiator transfers more heat than running it through two radiators twice as fast. I don't know why you want to run it through the radiator faster, you will just lose cooling capacity and increase the power loss due to the waterpump.

This is not correct. It goes against everything I've been taught/read. Also, there are many cooling discussions on the internet with engineers stating the same things I have posted.

Yes, I know what you have read and have been taught, I am also a mechanical engineer. I wasn't clear in that previous post, and reading it now I see how poorly (and incorrectly) worded it is. I am agreeing with you that the cooling rate is the same in the end, it is not dependent on the velocity of the flow through the radiator. That is what I am getting at, why try and speed up the velocity of the flow through the radiator? All that will do is make the waterpump work harder, using up more power, making the engine work harder, and thus building more heat in the engine. In that aspect you will lose out. In reality the amount of power used up by the waterpump is pretty small, and the extra heat build up in the engine would be very small as well, but on a small engine minimizing losses, no matter how small, is the name of the game when it comes to performance. However, there is no advantage to running the radiators in series.

[adynes]

Running in parallel the waterpump pumps the same amount of water through the radiators, half goes through one radiator, half goes through the other. The water travels through each radiator at half the speed... The same amount of water goes through the waterpump per unit of time.

I agree.

It's been a while since I took a heat transfer course, but if I remember correctly, the larger the difference in temperatures (water to ambient air) in a heat exchanger, the faster the heat transfer rate. The rate of flow through the heat exchanger doesn't really have an effect...

Exactly

The cooling capacity is the same either way, but you are adding unnecessary pumping losses to the system, which will take power from the engine to overcome, thus costing you a bit of power, or making the engine work harder to put out the same amount of power (thus building more heat). In the end there is no benefit to running the radiators in series.

So we are in agreement, exept you are neglecting the fact that an increased flow rate (through each radiator in series) can increase turbulance and therefore heat transfer.

What makes you think that the bike needs to be cooled more anyways? If it is not overheating then there is no issue.

I don't. I'm just replying to the original poster.

[adynes]

OK, sounds like we are in agreement, except:

However, there is no advantage to running the radiators in series.

You can't make this claim without performing a test. Maybe it will help, maybe it will hurt. My guess is that a series setup might help for lower engine speeds (lower coolant velocity/turbulence), but maybe not.

and:

...it (cooling rate) is not dependent on the velocity of the flow through the radiator.

velocity increases turbulence which increases heat transfer. There will be a curve of diminishing returns. I'm not making a statement that there is a problem with a certain bike, just stating the trend.

[adynes]

These new age four strokes make more power the hotter they run. I've seen this on a dyno many times, the hotter you can get the engine to run without melting parts, the more power it puts out. This is why James Stewart blocked off half of his left radiator with tape all season last year.

I agree. All heat engines have the potential to make more power with a larger temperature differential between combustion and ambient. This is called Carnot Efficiency, and is the idealized theoretical maximum efficiency a given engine could possibly reach, but never will due to all the different losses throughout the system.