By continuing to use this site, you agree to our use of cookies. Find out more
Forum sponsored by:
Forum sponsored by CML

Use Venturi effect for cooling air?

All Topics | Latest Posts

Search for:  in Thread Title in  
Nigel R04/08/2020 14:29:19
avatar
4082 forum posts
694 photos

"But I wonder if using the simpler shape would give just as good a result at model scales."

Probably give acceptably good result.

Bear in mind, lots of full size just have a bit of a flap, e.g. sea fury exhaust & cowl outlets.

sea fury from behind

Denis Watkins04/08/2020 14:49:12
4633 forum posts
129 photos

Good idea on aircraft

nx5330_lancair_03.jpg

Richard Clark 205/08/2020 03:30:44
426 forum posts
Posted by Nigel Sharp on 04/08/2020 12:31:22:

Doug,

Yes I'm on that same page I was just after you posted.

I'm trying to get as near to finished as possible on this build this week. My furlough ended last friday and I am on enforced holiday this week. Back to work next.

I can leave the rear lower fuselage uncovered for the time being and knock up a proper duct using the NACA proportions in fusion 360 to give me some patterns to cut out and fit.

But I wonder if using the simpler shape would give just as good a result at model scales.

N

Nigel,

I just cut a rectangular hole similar to yours and epoxied in (on the inside so the hole makes a neat rim) a piece of aluminium mesh from a car accessory shop. It's normally used to reinforce 'P38' type filler body repairs, such as holes caused by rust.

I carefully painted it the fuselage colour and it looks fine.

However:

Unlike an IC engine an electric motor and the ESC should run no more than 'warm' as the efficiency is greatly reduced if it is hot. You are not trying to make a room heater.

And carrying that heat away with a strong airflow merely 'disguises' the heat caused inefficiency as the motor is still generating the heat you are carrying away.

So if the motor and/or the  ESC  is getting hot enough to need a lot of cooling you should look at the suitability of the motor, the prop size, the number of cells relevant to  the prop size, and so on rather than trying to improve the cooling.

Edited By Richard Clark 2 on 05/08/2020 03:37:56

Edited By Richard Clark 2 on 05/08/2020 03:38:42

Denis Watkins05/08/2020 07:50:56
4633 forum posts
129 photos

It is mainly the ESC that needs cooling, as in its design is a heatsink, put there to avoid melting the soldered joints.

The ESC runs very hot as it operates controlling variable current during the flight.

It does get burny hot !!!

Richard Clark 205/08/2020 09:44:59
426 forum posts
Posted by Denis Watkins on 05/08/2020 07:50:56:

It is mainly the ESC that needs cooling, as in its design is a heatsink, put there to avoid melting the soldered joints.

The ESC runs very hot as it operates controlling variable current during the flight.

It does get burny hot !!!

It's a poor ESC that gets that hot, even with no airflow at all.

As I said we aren't making room heaters and we don't want cooker hotplates either.

In our usage all heat is wasted energy, Even if you take it away by airflow and/or heatsinks.

Nigel R05/08/2020 11:10:17
avatar
4082 forum posts
694 photos
Posted by Richard Clark 2 on 05/08/2020 03:30:44:

Unlike an IC engine an electric motor and the ESC should run no more than 'warm' as the efficiency is greatly reduced if it is hot. You are not trying to make a room heater.

And carrying that heat away with a strong airflow merely 'disguises' the heat caused inefficiency as the motor is still generating the heat you are carrying away.

So if the motor and/or the ESC is getting hot enough to need a lot of cooling you should look at the suitability of the motor, the prop size, the number of cells relevant to the prop size, and so on rather than trying to improve the cooling.

I'll keep an open mind.

I have several questions:

Can you explain how an ESC efficiency drops with temperature?

Can you explain how a strong airflow "disguises" this proposed inefficiency?

Can you explain why "number of cells relevant to the prop size" is a measurement worth investigating?

Can you explain why the power train design should be assumed to be at fault?

