How To Ice Up Your Motor? (Cooling Ideas)
- Thread starter safe
- Start date
safe
1 GW
- Joined
- Dec 22, 2006
- Messages
- 5,681
This also supports the "Ice Cooling" idea.If you could cool the motor shell to near freezing by using ice and use a good heat transfer material like copper you could force the motor down to a lower electrical resistance and help to make the magnets have more magnetic strength.
Colder Motor = More Powerful Motor
flip_normal
100 W
High temp magnets & heat pipes; these are gonna make for electric motors with awesome specs!
From the heat pipe article;
Nice finds safe, Dr. Shock. Thanks for posting.
From the heat pipe article;
Nice finds safe, Dr. Shock. Thanks for posting.
TylerDurden
100 GW
My laptop uses a heatpipe, about 6" long on each side of the CPU.
(Note that they use forced air cooling.)
8)
(Note that they use forced air cooling.)
8)
safe
1 GW
- Joined
- Dec 22, 2006
- Messages
- 5,681
From what I know the heat pipe uses radiation to transfer heat and since that happens in all directions it should not matter. Convection requires that the liquid circulates... the heat pipe doesn't move the liquid (or gas) at all.Dr. Shock said:
I suppose in theory I could wrap the motor with a heat pipe and take away heat remotely at the other end of the pipe, but since it's just as easy to place the ice near the motor I might as well keep the ice close.
The big advantage of ice over water is that ice has 80 times as much heat absorption ability for the same weight and at FREEZING. It's the fact that you can super cool the motor that is the real magic of the idea. Fans can only bring a hot motor down to room temperature. Ice can drop the motor temperature to below room temperature. This means that off the line at the start the first minute or two might allow your motor to be running stronger than the fan cooled alternative. Since our races will likely be short the advantage of super cooling with ice will be noticeable.
At least that's the idea...
TylerDurden
100 GW
Correct. There is little need for you to relocate the cooling surfaces, although circulation of the water/ice can only help.safe said:
Dr. Shock
100 W
- Joined
- Dec 28, 2007
- Messages
- 274
I thought the idea was to absorb heat through boiling the fluid inside the pipe, have the vapor wick up the tube and release its energy at the cold end through condensation, and have the liquid return to the hot end by force of gravity. I was under the impression some vertical slope is necessary to get an efficient cycle in a heat pipe. I only know what I read, however, as I haven't gotten around to making one of these things.
TylerDurden
100 GW
That's right... but the release of heat is accomplished by the large mass of available air passing the condensor elements. The evaporation and condensation facilitate convection without a circulation pump. It wouldn't be useful, if there weren't greater temperature differentials at the condensor end.Dr. Shock said:
This also points at the weakness of the ice-cooling concept: the ice eventually will reach ambient temperature as water, then act as an insulator, retaining heat. Forced-air cooling can indefinitely direct a limitless mass of air directly across the coils, where it is needed the most.
safe
1 GW
- Joined
- Dec 22, 2006
- Messages
- 5,681
TylerDurden said:
Eventually all the ice melts and ONLY THEN does the water temperature rise above freezing. Come on... don't you remember your physics lab class? That was the big "wow that's fascinating" lesson about ice and water which was that when ice is in water the water cannot rise above freezing until ALL the ice melts.
So the "insulation" of water idea is completely false... 
And it takes 80 cal/g to make ice melt, so for the same weight you get 80 times more cooling than if you used water alone.
Also... one assumes that if their "ride" is more of a "race" which is how I intend to use the machine that by the time the ice melts that my ride will be coming to an end. That's why I calculated the total energy needed and came to the conclusion that for the limited battery power I will have on board the bike and the heat production of the motor that I will need roughly one liter of ice to completely satisfy my cooling needs at roughly three times the rated heat. This agrees with the experimental results other people have done using water cooled motors.
Even if all the ice melts the water still increases the overall thermal mass of the motor so you still have to take the water all the way from freezing. To raise that one gram of water from freezing (0 C) to boiling (100 C) means that you have to add ANOTHER 100 cal/g to all that water. So if the one liter rule applies to the ice, then the water boiling rule means that now the idea has a 50% margin of error just to get to boiling. Once you actually start to boil off the water you give off 540 cal/g of heat and that means that you have a virtually unlimited safety factor as far as overheating. The main thing will be to make sure that the ice container vessel has an air vent that allows the phase changes to take place without causing any problems.
