Stevil_Knevil
10 kW
Yay math!
Geeks rule.
Geeks rule.
MXUS 3000 Hub Motor - V1 V2 V3
- Thread starter Allex
- Start date
John in CR
100 TW
bataanjr said:
Too old. Venting like that has been proven to create little flow. How many fans have you ever seen that consist of holes in a spinning flat plate?....None. I was one of those big into the hole shapes and angles of cut, because I had success, but it turned out that it was my blades on the interior of the covers that made the air spin enough to get centrifugal flow worth mentioning.
Then when I had motor with no room for interior blades, I had to put my hand almost touching the exhaust holes to feel the flow. Air can't carry much heat, so you need a lot of flow to move much heat, so I added just 6 exterior centrifugal fan blades that extend 1" beyond the perimeter of the motor sheet. Now I get tremendous flow with intake on one side and exhaust on the other with holes and slots small enough that nothing larger than 2mm can even enter the motor.
windtrader
1 kW
It does seem like a pretty simple and light finned rotor could be cast and attached to the side of the housing. Holes matching the fin placement could then be cut into the housing and thereby created a forced air flow into the hub. Exhaust could be simply some holes on the other side, maybe put in places to facilitate minimize restriction or maybe over areas that could use some more flow on that side. It's just a more manufactured way to do what has already been custom fit.
Allex
100 MW
What about car windows. When you open it(even half ways) you get a lot of air inside even tho it is flat. And when you open another window on the other side you have even more. Does it work because the opening is much bigger compared to the drilled hole in the motor?
John in CR
100 TW
windtrader said:
Centrifugal fans don't work that way. What you describe is an axial fan, which built onto a hubmotor would have quite small blades for the rpm. Plus axial fans don't work well operating perpendicular a significant air flow.
windtrader
1 kW
OK. back to old school the scooped ram air style but no cooling when not moving. Plus if grinding up steep hill, probably not enough speed to get enough flow while motor is cranking big amps. couldn't a fan be installed in an induction tube to drive air as required at all speeds and times?
John in CR
100 TW
Allex said:
If you really want to question every statement fine I'm game. While the airflow in through a car window is complex, it is primarily driven by the front pillar deflecting air to the side which creates a long area of low pressure near the front edge of the window. When you open it a little, air rushes out of the car along that front edge to fill the region of low pressure, and air flows in to replace that air along the top rear half of the window edge.
View attachment Air flow around cars.bmp
On our hubbies the circular shapes that are far smaller and don't match up together like the front pillar and the front part of the open window, as will as the air disturbance cause by the tire, combine to greatly reduce the similar effect on a hubmotor. That leaves only the leading edge and trailing edge of the circular holes to act like blades, something for which they are very ineffective, which is why you'll never see any fan consisting of a flat plate and holes.
I used to think that scoops and angled edges of holes would work, and they do, but only to a very limited extent. Interior blades work better and do create some centrifugal flow, as well as work to deflect air at the hottest parts of the stator for better cooling. These all worked to provide a flow I could barely feel with my hand almost touching the motors spinning up at no-load speed. That's when I introduced larger blades at the outside designed to fling air away from the perimeter just like any centrifugal fan, and they throw a tremendous amount of air. This creates a significant low pressure behind the blade, so air flows out of exhaust holes immediately behind the blades quite strongly, which is replaced by fresh air flowing into the intake. At the diameter of our hubbies it doesn't take very many blades or very high rpm to create a good centrifugal flow. A smoke test showed my latest attempt to clearly draw air through the motor above 10-12mph.
John in CR
100 TW
windtrader said:
You can use deflectors and scoops to direct air toward the motor, but mounted on the spinning motor they aren't very effective due to limited size, and the way they move through the environment. Plus the larger you make scoops or blades to make them more effective at forcing air into the motor, the more likely they are to send sand and grit into the motor.
On my latest iteration, if it wouldn't have been totally effective at keeping temps down, the next step was to be an air dam on the intake side to increase air pressure and reduce air flow velocity perpendicular to my numerous very small intake holes and slots. This should significantly increase flow through the motor, because the ease of intake has a tremendous impact on the flow of any centrifugal blower.
Absolutely, any passive system, will have limited flow while stopped or at very low speed. Very little power is required at low speed, so it's the accelerations from a stop or from very low speed that cause heat to build up rapidly in the motor. Some kind of filtered or screen blower is sure to work well, and some have even put fans in the motor between the stator spokes. Luckily I haven't needed to go to that extent, because I don't want to hear the noise. I found even the noise from tiny blowers to get cooling air through controllers to be bothersome, much less something like the 70W leaf blower that Toolman2 used to great effect to blow air through an RC outrunner. For low speed, I find it sufficient to make a conscious effort to try to keep speed as steady-state as practical. Just like controller settings, how you ride can take huge chunks out of the heat you generate at low speed.
