Water Coolling An RC Sub And Why I Do It

[AQUASUB]

Water cooling!

No matter what kind of motor or drive one installs in a RC Sub, the motor will get warm and continue to get warmer until it reaches a point that a balance between the torque imposed on the motor and the input energy used over a given amount of time is reached. That can be damaging to the motor and the watertight casing of the sub. The surrounding air in the casing gets warmer and expands, sometimes with a lot of pressure, unseating the watertight seals and shaft seals.
I have seen those very small cheap subs about 3 inches long actually burst their seal from just the heat of the motor!

Will continue soon.

[Slats]

Water cooling!

No matter what kind of motor or drive one installs in a RC Sub, the motor will get warm and continue to get warmer until it reaches a point that a balance between the torque imposed on the motor and the input energy used over a given amount of time is reached. That can be damaging to the motor and the watertight casing of the sub. The surrounding air in the casing gets warmer and expands, sometimes with a lot of pressure, unseating the watertight seals and shaft seals.
I have seen those very small cheap subs about 3 inches long actually burst their seal from just the heat of the motor!

Will continue soon.

Sorry mate but I completely disagree with your argument that there is any need for water cooling. AND the generalisation that "No matter what kind of motor or drive one installs in a RC Sub, the motor will get warm and continue to get warmer until it reaches a point that a balance between the torque imposed on the motor and the input energy used over a given amount of time is reached." Emphasis added.

I have been running subs for two decades. Yes all motors will get warm, BUT if they are set up via what is pretty much the established practice of running via a drive reduction gearbox and the assembly is done properly there is no finger burning heat, or sufficient heat to drop performance. The only time I have seen this happen in subs is when the operator set the boat up with
-1 too small a motor for too larger a diameter propeller;
-2 direct drive;
-3 over rating the input voltage of the motor i.e running a motor on say 12v when its max current spec is 7.2v;
-4 the motor fitted is crap
-5 all of the above

The other issue is "motors ain't all the same".
A Standard Johnson 540 Motor will warm up in a fraction of the time cf it's Graupner Cousin of the same can size, or cf it's Electronize Cousin, or cf it's Mabucchi Cousin.

I hold David's engineering skills as pretty much the gold standard of what the basic platform should be - if water cooling was mission critical we'd have seen it by now inside a subdriver. Further, where are all the posts (not just on this site but elsewhere) winging and *****ing saying my subs motor is stuffed - we need a better setup?

I should point out that my Alfa's were engineered along the lines of an underwater jet fighter, no holding back on the throttle for that thing in a pool environment, and whilst I flew at flank bell with that platform, the motor was never what I'd call hot. All my Alfa's were ran through either 3:1 5:1 and later 6:1 gear boxes, the later being only marginally slowly.

I accept the fact, with my 5 observed caveats above, of everything you mentioned about excessive motor heat being a POTENTIAL problem but the fact after 2 decades of patrols as often as a few times a week in summer, I just don't get your argument. Why? The problem simply has not eventuated. Set you boat up right with a gear box, operate your motor within its max voltage, and through a speed controller and in my extensive experience you should not have a problem either. Installing water cooling on a RC sub is another complexity for the already crammed WTC. Another key consideration is that how many RC submariners drive at flank bell anywhere other than a controlled pool type environment? I'd expect most don't in blue water ops where the boat is confined pretty much to PD. At such average lower speeds any heat will unlikely go beyond a warm status.


J

[AQUASUB]

Hi John

Yes I agree with all of the above you have mentioned and expected the response, that is why I started the thread, to bring up this issue.

Sure it is controversial, but no matter, the laws of physics apply here. A motor not connected to anything will generate heat and it should be dealt with, or dissipated in some manner, even if one feels it is not necessary it is still a fact of operating a RC Sub.

I have been building and operating RC Subs for well over 20 years and some 27 subs of all sizes and shapes, and all but 2 early ones had no water cooling. All have 12 volt can motors ,3.5 to 1 gearboxes and normal scale props . Many think that I use some kind of mis-matching and that is why I need to water cool them. It is not the case. It makes good sense to have the extra insurance policy that a cool motor is running cooler in that totally sealed container.

Believe it or not your sub is already water cooled because the cylibnder is submerged in water. Run it out of the water and it will get noticeably warmer.