Bob Cotsford05/08/2020 11:42:37
avatar
8746 forum posts
489 photos

I've got to agree with Richard to a point on this occasion. An ESC that gets hot is reducing the efficiency of the power train - heat losses are energy that's not going through the prop! ESC's are pretty close to digital devices, the FETs are either off or on so voltage drops should be minimal, nearly all BECs are switched mode these days which again means there should be minimal cause for heat buildup. If you need a lot of cooling then you are wasting a lot of energy!

Looking at a number of my electric models the inlet area is minimal, which means the airflow over the electric components is low velocity and low volume. Nothing seems to be complaining so far but I try to run with a reasonable overhead on all components.

Nigel Sharp05/08/2020 12:56:11
39 forum posts
166 photos

I agree with efficiency and heat dissipation etc.

Being new to electic setup (because I have for 40 years flown Slope soarers) I am aware that the ESC can get hot. Also that batteries get warm, and motors when pushed generate heat. All the heat adds up to inefficient. Yes.

I've had best part of 6 years away from the hobby. My life turned upside down and I'm picking up the pieces. Part of that has been Shielded lockdown and with mucho time now on my hand I've come back to my hobby.

I've modified a 1972 biplane design its a plan build and has been most enjoyable I mostly used to build old style balsa kits (I dropped away because every time I go in a model shop the kits are ARTF clones and that just doesn't float my boat at all.)

I've decided to try my hand at flying power to bring something new. But I hate sticky glow fuel residue smell and general mess involved with IC. Hence 40 years at Gliding. I was the same with Boats I raced model yachts for around 12 years so I use air pressure all the time for lift or motive power.

Now.... I believe I'll get a nice RAM air effect through the cowl.... its going to come in at at least 50mph, swirl around and through the motor (also augmented by a vented spinner).

Then the air will pass through the firewall via a constriction into the battery bay. The ESC is on the floor of the bay immediately in that stream of constricted air. Hence it should be given cooling. there is minimum around 6mm clearance all around the ESC.

Above the ESC is a shelf open in the middle that the battery will be held to with velcro strap. There is lots of space around the battery. So air should form eddies and help keep that cool.

Behind the battery is an open bulkead through which the battery pokes I've placed a small sub bulkhead over the U/Carrage plate to stop the battery sliding into the main over wing area. Above that bulkhead it is all open except for the RX and two servos.

Then the air should be vented out of the lower rear decking.

inside_airflow.jpg

My question was about augmenting SUCTION at that vent.

My Bipe currently weighs 4 lbs 14 oz less battery. with battery its about 5 lb 12oz..

I spoke at some length with George at 4 max so I'm sure its going to be ok

The motor is a 4-max 3547 700kv

The ESC is 60 Amp

Battery is 4s 3700Ah 40C

Prop is 13 X 6.5

Ecalc gives me 745 Watts on full charge and heat is warm with "EXCELLENT" cooling.

I dont envisage doing prop hanging or lots of vertical work but I do want to have as much cool air as possible because that to me is just sensible.

So.... I can try reverse NACA Duct it looks good.

A simple Duct ( It might be just as good as NACA since that was designed for inlet)

An External Cowl rear facing ( a bit draggy but who cares its a sport bipe not a racer)

A combination of all of the above...

and to add another into to the mix Vortex generators just upstream of the inverted duct.

How will I know which is best?

Well I can make a wind tunnel build a mockup of the fuselage, get a manometer and do some scientific analysis. Which I might do simply because I bought some black oracover EasyCoat and its pant's its stopped my build.

I just started to cover the battery hatch - worse than solarfilm for leaving black sticky glue residue on the balsa. 15cm squared used rest in the bin....

Anyone want 4.95 metres of the stuff!

Its not a patch on HK film and I've just been fleeced for some Black Oracover profilm.

