8.5mph motors sunrise - how much at cost ?

[Yennek]

Ah, but look again at the graph I posted. Say that motor is connected to a controller that limits the current to 0.5. At a stop the red motor will make a torque of 1, and the blue motor will make a torque of 0.25.

Here's something to ponder. Imagine if that green line is a normal 4mph powerchair motor. Now take that motor, and pull the 4mph gearbox and put a gearbox from an 8mph motor on it. How would you expect that graph to change?

[Yennek]

adding a little extra magnetism to the outside can probably does give an improvement everywhere.

That blue line, where you are getting more speed though, it's with less magnetism.

[Burgerman]

Its also unloaded with less torque. We dont drive unloaded unless going downhill.

[Yennek]

Its also unloaded with less torque. We dont drive unloaded unless going downhill.

It's also fully loaded with lots of torque. That green line applies in all cases of the motor spinning between zero speed and the no load max (assuming you have a controller that is powerful enough not to need to current limit at high torque). It doesn't just apply to downhill. And if you change your flux, you'll move towards either the red line or blue line, depending on which way you change it.

So lets work with more flat ground example. No, lets do going up a hill. Let say you are driving up a hill in a powerchair. I'm thinking the numbers here will make it a bit steeper than an ADA 1:12 spec ramp, with a pretty heavy chair and user. It's holding at a steady speed, so the torque needed to hold the speed is equal to the torque the motors are putting out. It's running on 24V. The it has 14" tires, the motors have a no-load rating of 4mph (or ~96rpm), the motor has an internal resistance of 65mOhms, and due to slope, drag, wind resistance, and whatever else, it is doing 90% of that (or 3.6mph/84.6rpm). So how much current will it draw to do that? and how much torque is it generating? and what is the back EMF that is being generated by the motor?

So, they are all related, and the easiest way to understand it is to look at the electrical model of a PMDC motor that engineers use. It's the figure attached.

It has a resistive element(R) to it, an inductive element(L), and an ideal voltage source(E). This all represents the motor. A battery or a power supply or whatever would get connected up to points A and D. For the steady state analysis we will be doing, the inductive element can be ignored. Not that it isn't important, it is, just that it isn't important for these calculations. The resistive element ends up being important and it is also a constant value. The voltage source is the back EMF that the motor generates. It is dependent on two terms, the back EMF Constant (Ke) and the speed of rotation. As covered before Ke is the same as armature constant (Ka) times the magnetic flux (phi).

First we have to calculate Ke. Ke=1/Kv=1/(speed/voltage)=1/(4mph/24V)=6V/mph

We then can calculate the voltage put out by E. It is Ve=Ke*speed=6V/mph*3.6mph=21.6V

Now that Ve is known, we can use ohms law (V=I*R), to calculate the current through the resistive element. And we know what R is, it's the internal resistance of 65mOhms. The voltage we need to use is the voltage across points A and B. The inductor really doesn't matter for steady state modeling like this, so the voltage at B is the same as the voltage at C (21.6V). The voltage at point A is our 24V supply for the chair. So I=V/R=(Va-Vb)/R=(24-21.6)/0.065=2.4/0.065=36.9Amps. And because there is no other places for the current to go, that is also the current provided by the battery (to one motor).

Calculating to find the torque we first have to calculate the torque constant. Which is the same as the back EMF constant, just converting some units. Skipping how one derives the unit conversions, Kt=(60*Ke)/(2*pi), and will give Kt in terms of Nm/A as long as Ke is in terms of V/rpm. So Ke=0.25V/rpm (because we know the tires are 14" and can convert from mph). That means Kt=(60*0.25)/(2*pi)= 2.387Nm/A

Torque = Kt * I = 2.387 * 36.9 = 88Nm (for one motor)

The math all still works out. You can repeat it as many times as you want, at different speeds, and it will always fall on a line that goes between the no-load speed, and the max torque. You can also do it the other way around and assume a torque and figure out the speed. If you change your Ke, by changing the magnetic flux, and repeat the math, you'll see the line move steeper or more shallow, and the current scaling on the left change as well.

[Burgerman]

Most of that seems pretty clear. And I agree.

Except for this bit.

It's also fully loaded with lots of torque. That green line applies in all cases of the motor spinning between zero speed and the no load max (assuming you have a controller that is powerful enough not to need to current limit at high torque). It doesn't just apply to downhill. And if you change your flux, you'll move towards either the red line or blue line, depending on which way you change it.

A typical large powerchair 6 or 8mph motor has a stall current 2 to 3 times the limit imposed by the controller.