You see all I want to do is remove "any heat" from the sub and keep the motors well below their normal operating temperature. If I need to realy put the pedal to the metal it will run cooler due to more water flow. Run at flank speed all day long and the motor is just fine and so is the casing.

I even water cool drill motors on very large subs where a huge or high pitched prop is used with 26 to 1 ratios, not because it was necessary, but because I could do it without much added complexity, cost or time.

If one can coil some aluminum tubing or copper and set up a simple means to pump water around like a forced system as on speed boats. It is a lot of fun, challenging to make and to see how your engineering worked out. Heck! Added complexity is my second name. .

So yes I agree with you but it is in my opinion one should water cool the submarine. Its just good sense to me.

Dave Amur Sub Yard

[AQUASUB]

Later on I will submit actual running temperatures done with real time Telemitry of runinng with and without water cooling ,with the same casing/hardwear,even the internal air pressure can be measured.

[Slats]

Dave,
a neat piece of engineering -I can see that if ever someone puts a fuel cell, or miniature pressure reactor on board - your our man. But we're going to have to agree to disagree. If you run mostly at walking speed, I can't see why there would be any excessive heat build up that would cause a drop in performance... In terms of battery consumption, my record with a standard sub driver at half a head was 3 hours out of a 4300nimH battery. I considered this and any longer as ridiculous as the gear as far as I am concerned needs the occasional down time (not just because of any heat build up), but that is the best Insurance you can have is having a break. That goes for the guy behind the sticks too.

My point still stands - PRACTICALITY - if it ain't broke why fix it - or fix at on as you say, "Insurance" that you don't need. The average Joe simply doesn't need all this extra on board the sub. So hats off to some great engineering but its not a practical mission critical essential. I have been running scale targets too for 25 odd years and at only at one time did I ever require active motor cooling. My point is much can be done in mating the right drive package to ship to avoid the need for active cooling. To me the already pretty much crammed WTC is heading into Rube Goldberg territory when there are ways around it - such as swapping the motor out for a more efficient one. David's Sub drivers are pretty much bullet proof in this respect with a fairly torquey motor through drive reduction gear box and the advantage of extra torque load shared over two motors if required. These I have never know to be hot - warm yes, but there well within the "normal" operational envelope of what most people will need.

Ok having said that we're going to agree to disagree, here's me "actively" disagreeing by offering up some advice. You could extend a small conduit down and aft the WTC end cap that penetrates through the base of the hull with a forward facing pick-up. You will find that the momentum of the sub will be enough to get good water circulation going and you could even prime the water outlet line with a bit of suction to ensure the loop works well. This would eliminate the need for a mechanical pump. I have employed this set up on a surface ship and the water flow even at very slow speed was fairly constant.

J

[Subculture]

In the cylinder I'm making up for the Aquabat, the brushless outrunner I'm fitting (this is a sporty sub) is rated at about 150 watts, or thereabouts. As these motors tend to be about 80-85% efficient, that means about 20 watts of heat kicking about, which is enough to make me consider some form of extra cooling.

I'm making use of natural heat convection by making a metal fitting for the motor to slot into, which projects into the wet stuff. Simple and cheap.

[Slats]

Andy,
for a standard scale sub travelling at scale speed would you consider active cooling - or in you case above passive cooling?
J

[AQUASUB]

Subculture

I would very much recommend that you water cool that motor and with a passive method.

From your stats, a sealed more aggressive coil to coil method will really do the job. Its better to have more cooling capability as heat and internal air pressure will build up accordingly. I have done that several times in my Graupner Shark. It suffered from overheating and short stunted run times until I installed the passive cooling plates/fins as seen in the pictures.

I know it works because I can feel the fins get warm when the sub runs out of the water and there is much less pressure build up that used to blow by the dive plane seals, (the built in pressure relief valve I guess). The modification to install 2ea. 2300 a lipos and water cooling did the trick. Now she literally leaps out of the water much higher and runs phenomenally longer than before.

On my latest build the Type XXI I am refitting the casing and have employed passive cooling by simply having the motors come into direct contact with the inner wall of the casing which will then transfer the heat out.

Since this sub is not intended to go at flank all the time, nor is the hull that type a 'watt-guzzler' like the U-32. A less aggressive approach to cooling was warranted.
On the U-32 I had to use a circulation pump and coils or a sealed liquid system to bring the temps down from 160 F without cooling to a cool 96 F and that was read at flank for 20 minutes. This sub was an exception as it turns a 5.5 inch 7 bladed monster of a prop and has a 12 volt Makita 26 to one gearbox giving this sub fantastic performance.