N

Nigel R05/08/2020 13:55:02
avatar
4082 forum posts
694 photos

I believe you'll be more than ok with what you already have. Just make sure the air must go right by the motor and the ESC. Battery is less important. You may need the odd baffle around motor or ESC.

I have a very similar powertrain on a 3547 size motor (not the one I posted earlier in thread) putting out 750W (ish).

The inlet is a narrow scoop by prop on underside of cowl area.

A slot in firewall allows air through. The ESC is right behind the slot.

The outlet is a simple square hole in fuselage, not even a ramp on this one.

Hope that helps.

BTW the Wayfarer is a nice looking design.

Mike T05/08/2020 14:08:05
553 forum posts
35 photos

OK...

I appreciate your drawing is 'illustrative', but looking at it, all I can see are a couple of better solutions to getting air in and out.

The obvious one is to fix your ESC to the underside of the inclined battery shelf (I assume that's what I can see?), then vent through the cowl bottom between (A-A and B-B). A semicircular cutout with a lip at the front edge should do it.

The second is an internal ramp aft of C-C, to vent through the cockpit, but in all honesty, despite the 6mm gap you've left around the battery, any air that gets beyond A-A is just going to be chuffing around going nowhere (IMV).

Again IMV, the only way you're going to get air through the exit hole you've made is to drag it out with a fan - pressure differentials just aren't going to cut it.

Happy to stand correction on all or any points...

Nigel Sharp05/08/2020 15:16:13
39 forum posts
166 photos

Everyone

Yes. I'm going with what I've got.

I liked the Wayfarer - I like the AcroStar better but hadn't discovered that before I ordered the balsa for this. They are more or less the same anyway and I've graphted the AcroStar nose onto the Wayfarer.

I added an internal ramp to the vent that means its not a black hole.

I can easily add more vent to the open cockpit if needed and or add some exit to the front underbelly diguised as large bore exhaust pipe(s).

Take the point about the ESC on the underside of battery but its essentially in the same place +/-6mm.

When flown I'll know for certain. Got to rejoin a club but I want to finish this build first.

The engineer in me would like data... so I probably will make a wind tunnel out of mdf, motor, transformer manometer etc. I've loads of stuff in the garage. If I do I'll post the results here to complete the thread (no promises though). If the competitive slalom kayaking gets underway I'll have literally no weekend time to do my hobby instead supporting my son at his.

N

Doug Moss05/08/2020 15:38:16
13 forum posts

Wind tunnel sounds like fun.

No matter what shape or position you decide on for the duct, remember the exit needs to be bigger than the entry.

Peter G Simpson05/08/2020 15:57:42
avatar
36 forum posts
193 photos

This is a really interesting topic, and quite important. Over the years I’ve had no end of problems with engine cooling and I’ve seen lots of other modellers suffer the same problems, engines quit because they overheat.

I picked up a tip somewhere which I’ve used to good success on several close cowled models. Regarding the air outlet, just having a plain hole to let air out doesn’t always work, especially if the hole is angled slightly forward. The solution is to have a slight lip on the leading edge of the outlet hole which causes a draw effect.

A really good full size example of this is on the Pitts Special, there is a clear lip on the leading edge of the hole.

01bf2ce9-c028-45d7-8744-62ad5317cff9.jpeg

ab9df2cd-276b-4e7a-a2cd-4229c2e76018.jpeg

Edited By Peter G Simpson on 05/08/2020 16:01:33

Simon Chaddock05/08/2020 16:05:13
avatar
5776 forum posts
3055 photos

Nigel

The actual electronic "switch" in a Mosfet has an efficiency (which tends to drop as the current increases) so it generates heat at each on/off cycle. In general the actual switch has an operating limit of 150 Centigrade but of course it is contained in a moulded resin case so the rate at which this heat can escape is limited. The bigger the temperature difference between the outside of the component and the inside the cooler it will run.

It the Mosfet over heats it fails, usually open circuit but not always hence the odd case of an ESC fire.

The reliability of a Mosfet is effected by its operating temperature so the cooler it runs the long the time between failure.