Finally we need to remember WHY we do this... the only reason to trick out a small motor this way is because we want to stay within the Federal Law of 750 watts as the baseline for the motor. My first test for ice will be on a 350 watt motor, but ultimately the idea is to take a perfectly LEGAL motor and trick it out so that it has much more power. This fits into my ideal of a "Modified" electric bike racing class that requires the motor that you start with as a base to be legal. And the crazy way the law is written it makes the "rated load" really turn out to be about a "rated heat". The law has unknowingly created it's definition of performance based on heat.
PS: It's something like 10 degrees outside so I stayed home rather than go out and party. :|
TylerDurden
100 GW
Blah-blah...
Just stick yer motor in a baggie and put it in a bucket with ice and see what happens.
Just stick yer motor in a baggie and put it in a bucket with ice and see what happens.
Link
1 MW
Jeez, this is still going on (yes, I read the whole. damn. thread.)?
IMO, ice will work for a small motor for a short period of time. Namely, the time it takes for the armature to heat up. After that, you will get some cooling from the magnets. After that, it will fry. This is assuming you're really cranking the amps (e.g. saturating the core, which means you're killing efficiency). If you keep it to a reasonable power draw, then the ice probably won't help much more than air.
The air might be better for bigger motors, since there is too much magnet between the armature and case. On something like a pancake motor (e.g. Etek) air will probably work as well or better than ice, since there is plenty of surface area on the rotor.
However, in the case of a brushless motor where the coils are mounted to the case (like a Kollmorgen), ice will be far superior for as long as it stays ice.
But, as long as you keep power at levels the motor can handle and don't saturate the armature, it probably won't matter much if you use air or ice.
It doesn't really matter, though, because I have an idea that pwns both ice and air. Liquid air 8). Inject some liquid gas (oxymoronic, much?) into a motor. Assuming you don't crack something, it would be nearly impossible to fry the thing. Not nearly as cheap as ice or a small battery and fan, but I'd like to see something that could out-cool liquid N2. (Shut up. I know liquid helium or hydrogen or what have you will out cool it. I was trying to make a point.)
IMO, ice will work for a small motor for a short period of time. Namely, the time it takes for the armature to heat up. After that, you will get some cooling from the magnets. After that, it will fry. This is assuming you're really cranking the amps (e.g. saturating the core, which means you're killing efficiency). If you keep it to a reasonable power draw, then the ice probably won't help much more than air.
The air might be better for bigger motors, since there is too much magnet between the armature and case. On something like a pancake motor (e.g. Etek) air will probably work as well or better than ice, since there is plenty of surface area on the rotor.
However, in the case of a brushless motor where the coils are mounted to the case (like a Kollmorgen), ice will be far superior for as long as it stays ice.
But, as long as you keep power at levels the motor can handle and don't saturate the armature, it probably won't matter much if you use air or ice.
It doesn't really matter, though, because I have an idea that pwns both ice and air. Liquid air 8). Inject some liquid gas (oxymoronic, much?) into a motor. Assuming you don't crack something, it would be nearly impossible to fry the thing. Not nearly as cheap as ice or a small battery and fan, but I'd like to see something that could out-cool liquid N2. (Shut up. I know liquid helium or hydrogen or what have you will out cool it. I was trying to make a point.)
safe
1 GW
- Joined
- Dec 22, 2006
- Messages
- 5,681
It's All About The Air
Whether you use Ice or forced Air it's the Air that actually does all the work.
To really understand what happens we have to look at the behavior of air in the context of how it transports heat. We are after all (for a brushed motor) talking about a rotor that is getting hot because the coils heat it up. The heat transfer rate from the coils to rotor is going to dwarf the heat transfer rate of the coils to air. The heat transfer rate is something like 20 times faster... I forget:
"You cannot isolate the rotor from the coils." (brushed)
...so if we are dealing with a large mass of metal that spins around then the first approach for cooling is the idea of passing air past this rotor. The air can only carry a fixed amount of energy ACROSS the boundry which is the shell of the motor. This is important to realize... the barrier is the ability to move high volumes of air so as to drop the high temperature rotor (75 C) down to room temperature. (25 C)
But there is also the property of radiation that goes on. If the shell is made cold (freezing) then the rate of direct thermal transfer (think heat pipe here) is faster. Not all forms of heat transfer have to do with the convection of air, sometimes you can get direct transfer. (this only amounts to a small percentage though)
The last and most important property is the EDDY CURRENTS. While normally we think about eddy currents of air as being a bad thing because it tends to count as a negative towards efficiency, in the case of cooling we are pleased to discover that air that is highly active and swirling around tends to transfer heat at very high rates. At HIGH RPM the heat transfer from rotor to shell will be three to four times the transfer rate when at slow speed.