The Type XXI is very different, it's sleek bow and shape slips through the water with virtually no wake, hence less power needed to go the same speed and hence cooling of the 2 motors was much simpler with just passive cooling methods,
So I have used mini pumps to the more efficient direct off the the shaft drive pumps that way there is virtually no energy loss from pumping.

I also tried the system slats mentioned of a forced intake. It was not very effective with scale subs as the scale speed was just not enough to effectively pump the water all the way around. It works on speed boats and if done right on subs it could be good.

To me the best method is to totally submerge the motor and gear box.

That can be done easily and without much expense by using a modified Rule bilge pump and a sealed gearbox as shown. The completed unit is doing fine in tests as I test to the extreme all the components installed in my subs.

I want to know whats going to fail and why at the Yards Lab before it goes to sea. I am also sourcing out other watertight motors that might be suitable.

If I were to answer for Subculture, it would depend on the sub and what was its intended mission. A speeder should have a good active cooling system while a scale one a passive type system would suffice,however there are exceptions to this as well since the running environment also dictates its design.
I often run my subs in the crystal clear rivers fed by springs. In order for me to run effectively in a relatively high current river, I had to make the subs water cooling system more effective or by using 2 pump on some subs.

It is fun to run in a clear river. One does not need to walk or have the sub go too far to have a blast, as the sub can be at station keeping while at near or at full and not move an inch!
Its also the best way to test the high speed dynamics of the sub as is done in a water current test tank, loads of fun!

Dave Amur

[Subculture]

For scale subjects, unless I was going very big I personally wouldn't bother cooling beyond the air that surrounds the motor, this is assuming that you have a good combination of an efficient motor running within about 75% or higher of its free running RPM range. Generally most scale boats at the size we model to only need about 5-10 watts of shaft power to move along nicely. If we assume that our motors are only 50% efficient that still equates to no more than about 5 watts of heat, which is equivalent to a modest torch bulb.

In the case of the Graupner Shark that Dave mentioned, I quite agree that some form of cooling would be desirable. This boat is a sub aquatic missile, and was originally designed to operate for a few minutes at the most bearing in mind the limits of battery technology 20 years ago.

With Lithium or NiMH battery technology, it's possible to keep this boat swimming around for an hour or so, these days in which case things could get a bit funky inside, temperature wise.

The majority of heat produced in a conventional brushed DC motor is generated by the friction of the brushes rubbing on the commutator and in the windings themselves. Unfortunately this is also the most complicated to cool- running a coil around the can of the motor will merely serve to cool the magnets, as the coil runs in the middle. Brushes can be cooled, and if you look at very high quality brushed motors intended for high performance e.g. Plettenberg, you will see that the brushes to indeed have their very own little heatsinks fed by water.

Brushless motors are a different kettle of fish. For a start they are electronically commutated, so no brushes or commutator to concern ourselves with. Secondly the coil remains static on both inrunner and outrunner motors, the magnets rotate, therefore we can employ methods of cooling directly on the coil. This is one of the main reasons we can get such huge amounts power out of these motors without cooking them.

In the the case of my Aquabat, I see no advantage of employing a pumped water cooling system, as my boat will be enveloped by water, and basic thermodynamics dictates that the heat will flow through to the cooler body by natural convection if I have the motors coil/armature in thermal contact with the water.

If I mounted the motor deep inside the wtc, then I might need some active means of getting the water to the motor. In the case of fast electric boats, they spend 99% of their time out of the water, or skimming just over it, so once again, some form of active pumping is required in the form of a scoop.

[AQUASUB]

Could you show some pictures of your Aquabat?
I like that title. I have that name also reserved for a special ultra fast sub, a larger version of the Graupner Shark.

Yes what you have stated also reflects in what I have been writing about in that it depends on the mission,size and designed /desired end capabilities of the craft as to what kind of water cooling is best.

I might suggest that since you like to express alot of numbers and percentages which is very good that I would think you would really like to install a Telemetry system in your Aquabat. It takes up very little space and uses very little power yet gives you a complete display of just what is going on in your sub in real time completely wireless. You can record it and go over it later to find any data needed, even the internal pressure of your casing as it warms up or descends into the deep. This system is no push over or a cheap, poorly made item but a true scientific tool that is invaluable in the development of RC Subs. You will love it, expandable capabilities and super easy to install.