Cooling does not impact on the ESC's working efficiency for a given duty (that is determined by its design) but only its life expectancy.

As the efficiency of a MOSFET drops slightly with load and hence the heat generated increases there is an advantage leaving headroom from its max design value particularly if that figure has been set with 'marketing' in mind. wink 2

Nigel R05/08/2020 16:23:46
avatar
4082 forum posts
694 photos

"Cooling does not impact on the ESC's working efficiency for a given duty"

Well, given that if you force cool the device it's junction temperature will drop thus leading to reduced drain/source resistance, I would disagree, but I was rather hoping for richard to explain his 'efficiency is greatly reduced' and 'disguises inefficiencies' statements, and why any of that meant examining the surrounding powertrain would be needed.

Nigel Sharp05/08/2020 16:48:23
39 forum posts
166 photos

Yes I agree it is most interesting.

Problem for us model makers is that we on the ground and cannot actually feel the passing air inside the model. If by luck at landing everything is cold you may assume you got everything correct.

Looking at a river like at a slalom event you get eddies at boulders they go round and round and not much else...but there is an eddie line too. Catch that wrong you get dunked. Thats a skill the competitors learn how to navigate the gates deliberately placed there to best use the water flow.

That's one of the reasons why I was thinking of suction the fuselage must be filled with eddies. The only real way to tell is get data. For that you need to know the pressure differential

Also If I think of slope soaring... at the lip of the slope you get a high velocity rip wind... go 20 to 50ft down the front its much less.

Behind the slope its slow moving possibly turbulent. Dynamic soaring uses these phenomenon to produce really high speeds.

Then

Look at the leading edge Slats they are there to speed airflow near the surface to increase lift. They are used to allow slower takeoff and landing speeds. Go back several hundred years the principle is the same but not really understood e.g.. with a Yacht Jib that slot between the main sail and Jib is all important. I know because you win or lose by bad sail trim. They are just vertical wings - new sailing boats are being designed with hard wing sails.

Could you have a through box with the standard venturi cone input and output angles suspended above the inverse duct? Perhaps this would increase the airspeed locally and therefore create substantially lower pressure.

The key here is to get the optimum position of vortex generation hence suction. Mr Dyson cornered that market with his Cyclone Vacuum they are certainly powerful things.

In most reading in RCME and other magazines you see lots of holes... but not much else.

N

Richard Clark 205/08/2020 16:59:17
426 forum posts
Posted by Nigel R on 05/08/2020 11:10:17:
Posted by Richard Clark 2 on 05/08/2020 03:30:44:

Unlike an IC engine an electric motor and the ESC should run no more than 'warm' as the efficiency is greatly reduced if it is hot. You are not trying to make a room heater.

And carrying that heat away with a strong airflow merely 'disguises' the heat caused inefficiency as the motor is still generating the heat you are carrying away.

So if the motor and/or the ESC is getting hot enough to need a lot of cooling you should look at the suitability of the motor, the prop size, the number of cells relevant to the prop size, and so on rather than trying to improve the cooling.

 

I'll keep an open mind.

I have several questions:

1) Can you explain how an ESC efficiency drops with temperature?

2) Can you explain how a strong airflow "disguises" this proposed inefficiency?

3) Can you explain why "number of cells relevant to the prop size" is a measurement worth investigating?

4) Can you explain why the power train design should be assumed to be at fault?

Here goes. I have numbered your questionw so it's simple to connect question and answer.

1) Unless 'superconducting' (Google is your friend) every electrical circuit has 'resistance' to the passage of electricity. This means the circuit wastes some of the electrical in heating itself which i not what you want.

2) The inefficiency is not "proposed", it is real and measurable. Having hopefully accepted 1 above it follows that passing the generated heat along to something else, in this instance the air has not stopped this heat being  generated in the first place.