John in CR
100 TW
bataanjr said:
Anyone who wants a bit of help designing an approach that's very effective, just send a pm until I do my next motor and post pics and video. Do a calculation on how much air flow is required to move a given amount of heat at a certain temperature, and you quickly realize that you want to go way beyond something with limited flow. Plus if you can draw a good pressure then you don't have to open the motor too much to the elements. A small effort can go a long way compared to about the same effort of simply drilling holes.
Allex
100 MW
John if I understand your text correctly we need to do something like this to "create a significant low pressure behind the blade, so air flows out of exhaust holes immediately behind the blades quite strongly, which is replaced by fresh air flowing into the intake".
The only problem would be Rear caliper in the way, so probably something like this if you want to fit a brake there?
The only problem would be Rear caliper in the way, so probably something like this if you want to fit a brake there?
Attachments
macribs
10 MW
- Joined
- Jul 22, 2014
- Messages
- 3,702
For me it seems that battling with various way to air cool closed hub motors is not time well spend.
At best you might get some extra cooling - but at the same time you open the motor wide open for the elements.
If you really are up for putting a hub to the max there is only one solution to remove heat - some sort of liquid.
We have seen a few do oil filled hubs. Some with good effect others not so much when you look at the problems oil created.
Then we have Linukas and a few other that have taken liquid cooling a step further. With heat sink inside the hub, a pump and a radiator. Have a look at his simple approach. And he has done liquid cooling of various motors in the past, including cro motor. So my guess is that he is on tested ground when he publish a picture like this.
A simple copper coil inside the hub, attached with some clever adhesive. That might be all you really need.
For this motor in particular I recon the numbers sold here on ES alone is so high no that it should be possible to take an even more serious approach. Take one motor apart and make a 3D drawing of the motor and then contact manufacturer of real heatsinks. Like computer grade heat sinks. They deal with extreme amount of heat that need to be removed during over clocking and benching. so those people know how to make liquid flow inside a heat sink in the best possible way for maximum heat capture before circulating through the radiator.
That heat sink could have a really large surface area compared to those pipes used by Linkus. All of this marked area could in fact be a heat sink.
If someone organize a group buy I am sure some shops will do the CNC work. You would never be able to overheat again. Even after several red light runs in row, up steep hills and then a complete dead stop for 5 minutes waiting for traffic the motor would be cool due to circulating pump, radiator and a radiator fan.
No more melted Halls, no more burned wires or de-magnetized motors due to over heating.
And even if you don't get turned on by this idea - do as Linkus does. Use pipes and radiators. Remove the heat as fast as possible.
At best you might get some extra cooling - but at the same time you open the motor wide open for the elements.
If you really are up for putting a hub to the max there is only one solution to remove heat - some sort of liquid.
We have seen a few do oil filled hubs. Some with good effect others not so much when you look at the problems oil created.
Then we have Linukas and a few other that have taken liquid cooling a step further. With heat sink inside the hub, a pump and a radiator. Have a look at his simple approach. And he has done liquid cooling of various motors in the past, including cro motor. So my guess is that he is on tested ground when he publish a picture like this.
A simple copper coil inside the hub, attached with some clever adhesive. That might be all you really need.
For this motor in particular I recon the numbers sold here on ES alone is so high no that it should be possible to take an even more serious approach. Take one motor apart and make a 3D drawing of the motor and then contact manufacturer of real heatsinks. Like computer grade heat sinks. They deal with extreme amount of heat that need to be removed during over clocking and benching. so those people know how to make liquid flow inside a heat sink in the best possible way for maximum heat capture before circulating through the radiator.
That heat sink could have a really large surface area compared to those pipes used by Linkus. All of this marked area could in fact be a heat sink.
If someone organize a group buy I am sure some shops will do the CNC work. You would never be able to overheat again. Even after several red light runs in row, up steep hills and then a complete dead stop for 5 minutes waiting for traffic the motor would be cool due to circulating pump, radiator and a radiator fan.
No more melted Halls, no more burned wires or de-magnetized motors due to over heating.
And even if you don't get turned on by this idea - do as Linkus does. Use pipes and radiators. Remove the heat as fast as possible.
kiwifiat
100 W
teslanv said:
There are no myths in science or maths just plenty of opportunity for misinterpretation and errors.
Lets agree that the torque constant for two electric motors that differ only in the number of turns will be higher for the motor with the highest turns. This is only a myth if the Lorentz force equation is false, which it simply is not and it can be easily verified with a strain gauge, a magnet, a length of wire and a variable current source.
