Cooling opening size?

[Sail 'n Soar]

Does anyone have a good experience base or rule of thumb that can be used to determine motor and battery cooling air intake sizes?

Seems there should be a correlation between flight speed, motor power and efficiency and battery internal resistance losses with cooling air inlet minimum required area. Several of the sport scale subjects I'm presently considering have relatively small inlets. I would like to keep things as close as possible.

[vintage1]

The difference between no airflow, and even a little bit, is dramatic. Ther was a long discussion about all this, and te consensus as far as I can remeber was

- exit LARGER than inlet

- Put inlet where pressure is highest (towards front and bottom) and exhaust where its lowest (top and rear, especially over the wings).

Make sure that the path the air takes goes over the stuff to be cooled :D

[Sparky Paul]

Having the large cooling volume commensurate with an IC heat source probably isn't needed for electrics, as long as there's some airflow over the motor, ESC and battery.
As Vintage recommends, high pressure in, low pressure out.

[Salto]

For good airflow, duct location is just as important as duct size. There's some more info here http://www.rcgroups.com/forums/showthread.php?t=257604&highlight=cooling+air+inlet with a link to even more info in another post.

Sparky, that's an interesting comparison. A typical IC engine in a model is probably only about 15% efficient, so for 200 Watts output there is about 1100 Watts of waste heat to get rid of. On the other hand, a good brushless electric motor will be about 80% efficient so for 200 Watts out there is only about 50 Watts of waste heat to get rid of. So cooling ducts in electrics only need to carry 5% of the heat that those in IC models do. This ignores heat generated in the battery, but I think this is minor compared to motor heat.

Yet another good reason not to go near a glo engine ever again!

Graham.

[Sail 'n Soar]

For good airflow, duct location is just as important as duct size. There's some more info here http://www.rcgroups.com/forums/showthread.php?t=257604&highlight=cooling+air+inlet with a link to even more info in another post.

Sparky, that's an interesting comparison. A typical IC engine in a model is probably only about 15% efficient, so for 200 Watts output there is about 1100 Watts of waste heat to get rid of. On the other hand, a good brushless electric motor will be about 80% efficient so for 200 Watts out there is only about 50 Watts of waste heat to get rid of. So cooling ducts in electrics only need to carry 5% of the heat that those in IC models do. This ignores heat generated in the battery, but I think this is minor compared to motor heat.

Yet another good reason not to go near a glo engine ever again!

Graham.

But not the best comparison. Heat transfer is a function of delta T, among other things. We don't want our electric's to get beyond warm, let alone flesh-burning hot. Thus, our delta T's are an order of magnitude larger - measured in 10's of degrees, rather than 100's.

In terms of duct locations, the areas I'm considering are ram air sources - perpendicular to and directly behind the prop, for example. But there still has to be some measure, such as inlet equivalent area times prop speed or flight speed per dissipated watt.

[Sail 'n Soar]

Oops, we want our to be an order of magnitude SMALLER, not larger. :o

[vintage1]

Well, if you want to do the math, you can work out for a given mass of air per second given its specfic heat how hot it will come out to get rid of X watts, I suppose.

But its not a big dewal: Most components are rated free air only: All you are doing is making sure the air is free, and not constrained, and gets emptied out now and again.

How fast does the air come out of your heatgun, and what what temperature and diameter and how many watts is the gun?

That gives you an idea of watt (:D) sort ofairflow is needed for a given power dissipation.

Never mind, nexts years motor will come with radial mounting, a big fat finned cylinder and a hor air pipe at the back, so as to make glo convesrions easier :D

[markdrela]

The mass flow going through the cooling ducting depends directly on the pressure difference between the inlet and outlet, and on the inlet and outlet areas. The usual goal is to obtain a given cooling mass flow, with a minimum of total pressure loss which causes added drag. The best (lowest drag) installation has the following features:

1) A forward-facing inlet which receives the full freestream stagnation pressure air.
2) An aft-facing outlet "nozzle" which discharges the air cleanly.
3) The inlet area is 1.5-2.0x larger than the outlet area.
4) The size of the outlet area controls the mass flow.

The fact that the combination of a relatively large inlet area and small exit area gives the least drag is counterintuitive, but true. The reason this works is that causes the air to flow into the interior at a relatively low velocity. Since mixing losses (= total pressure losses) for a given mass flow are proportional to velocity^2, the low inflow velocity keeps the drag penalty small. In contrast, a small inlet area will produce a high-velocity jet squirting into the interior, which will mix out and ultimately produce drag.