3) Given a fixed voltage (number of cells) and  full throttle a  given prop will cause a fixed load on the motor. You can s reduce this load (and thus  heat)  by using  a smaller prop or less cells.  And vice-versa of course.

4) It's  not  a fault  it's  the  laws of the universe  (as so far discovered). Some  people understand  them  better than others so designs vary in competence.

Edited By Richard Clark 2 on 05/08/2020 17:18:25

Richard Clark 206/08/2020 04:10:49
426 forum posts
Posted by Mike T on 04/08/2020 13:30:07:
Posted by trevor wood 2 on 04/08/2020 01:22:16:

image.jpegPerhaps someone should have told Colin Chapman that they are only for intakes before he designed his F1 Championship winning Lotus 49B...

On the basis that neither you nor I know what CC's thinking was with this, I'm going to assume that he needed to get some air out using the lowest drag solution, not necessary a solution which promoted maximum airflow through the duct.

If you want to actively promote airflow, you need a turbulator to reduce pressure at the duct aperture, otherwise a NACA type duct is just a cool-looking hole... smiley

That's certainly all the Lotus ones are. In addition to the air going the wrong way they are not even close to the very specific details of a NACA duct.

Richard Clark 206/08/2020 04:45:50
426 forum posts
Posted by Richard Clark 2 on 05/08/2020 16:59:17:
Posted by Nigel R on 05/08/2020 11:10:17:
Posted by Richard Clark 2 on 05/08/2020 03:30:44:

Unlike an IC engine an electric motor and the ESC should run no more than 'warm' as the efficiency is greatly reduced if it is hot. You are not trying to make a room heater.

And carrying that heat away with a strong airflow merely 'disguises' the heat caused inefficiency as the motor is still generating the heat you are carrying away.

So if the motor and/or the ESC is getting hot enough to need a lot of cooling you should look at the suitability of the motor, the prop size, the number of cells relevant to the prop size, and so on rather than trying to improve the cooling.

 

I'll keep an open mind.

I have several questions:

1) Can you explain how an ESC efficiency drops with temperature?

2) Can you explain how a strong airflow "disguises" this proposed inefficiency?

3) Can you explain why "number of cells relevant to the prop size" is a measurement worth investigating?

4) Can you explain why the power train design should be assumed to be at fault?

Here goes. I have numbered your questionw so it's simple to connect question and answer.

1) Unless 'superconducting' (Google is your friend) every electrical circuit has 'resistance' to the passage of electricity. This means the circuit wastes some of the electrical in heating itself which is not what you want.

2) The inefficiency is not "proposed", it is real and measurable. Having hopefully accepted 1 above it follows that passing the generated heat along to something else, in this instance the air has not stopped this heat being generated in the first place.

3) Given a fixed voltage (number of cells) and full throttle a given prop will cause a fixed load on the motor. You can s reduce this load (and thus heat) by using a smaller prop or less cells. And vice-versa of course.

4) It's not a fault it's the laws of the universe (as so far discovered). Some people understand them better than others so designs vary in competence.

ADDENDUM:

It all boils down to this: ALL heating in electrical systems other than those specifically deaigned to produce heat is a sure sign of inefficiency, be that inefficiency small as in FETs, or big, as in a badly matched motor, propellor, and supply voltage combination.

EG: My Black Horse 90m fan 55 inch span 10 cell high power EDF ViperJet.

First try - Hacker B50 14L inrunner motor, OS 1100HV ESC. Thrust at full power 9 pounds, 'cruising around' duration 7 minutes. Motor hot even with a heatsink, ESC 'warmer than I would like'.

Second try - Hacker E50L 2D inrunner motor, (3mm larger diameter and 20mm longer than the B50 above), Kontronik Jive Pro ESC. Thrust at full power 10 pounds, cruise duration 10 minutes, no motor heatsink, motor and ESC slightly above ambient temperature.