So can we all agree that if you drive both motors with the same amps then the motor with the highest turn count will produce the highest torque. And it follows that in the same diameter wheel, the higher turn count motor will produce the higher tractive force at the ground since F = T/r. No myths here just irrefutable verifiable facts.
You can argue that you can push more amps into the lower turn count motor so that it produces the same torque as the higher turn count motor did at a lower current and that is quite true. It is also true that if you push the same higher current into the higher turn count motor you get more torque than the lower turn count motor. The torque constant is not a myth. You can play around with controller settings and wheel diameters until you go blue in the face but the one thing that remains true is that for the same current and the same diameter wheel the higher torque constant motor with generate the highest tractive force.
The takeaway from all of this is that if you ride around a hilly terrain you are better off with a higher turn count motor. You can be sure that the Kv of an electric motor has nothing to do with the size wheel you put it into and equally Kt has nothing to do with the size of the wheel you put it into.
ohzee
100 kW
+1 for the smiley face.
Sounds interesting just not sure if id want to deal with a radiator and tubes.
You'd need to run that to the front of the bike to get idea airflow would you not ?
Still while my needs are pretty simple I find it very interesting.
Sounds interesting just not sure if id want to deal with a radiator and tubes.
You'd need to run that to the front of the bike to get idea airflow would you not ?
Still while my needs are pretty simple I find it very interesting.
macribs
10 MW
- Joined
- Jul 22, 2014
- Messages
- 3,702
ohzee said:
Well really it is less of a pain then you might think. Say you have a bike with custom frame and a battery box.
You can run inside piping via batterybox to hub, radiator can be mounted where motorcycles have the headlight. Or tuck your controller inside the batterybox and put your radiator where the controller used to sit.
Pumps are cheap and almost bullet proof. Pumps used for computer overclocking is constructed to run 24/7 for years. Many uses aquarium pumps - they are like 10 $ and up. Radiator and if you like add a fan to the radiator. I would say that a setup like this is less vulnerable for failure then a heavily modified hub motor with pierced side covers. Dust, water and what not just waiting to make problems. If not a sudden overheat issue releated to poor ventilation renders your magnets useless before that.
A custom made heat sink would for sure be most efficient but it would also add some extra costs, plus the pain of a yet another group buy. But hey, if someone know a bloke with a CNC machine aching for more business maybe a good 3D model of motor is all he need to go ahead and make to order.
I've never seen Linukas approach used before, and the first time I saw that my thoughts was those of an computer overclocker. I saw so many places where things could be done better. Then I read something about efficiencies. Most motors are efficient in the high 80's to the mid 90's. Of all your battery power hardly none goes to generate heat - and that mean it is really not a vast amount of heat to remove, for liquids that is. As many here can testify to air will struggle to remove heat build up. It's just a matter of moving that access heat in an orderly and efficient way, and fluid material is great for that purpose. Sous vide anyone? And thinking it through a few more times I am certain that even an easy and low cost approach like the copper tubing inside the hub will remove way more heat then what will possible with ventilation and piercing. Air is just not a good conductor.
The only 2 other things that I can think of that might also work to some degree is 1: Ride better, less heavy acceleration and use your legs up steep hills. 2: Fill that hub with oil and deal with the leakage.
kiwifiat said:
I don't think anyone would disagree with that. However, one of the basic tenets of the argument that has been raging is that the current source is not fixed - you push more current into a low-wind motor (at a lower voltage, producing the same amount of work). Or, relatedly, reduce wheel size and increase voltage.
It helps me to think it terms of amp-turns rather than current in amps. It's amp-turns that produce magnetic flux, which produces torque. The only limit (until the stator iron reaches magnetic saturation) is burning up the winding, which is caused by a lack of copper (thickness times length). Ergo, whichever winding has the most total copper wins.
I like those technical Battles of "iam and you are not" 
My solution about efficiency?
If my Battery is empty, i'll charge it.
If i need more Juice for better Range, Update your Battery Capacity
So much (MXUS3000)discussion Offtopic again on How-is-the-best
and the conclusion is always the same....
Theres not Motor Outside to fit all requirements.
But the Best comes, when some People (noobs or not) ask between the Offtopic Battles something about relevant to the Topic and no one answers.
Has anyone built in this MXUS 3000 in a normal 135mm dropout?
My solution about efficiency?
If my Battery is empty, i'll charge it.
If i need more Juice for better Range, Update your Battery Capacity
So much (MXUS3000)discussion Offtopic again on How-is-the-best
and the conclusion is always the same....
Theres not Motor Outside to fit all requirements.
But the Best comes, when some People (noobs or not) ask between the Offtopic Battles something about relevant to the Topic and no one answers.
Has anyone built in this MXUS 3000 in a normal 135mm dropout?