So the "smart guys" are going to exploit these facts... as long as you run the motor at high rpm and use ice to supercool the shell you will have the highest heat transfer rate that you can reasonably expect. The eddy currents at high rpm are like your own unharnessed wind that is now suddenly harnessed.
A good analogy is passive solar homes.
In a passive solar home you exploit all the qualities of nature to your advantage. When the sun is out you store it's energy into a trombe wall. At night you recirculate the heat. In summer you use the earths 55 degree temperature as air conditioning. Using forced air cooling for a motor is the same (in this analogy) as using solar panels to run an air conditioning unit. The end result is the same (you cool off your house) but the strategies for making it happen are very different.
To understand what I'm doing you need to think in a way where you exploit EVERYTHING that there is to exploit.
Of course... I could be wrong... I'll only be able to prove myself when I actually build it.
Whether you use Ice or forced Air it's the Air that actually does all the work.To really understand what happens we have to look at the behavior of air in the context of how it transports heat. We are after all (for a brushed motor) talking about a rotor that is getting hot because the coils heat it up. The heat transfer rate from the coils to rotor is going to dwarf the heat transfer rate of the coils to air. The heat transfer rate is something like 20 times faster... I forget:
"You cannot isolate the rotor from the coils." (brushed)
...so if we are dealing with a large mass of metal that spins around then the first approach for cooling is the idea of passing air past this rotor. The air can only carry a fixed amount of energy ACROSS the boundry which is the shell of the motor. This is important to realize... the barrier is the ability to move high volumes of air so as to drop the high temperature rotor (75 C) down to room temperature. (25 C)
But there is also the property of radiation that goes on. If the shell is made cold (freezing) then the rate of direct thermal transfer (think heat pipe here) is faster. Not all forms of heat transfer have to do with the convection of air, sometimes you can get direct transfer. (this only amounts to a small percentage though)
The last and most important property is the EDDY CURRENTS. While normally we think about eddy currents of air as being a bad thing because it tends to count as a negative towards efficiency, in the case of cooling we are pleased to discover that air that is highly active and swirling around tends to transfer heat at very high rates. At HIGH RPM the heat transfer from rotor to shell will be three to four times the transfer rate when at slow speed.
So the "smart guys" are going to exploit these facts... as long as you run the motor at high rpm and use ice to supercool the shell you will have the highest heat transfer rate that you can reasonably expect. The eddy currents at high rpm are like your own unharnessed wind that is now suddenly harnessed.
A good analogy is passive solar homes.In a passive solar home you exploit all the qualities of nature to your advantage. When the sun is out you store it's energy into a trombe wall. At night you recirculate the heat. In summer you use the earths 55 degree temperature as air conditioning. Using forced air cooling for a motor is the same (in this analogy) as using solar panels to run an air conditioning unit. The end result is the same (you cool off your house) but the strategies for making it happen are very different.
To understand what I'm doing you need to think in a way where you exploit EVERYTHING that there is to exploit.
Of course... I could be wrong... I'll only be able to prove myself when I actually build it.
TylerDurden
100 GW
Direct transfer would be "conduction". The rotor's rate of radiation will not change with a change in stator temperature; the stator's rate of return radiation will be reduced.
As for how much air you can pump through a motor... been to a truck-stop lately?
How's 16,000LFM sound:
Construction Features:
Cover – One-piece, heavy-duty 18 gauge steel cover
is lightweight and virtually indestructible.
Finish – Durable, electrostatically-applied, chip- resistant
white epoxy finish (M), or chrome plated satin finish (C).
Motor – Powerful series commutated through-flow
discharge vacuum motor/blower with automatic resetting
thermal protector. 5/8 HP, 20,000 RPM.
Automatic Sensor – 100% solid state design eliminates
relay contact failures. Microprocessor controlled infrared
sensor turns dryer on when hands are under the air outlet.
Dryer turns off when hands are removed. Override turns
off in 80 seconds if hands are not removed.
Heating Element – Constructed of NiChrome resistance
wire with automatic resetting thermostat. Heating element
operates in the black range for maximum life. Heating
element is enclosed within blower housing and therefore
inaccessible to vandals.
Tamper Resistant – Two tamper resistant screws secure
cover to steel base. Blower air intake is shielded for
additional safety.
http://www.katom.com/155-GXT6M.html
(It sounds like a friggin Boeing 747)
As for how much air you can pump through a motor... been to a truck-stop lately?