One very cool thing I use mine for is the Depth Read-out, which is sensed by the Altimeter in reverse. It is an extremely precise depth measurement since water does not compress and the sensor picks it all up!


Before I only had an idea just how warm or hot my subs got after I pulled them out of the test tank ,opened them up and then touch the motor, man is that a pain and antiqaited in this day of age. Thanks to Nasa this type of teck is now available to all and not costly. It replaces so many types of bench electronic tools ,too many to list.


The System I use is the EagleTree Seagull, in my opinion its the best out there cost around 350.00$ for a good kit.


I like brushless motors also and will employ more of them in future builds, but for a good old reliable cheap motor for general use one can't go wrong with the can motor coiled or not it keeps on going even if it gets submerged,while a bruschless will ruin its ball bearings and other internals a much more delicate motor,so just be sure they do not get wet.




Laptop display of the EagleTree and other sensors, all can be alarm set so one is notified when for instance the Lipos are getting too hot or have low voltage.


Dave Amur Sub Yard

[Subculture]

Aquabat was a kit produced about 20 years ago by Ron Perrott (he is on this forum posts as 'RonP'). The boat is about 20" (50cm) long.

A vac-formed HIPS model designed for simple 2-channel operation (multi-channel sets were big money 20 years ago). Despite the small number of channels, Ron managed to squeeze in water pump ballast system with a spherical tank.

A 540 motor and large control surfaces gave it excellent underwater performance.

Ron brought this bat along as a freebie, but needed quite a bit of work to put it back in the water. Unfortunately the original dry hull design had to be ditched as this was beyond repair- whoever built this one originally glued it together with cyano, and there were a lot of fractures all over the lower half of the hull. Suffice it to say, I didn't fancy trusting my radio kit to this, so a new approach was sought.

This refreshed bat is getting get an acrylic cylinder, a brushless motor will replace the brushed 540, with lithium cells replacing the originals lead acid battery. I've also decided to give the ballast system the elbow, as I will be happy with this as a dynamic diver bearing in mind I tend to sail in pools and this subject has no 'scale' waterline.

Anyway, still a work in progress at the moment, but here are some shots I posted a few weeks back which should give you an idea of what the boat looks like. The second one shows the new bottom hatch master I made. The boat was missing this and Ron didn't have any spares, so I fashioned a new one from some MDF and acrylic, and will mould this in GRP. At the moment I'm making up a new kort nozzle and rear plane assembly.

I use eagletree loggers on model aeroplanes. I have a V3 micrologger, and it's a very useful bit of kit, especially for checking motor and prop combinations. I think telemetry is a bit of gimmick for models, but logging which enables me to monitor things back home behind my computer is definitely useful for new models, and I will be using this to check how things are working in the Bat.

[AQUASUB]

I use Eagletree loggers on model aeroplanes. I have a V3 micro-logger, and it's a very useful bit of kit, especially for checking motor and prop combinations. I think telemetry is a bit of gimmick for models, but logging which enables me to monitor things back home behind my computer is definitely useful for new models, and I will be using this to check how things are working in the Bat.

Why do you feel that Real Time Telemetry is a gimmick?

I find that rather puzzling since you allready use and enjoy the Elogger or a half of the system. It would be imperative to see and know right away that the lipos are getting close to their lower limit, or that the motor is running a bit too hot. One cannot see those things after the fact. It must be reported live as it occurs, just like it is done with all experimental aircraft, missiles,submarines,drones,cars,boats,robots, test models - even test dummies have extensive logged back up.

Its just not for aircraft though it came from that field,it can be applied to your toaster and to your PC.
If I thought it was a gimmick and found it to be so I would have simply returned it, but no way I am buying more sensors and TXs etc. to do at least 3 to 4 subs.

With the above package I can receive data at a distance of about 60 meters while at periscope depth, and much further while on the surface. In the clear water the range is much better and one can still get reception even at 10 feet down.
There is an even more powerful TX unit on 900mghrz like the one I have. Its about a watt of power and was quite epensive. Later I might get that one.
The antenna can be extended to the sail as the subs comunication mast. I did this on the Robbe U-32.