Notes: (1) If your ESC is getting  as hot as you say  it either doesn't use FETs or something is badly wrong. (2) Though specified to run at a high maximum current or wattage  motors are far more efficient with less - the B50 above was being driven too hard,

Edited By Richard Clark 2 on 06/08/2020 05:09:27

Nigel R06/08/2020 08:17:51
avatar
4082 forum posts
694 photos
Posted by Richard Clark 2 on 05/08/2020 16:59:17:
Posted by Nigel R on 05/08/2020 11:10:17:
Posted by Richard Clark 2 on 05/08/2020 03:30:44:

Unlike an IC engine an electric motor and the ESC should run no more than 'warm' as the efficiency is greatly reduced if it is hot. You are not trying to make a room heater.

And carrying that heat away with a strong airflow merely 'disguises' the heat caused inefficiency as the motor is still generating the heat you are carrying away.

So if the motor and/or the ESC is getting hot enough to need a lot of cooling you should look at the suitability of the motor, the prop size, the number of cells relevant to the prop size, and so on rather than trying to improve the cooling.

 

I'll keep an open mind.

I have several questions:

1) Can you explain how an ESC efficiency drops with temperature?

2) Can you explain how a strong airflow "disguises" this proposed inefficiency?

3) Can you explain why "number of cells relevant to the prop size" is a measurement worth investigating?

4) Can you explain why the power train design should be assumed to be at fault?

Here goes. I have numbered your questionw so it's simple to connect question and answer.

1) Unless 'superconducting' (Google is your friend) every electrical circuit has 'resistance' to the passage of electricity. This means the circuit wastes some of the electrical in heating itself which i not what you want.

2) The inefficiency is not "proposed", it is real and measurable. Having hopefully accepted 1 above it follows that passing the generated heat along to something else, in this instance the air has not stopped this heat being generated in the first place.

3) Given a fixed voltage (number of cells) and full throttle a given prop will cause a fixed load on the motor. You can s reduce this load (and thus heat) by using a smaller prop or less cells. And vice-versa of course.

4) It's not a fault it's the laws of the universe (as so far discovered). Some people understand them better than others so designs vary in competence.

Edited By Richard Clark 2 on 05/08/2020 17:18:25

hi richard

1) hasn't answered the temperature related efficiency question

I mean, I'll state my POV; heat relates to power relates to I2R(approximately, because reactance etc), so at any given sensibly sized ESC the current will not affect junction temperature within ESC until a point on that curve where the power dissipation begins to rise rapidly due to nature of second order curve. Maybe you have a more sophisticated understanding, hardware isn't my speciality.

2) hasn't answered why you say it 'disguises' the inefficiency

Again, my POV, you just need to cool the device a certain amount. If you need X amount of cooling - in our case facilitated by airflow - to maintain a low enough junction temp that your ESC operates "efficiently" then, well, er, so what? Airflow isn't disguising anything; it is just a means of achieving an end. Again, maybe you can help me understand what I'm missing here, but this is the basic process the hardware guys at my work operate with.

3) ok - but that means that the ESC isn't sized correctly for the current

4) ok - but again means that the ESC isn't sized correctly for the current

Your example in the second post appears to concur with my reply to 3 & 4 above, as things relate to the ESC, ignoring the motor aspect of your setup.

Edited By Nigel R on 06/08/2020 08:19:41

All Topics | Latest Posts

Please login to post a reply.

Support Our Partners
CML
Sussex Model Centre
electricwingman 2017
Slec
Sarik
Advertise With Us
Latest "For Sale" Ads
NEW POLL - has the pandemic altered your event safety perceptions?
Q: Has the covid pandemic deterred you from attending shows and events in 2021?

 No, I'll be attending just as many as I usually do
 No, but I'll choose my event with greater care
 Yes, I'll attend fewer events going forward
 Yes, I wont attend any where previously I have

Latest Reviews
Digital Back Issues

RCM&E Digital Back Issues

Contact us

Contact us

Magazine Locator

Want the latest issue of RCM&E? Use our magazine locator link to find your nearest stockist!

Find RCM&E!