How's 16,000LFM sound:
Construction Features:
Cover – One-piece, heavy-duty 18 gauge steel cover
is lightweight and virtually indestructible.
Finish – Durable, electrostatically-applied, chip- resistant
white epoxy finish (M), or chrome plated satin finish (C).
Motor – Powerful series commutated through-flow
discharge vacuum motor/blower with automatic resetting
thermal protector. 5/8 HP, 20,000 RPM.
Automatic Sensor – 100% solid state design eliminates
relay contact failures. Microprocessor controlled infrared
sensor turns dryer on when hands are under the air outlet.
Dryer turns off when hands are removed. Override turns
off in 80 seconds if hands are not removed.
Heating Element – Constructed of NiChrome resistance
wire with automatic resetting thermostat. Heating element
operates in the black range for maximum life. Heating
element is enclosed within blower housing and therefore
inaccessible to vandals.
Tamper Resistant – Two tamper resistant screws secure
cover to steel base. Blower air intake is shielded for
additional safety.
http://www.katom.com/155-GXT6M.html
(It sounds like a friggin Boeing 747)
safe
1 GW
- Joined
- Dec 22, 2006
- Messages
- 5,681
Do you agree with my:Passive Solar vs Active Solar
...analogy?
I think that you can do the heat transfer either way. Using a fan requires additional energy to be supplied which detracts from the (scarce) battery resources. Using Ice Cooling is in effect adding another battery that costs less. Ice is "cheap" because it can be taken from your freezer. Batteries cost a lot of money. Both get the job done. (and there's nothing to stop someone from doing both)
Passive Solar home design is (in my opinion) a better approach than using solar panels to generate electricity and to then have to go back and recreate air conditioning which does the actual cooling.
Different strokes for different folks I guess.(some people use the "bigger hammer" approach while others use the passive use of what's already there)
Let's not forget that the supercooled motor starts off with a big performance advantage too. So there's the "I get to win the race" angle on Ice Cooling over Forced Air Cooling. (frozen motors run stronger)This is the 1016Z3 motor running "stock" at 30A and the chart shows how increasing temperature degrades the motor power. Those extra few degrees of cold might translate to an extra 10% of power... even if it only lasts for a few minutes. :?
Attachments
TylerDurden
100 GW
Not buyin a bit of it...
First, there aint nothin passive about using referigerated ice, and the convection in a motor is forced, not natural.
In passive solar homes, the thermal mass (trombe-walls, tile floors or whathaveyou) serve as heat retention. Plus, very few off-the-grid PV homes would be stupid enough to use any compressor-based A/C. Grid-tied homes may, but they are only using PV as cost reduction.
The ice-cooling concept is used in building systems referred to as "Cold-Stores", whereby costs are reduced by refrigeration being run at night when utility rates are lower.
That chart is totally bogus... the resistivity of copper increases >30% at 100C.
First, there aint nothin passive about using referigerated ice, and the convection in a motor is forced, not natural.
In passive solar homes, the thermal mass (trombe-walls, tile floors or whathaveyou) serve as heat retention. Plus, very few off-the-grid PV homes would be stupid enough to use any compressor-based A/C. Grid-tied homes may, but they are only using PV as cost reduction.
The ice-cooling concept is used in building systems referred to as "Cold-Stores", whereby costs are reduced by refrigeration being run at night when utility rates are lower.
That chart is totally bogus... the resistivity of copper increases >30% at 100C.
http://www.sigcon.com/lib/htm/RESIST.htm
safe
1 GW
- Joined
- Dec 22, 2006
- Messages
- 5,681
MY1016Z3
24 Volt, 350 Watt, 3300 RPM,
19.2 Amp, permanent-magnet motor.
9 tooth sprocket for #35 chain.
12 gauge power leads.
Gear Ratio: 88:9 (9.778)
Resistance @ 0C - 0.191 Ohms Resistance @ 100C - 0.266 Ohms Difference 0.266 to 0.191 - 28.20% (less) Difference 0.191 to 0.266 - 39.30% (more) Peak Power @ 30A @ 0C - 497 watt Peak Power @ 30A @ 100C - 442 wattGot a problem with that? :?
0.191 * (1 + (0.00393 * 100)) = 0.191 * 139.3% = 0.266
(0.266 - 0.191) / 0.266 = 28.20%
0.191 / 0.266 = 71.80%
Attachments
Ben
100 W
Seriously, I'm getting mighty annoyed with this thread.
It just keeps going round in circles without the slightest whiff of a test.