I run in my pool and from my recliner in the living room. I can see on the Seagull's readout video feed, live real time updates sampled in milliseconds all pertaining data. Now that's taking it easy on a rainy day!

The live option really shows its strengths on an RC sub. One needs not to wait till the sub is out and opened from a tank or pond run to feel if the motors are too warm, by that time its way too late and not at all precise.

Dave Amur Sub Yard

[Albion]

The majority of heat produced in a conventional brushed DC motor is generated by the friction of the brushes rubbing on the commutator and in the windings themselves. Unfortunately this is also the most complicated to cool- running a coil around the can of the motor will merely serve to cool the magnets, as the coil runs in the middle. Brushes can be cooled, and if you look at very high quality brushed motors intended for high performance e.g. Plettenberg, you will see that the brushes to indeed have their very own little heatsinks fed by water.

Thats a good point, going back to my slot racing days, we often had large aluminium heatsinks which formed part of the brush housing.
No other cooling required

[AQUASUB]

Since as I can see that a brushed motors heat begins in the brushes and armature, and on to the bushings and bearings, then to the magnets and can, a coil would be very effective since it is all connected and in constant contact. Hheat extraction is transferred directly and naturally just like the manufacturers intended it to be.

[Slats]

For scale subjects, unless I was going very big I personally wouldn't bother cooling beyond the air that surrounds the motor, this is assuming that you have a good combination of an efficient motor running within about 75% or higher of its free running RPM range. Generally most scale boats at the size we model to only need about 5-10 watts of shaft power to move along nicely. If we assume that our motors are only 50% efficient that still equates to no more than about 5 watts of heat, which is equivalent to a modest torch bulb.

In the case of the Graupner Shark that Dave mentioned, I quite agree that some form of cooling would be desirable. This boat is a sub aquatic missile, and was originally designed to operate for a few minutes at the most bearing in mind the limits of battery technology 20 years ago.

With Lithium or NiMH battery technology, it's possible to keep this boat swimming around for an hour or so, these days in which case things could get a bit funky inside, temperature wise.

The majority of heat produced in a conventional brushed DC motor is generated by the friction of the brushes rubbing on the commutator and in the windings themselves. Unfortunately this is also the most complicated to cool- running a coil around the can of the motor will merely serve to cool the magnets, as the coil runs in the middle. Brushes can be cooled, and if you look at very high quality brushed motors intended for high performance e.g. Plettenberg, you will see that the brushes to indeed have their very own little heatsinks fed by water.

Brushless motors are a different kettle of fish. For a start they are electronically commutated, so no brushes or commutator to concern ourselves with. Secondly the coil remains static on both inrunner and outrunner motors, the magnets rotate, therefore we can employ methods of cooling directly on the coil. This is one of the main reasons we can get such huge amounts power out of these motors without cooking them.

In the the case of my Aquabat, I see no advantage of employing a pumped water cooling system, as my boat will be enveloped by water, and basic thermodynamics dictates that the heat will flow through to the cooler body by natural convection if I have the motors coil/armature in thermal contact with the water.

If I mounted the motor deep inside the wtc, then I might need some active means of getting the water to the motor. In the case of fast electric boats, they spend 99% of their time out of the water, or skimming just over it, so once again, some form of active pumping is required in the form of a scoop.

Thanks Andy I appreciate it. But Dave reconsidering your initial post and my whole point of a counter view you said:

No matter what kind of motor or drive one installs in a RC Sub, the motor will get warm and continue to get warmer until it reaches a point that a balance between the torque imposed on the motor and the input energy used over a given amount of time is reached.

this is a huge call - and it simply is not true for all applications and sub setups. Yes the motor will get warm, BUT depending on the motor, the setup, and how the sub is driven, or what type of sub it is, you can't simply state that it will "continue to get warmer until it reaches a point that a balance between the torque imposed on the motor and the input energy used over a given amount of time is reached." Your statement above is my main issue. Prefaced by "No matter what kind of motor or drive one installs in a RC Sub" it is an all encompassing generalisation that is simply not in evidence for all RC sub applications and in my view not helpful to the new guy that may read this as a "given" for ALL situations. There are too many bum steers in this hobby, too much generalised noise. I admire your skill and your proven track record and please do show us how one might water cool - where it is necessary or is indeed crucial, but please lets leave the assumptions about the whole world being one size fits all. Lets not over complicate things for the already confused new guy.

Best

J