[jlk]

I would never hold my own in a debate with Mark as he is the expert. I will do the "Oprah" thing and present anecdotal evidence against the expert. Over my several decades of modeling with IC engines the exact opposite for exit and entrance proportion was always the rule for good cooling (not sure about the drag). The same was applied to electric cooling in almost every design I have flown. In my full size flying we also find that cowl inlets are about 1/2 the size of outlets and in the newest models, Cirrus and others the cowl inlets are even smaller and the outlet large. Cowl flaps on full size planes increase the exit area to increase cooling when velocity is low. NACA duct specs call for larger outlet areas than their inlet. There also is the consideration that on a sleek model the larger inlet area may not be available in a high pressure area whereas a larger outlet in a low pressure area usually is.

I don't disput your post Mark, I would like more education about these cases.

Thanks,

Jordan

[Salto]

Jordan,

I think your observations are entirely consistent with Mark's statements.

He described what's needed for the lowest drag installation. But lowest drag is not always the primary requirement. In full size GA aircraft and on many models the requirement is to get maximum cooling airflow, and the drag penalty is not so important. So making both duct areas large is a good approach. And as you say, for structural/styling/space reasons the inlet size is sometimes limited, so the next best thing to do is increase the outlet area.

Graham.

[Sail 'n Soar]

Thanks, all, for the inputs. Most recently I am focusing on either a P-51 AorB or an A-36 sport scale, in which case the inlet would be the scale carburetor inlet, and the exit, the scale radiator exit appropriately opened for the ~1.5-2x. I will most likely go with the A-36 because the carburetor inlet ducting was roughly double the size of the Allison-powered P-51's to allow for an air cleaner and bypass port, since they flew down low in the dust.

All in all, the locations are close to ideal in terms of your various inputs. My main concern is that they are large enough, since they are smaller than what I have used in the past.

Anyone wish to relate their practical experience relative to 200W and 500W sized brusless electrics?

[vintage1]

All in all, the locations are close to ideal in terms of your various inputs. My main concern is that they are large enough, since they are smaller than what I have used in the past.



That's because the full size designers knew what they were doing by those days.

At say 600W amnd 0% efficiency you need worst case to dump 180W.

A couple of light bulbs worth,but not a hairdryer worth :D

I can;t say exactl,but my expereince with forced cooling on semiconductoirs suggets that teh real difference is between any airflow at all, and no airflow!

Theefficiency ofheat dissipation does not get markedly better as flow volume is increased, because in tehend, you start to get better heat transfer to the air than from the hot spots inside the motor to its casing. Beyond that you cannot go.

In semiconductor terms, limited to say a 175C internal temperature, a typical free air rating would be 5W, and on an infinite heatsink (case held rigidly to 25C) the device might be rated at say 100W. Typically it would to 25-30W on a simple heatsink and only about 60W on a blown one.

This works well on inrunners where the windings are outside near the case surface, but is of course impractical on ourunners with the case spinning. Thse tend to suck air in anyay and sling it out through the perpihery of the back of the bell.

Remember that a typical glo engine is probably generateing 3-5 times as much heat as its delievring: A good brushless should be delivering twice its heating AT LEAST.

So the electricneeds between 6 and 15 times less colling for a similar utput and temperature rise.

Of course they need to run cooler than glo, so probably 3-5 is a better real estimate of cooling differential.

There is o way to real esotamte without a lot of complex maths, how hot a given motor will get. Its probably best to get spme of those tempereature strips that are used to mintor temeperature on electronic equipment, and fly with them stick to the motor to actually see how hot its getting.

Finally of course, if you spend the money on e.g. a Hacker, you get a cooler runniing motor anyway.

Ort a cobalt, which can run hotter without damage.



If you are concerened about airflow, a finned heatsiink may be worth consuidering.

[Sail 'n Soar]

Finally of course, if you spend the money on e.g. a Hacker, you get a cooler runniing motor anyway

Vintage1,

Thanks for the expanded thought piece. Right now I'm flying a couple of AstroFlight 020 brushless, which aren't that bad either. Motors and batteries are just slightly warm to the touch post flight. The thing I rember from my distant past is the sound of my neighbor's nicad battery sizzling from boiling its internal juices after each flight - from back when the AstroFlight 10 ferrite motor was news.

Gerry