It just keeps going round in circles without the slightest whiff of a test.
safe
1 GW
- Joined
- Dec 22, 2006
- Messages
- 5,681
In the Northern Hemisphere we are in winter right now. It's 13 degrees outside and it's the warmest it's been all day. (it was 1 degree this morning) We have no ability to test now.Ben said:
You are in summer, we are trapped inside and arguing theory. It will flip in a six months...
Link
1 MW
Ben said:
No kidding. I'm this close to getting another MY1020 to fry.
Simple test: Set up a two cheap identical motors. Force air cool one as best you can. Strap an ice jacket to the second. Bog them down with something. Pony brake, fan, I don't care what. Plow some amps through them. The one that lasts longest wins.
You'll have to start them right after you strap on the ice. Otherwise the air one might fry first because the rotor is cold, which wouldn't prove anything except that a cold rotor takes longer to fry than a warm one (duh).
I'm pretty sure I could do this inside.
safe
1 GW
- Joined
- Dec 22, 2006
- Messages
- 5,681
Link said:
Well the "duh" is that pre-cooling the motor with ice is a way to get better performance from the beginning. That's the idea. It should not take more that a minute or two after you put on the ice for the shell to be at freezing. It might take 5-10 minutes before the rotor itself drops to freezing. Now when you start your ride the heat that is produced first has to "undo" all that pre-cooling and only after that can it make an attempt to raise the temperature above room temperature.
At some point a balance between heat production and the cold shell will take place and this will be the final temperature that you have at the end of your ride. If heat production was perfect and linear (which it is not) you would expect a temperature of about 50 C as the maximum possible (in the middle of 0 C and 100 C) compared to the 100 C that is recorded (elsewhere) as the point that most motors fail.
Failure occurs at 100 C Shell is at or near 0 C Rotor is at or near 50 C (max) or 0 C (min)It should work...
It got down to 0 F last night... 0 degree Fahrenheit = -17.7777778 degree Celsius... the very thing that keeps me from riding in the winter might make the bike faster in summer.
Link
1 MW
*head explodes*
safe
1 GW
- Joined
- Dec 22, 2006
- Messages
- 5,681
Motor Pre-CoolingLink said:
If you are simply tired of the topic then you should just ignore this thread from now on.
:? But if you are perplexed then let me say it again...
Basically you can pre-cool the motor with ice. All heat transfer is slow and thermal mass of something like a motor is very large relative to the heat that the motor produces. So any time you deal with heat issues you are dealing with AVERAGES over time.
If you can pre-cool the motor with a long period of pre-ride icing (say 5-10 minutes) then the thermal mass of the motor will drop from room temperature down to near freezing. When you begin your ride the motor now has to apply heat to UNDO all the pre-cooling before it can even get back to room temperature. And even while the motor is trying to heat it's best from freezing up to above freezing the ice is still trying it's best to prevent it from happening.
By the time the motor finally gets hot enough to counteract the pre-cooling and actually manages to melt all the ice and then actually begins to get the motor hot the battery will be running empty.
Ride over. Ice melted. Time to recharge the battery and place your water container back into your freezer to prepare for the next ride.
safe
1 GW
- Joined
- Dec 22, 2006
- Messages
- 5,681
It's Hot Again
Damn... the midwest is either freezing cold or hot and humid. The spring and fall are really very short.
So now my #001 bike with it's 48 volts is getting really hot and losing power based on heat caused resistance. (towards the end of the ride)
Ice Cooling is a great idea I think... I'm going to have to get around to testing it someday. (as if I didn't already have 14 zillion ideas being tested) I might try just stuffing an ice bag onto the outside of the motor to just try it out using my #001 bike.
This stuff takes goobs and goobs of time to develop... it's no wonder I'm about to enter my third year on it all and only have one fully working project and two others partway there.
Oh well... some things cannot be rushed... 8)
It's about 90 degrees now and with full leathers that really seems to amplify it...
Damn... the midwest is either freezing cold or hot and humid. The spring and fall are really very short.
So now my #001 bike with it's 48 volts is getting really hot and losing power based on heat caused resistance. (towards the end of the ride)
Ice Cooling is a great idea I think... I'm going to have to get around to testing it someday. (as if I didn't already have 14 zillion ideas being tested) I might try just stuffing an ice bag onto the outside of the motor to just try it out using my #001 bike.
This stuff takes goobs and goobs of time to develop... it's no wonder I'm about to enter my third year on it all and only have one fully working project and two others partway there.
Oh well... some things cannot be rushed... 8)
It's about 90 degrees now and with full leathers that really seems to amplify it...
Similar threads