John in CR
100 TW
Allex said:
Allex,
Your paper blades are turned the wrong way. You want the exhaust holes immediately behind the blades, where the lowest pressure is. You have some issues making them stick out the side like that:
1. They're too small to move much air.
2. They get in the way of the brake caliper
3. Air can too easily rush in from the side to fill in the low pressure region resulting in little suction to draw air out of the motor.
I'd suggest changing the holes slightly and orienting much larger blades differently so they don't affect the brakes:
- cut the leading half of the holes larger by making 2 cuts through that corner, one at the leading edge of the hole and one at the halfway point, and remove the material between.
- make a long angled very thin cut through the cover just in front of each hole for mounting the blades. This allows the exterior mounted blades to extend inside the motor to force the inside air to spin better and create a faster more turbulent flow at the end windings.
- drill and tap a bolt hole for each blade.
The result would be similar to what I did but better. I used 1"x1" thin angle steel used in roofing to make my blades that are mounted with one side flush with the side of the motor and the other side extending inside the motor, and past the past the perimeter of the cover toward the spokes.
Here's a crude drawing showing one modded hole and cut, along with one with the blade in place.
Here's a picture of mine
I did slots on mine and also made my blade cuts and tapped bolt holes all at the same time. I didn't want to have to open the motor again (I don't like opening motors any more than absolutely necessary, because it scrapes the coating on the magnets and the waterproofing on the stator steel). I wanted to see how much air 100 slots would move, but those tiny effective blades are too small to move much air at wheel rpm. With the motor back together I used thin cardboard to mock up blades in order to shape the part extending inside the motor so they get close but don't touch the stator. The purpose of the side facing part of the angle steel is to make the blade rigid, and block air from the side rushing in behind the blade. I did bend mine out away from the side slightly to expose more of the slot area behind the blade. Next time I will just make 6 or 8 larger holes with an angled blade in front of each. The angle would be so the larger hole isn't exposed for something to get in, especially while parked. Near the blade tips I would curve them forward for better flow.
The textured metal of the angle I used works like a lock washer, so it's just bolts, blades, and side of the motor, and in over a year not a single bolt has loosened. The blades fit the slot tightly at the bend in the angle, so they don't rattle or make any noise. There's just a bit of "whoosh" of the air that I can hear only when spinning it up no load at 2krpm.
John in CR
100 TW
kiwifiat said:
Kiwifiat,
You're mixing facts with fiction trying to confirm a myth. It is invalid to compare two different windings at the same current, because the slow wind motor you're trying to say is better on hills cannot handle the same current as the speed wind. For the speed wind you simply lower the voltage and increase the current for the same power input and rpm, and then torque and efficiency will be the same, ie same torque and rpm and same amount of heat produced.
If what you myth promoters say was actually true, then speed wind motors wouldn't even exist. The torque limits of our hubmotors are set by the iron and the magnets, and have nothing to do with the copper. How the copper is wound only matters to the extent of the voltage and current you want to run. As long as the total copper is the same, then they can make an identical amount of torque for the same amount of heat loss in the copper. Saying that a slow wind motor makes more torque at the same current is irrelevant, because in that comparison the slow wind makes more heat to produce that greater torque. Apples and oranges comparisons simply don't cut it. Your understanding is incomplete, because you can't ignore copper losses.
The only ways to make a hubmotor better able to climb a hill does not include rewinding it with more turns on each tooth, which accomplishes nothing, because you decrease current handling by the same amount you increase the torque constant. Decreasing wheel size, and improving heat rejection are the only things that will help a motor climb hills better.
Kingfish
100 MW
John, I struggle to believe the stuff you write. You have mixed metaphors and facts and your explanations do not add up. Forget trying to explain this myth business… you are not conveying a message that is believable.
You appear to be unable to provide us with Math that proves your theory, yet you mock those of us that can. I asked you to clarify and you ran away. I want to see your math. Show us your evidence where you got this idea that all winds perform the same under the same power and load, and this relationship with copper-fill.
Be concise.
Where's the beef? KF
You appear to be unable to provide us with Math that proves your theory, yet you mock those of us that can. I asked you to clarify and you ran away. I want to see your math. Show us your evidence where you got this idea that all winds perform the same under the same power and load, and this relationship with copper-fill.
Be concise.
Where's the beef? KF
Merlin said:I like those technical Battles of "iam and you are not"
My solution about efficiency?
If my Battery is empty, i'll charge it.
If i need more Juice for better Range, Update your Battery Capacity
wow! thats a very good solution regarding efficiency. please also let us know what you do if your motor is overheated.
Kingfish
100 MW
:lol:Punx0r said:Astroflight say it's so and that's good enough for me
~KF
Similar threads