[Yennek]

Most of that seems pretty clear. And I agree.

Except for this bit.

It's also fully loaded with lots of torque. That green line applies in all cases of the motor spinning between zero speed and the no load max (assuming you have a controller that is powerful enough not to need to current limit at high torque). It doesn't just apply to downhill. And if you change your flux, you'll move towards either the red line or blue line, depending on which way you change it.

A typical large powerchair 6 or 8mph motor has a stall current 2 to 3 times the limit imposed by the controller.

Okay I'll modify the plot to show it with a controller limit of 0.5 current. See below. The dashed line is where it is limited by the controller to something less than full voltage, in order to keep the current to 0.5.

[Burgerman]

So has the same effect as gearing. This isnt what I see in practice on my RC truck for e.g.

Graph is still wrong. Current isnt limited to .5 of eac motors stall. But to a constant of x amps. So all would have same peak torque?

[Yennek]

So has the same effect as gearing. This isnt what I see in practice on my RC truck for e.g.

Graph is still wrong. Current isnt limited to .5 of eac motors stall. But to a constant of x amps.

Double check the graph. Look at the current scale on the right hand side. The scale for current isn't the same for each line. For each line, the current is limited to the same amount - 0.5.

[Burgerman]

see it. But it makes the graph confusing using 3 scales for current!

[Yennek]

So has the same effect as gearing. This isnt what I see in practice on my RC truck for e.g.

I've not done much with RC stuff, but I have tried driving a friend's car a little bit, and my biggest observation was that OMG it is quick! Especially for someone like me who has always been a little bit of a klutz. It was quick enough, that if I didn't want to spin the tires at start, I had to slowly feed in the throttle. I'm guessing that it pretty much the case with any serious RC car these days. This means that at low speeds, the limiting factor for torque was neither the controller, nor the motors, but rather the friction with the ground. With that being the case, he could have swapped in an otherwise identical motor with less flux (or changed for higher gearing),and have it perform better for most things, unless he wanted to be the burnout king or something.

One thing to watch out for is making sure you are truly doing an A-B comparison with the RC truck. Are you really applying the same voltage when comparing the performance? If you are doing something at less than max throttle, how are you really sure that the stick is in the exact same spot?

[Yennek]

see it. But it makes the graph confusing using 3 scales for current!

You can have either multiple scales for current, multiple scales for torque, or multiple scales for speed. Or make it a 3-D graph if you want to look at 3 variables. Because the graph started out as a torque vs. speed graph, I made the current have multiple scales. I can probably re-draw it another way pretty easily if you think a different view would assist in comprehension. I don't think I have any handy tools to make a 3-D graph.

[Burgerman]

Dont worry I get what you are saying.

[steves1977uk]

Pretty much.

You need to look inside and see where the magnets are though. Or "feel it" for the magnetism with a bit of steel outside through the casing.

Or you will be randomly guessing. And the got to be the right side up! Or you will make it slow and overheat. It actually increases efficiency if you do it correctly.

You use one magnet to find out which is the north south as ithey will repel or attract each other. Or safer to use a compass. Mark with black pen so you can see which is the same face. Stick them on magnetically and see if its faster. Do one side of the motor first. Then the other side of the motor. The chair will drive oddly and turn towards the slow side. Then do the opposite motor!

You could use these? Or any that you think are easier to get in there. Try to cover the whole magnet. Say 4 of these on each side of the motor can. https://www.ebay.co.uk/itm/5-x-Neodymiu ... 1195.m1851

The bigger the magnet, and the more of the magnet that you cover, the better it will work. The hobby motors I use in quadcopters etc are built with Neobdnium magnets and thats why they are so powerful.

Just bought 2 sets of those magnets, will experiment on an old set of motors. Will be fun! Are the magnets inside a motor located between the brushes or near them?

Steve

[Burgerman]

Yennek says it wont work. And the theory he presents seems sound. Yet what I see on the bench is the same as those vids. So who knows!

[Yennek]

Yennek says it wont work. And the theory he presents seems sound. Yet what I see on the bench is the same as those vids. So who knows!

It's been a while since I watched that video (videos? Was there more than one?) but from what I remember, I think I can explain everything he is seeing in the video.

If I took the time to do sort of a point by point of everything in the video and compared it to the electrical model of a motor we have been using, would you want to read it? If so, repost a link to whichever video you'd want me to analyze. It's a rather rainy windy Saturday today...

[Yennek]

Pretty much.

You need to look inside and see where the magnets are though. Or "feel it" for the magnetism with a bit of steel outside through the casing.

Or you will be randomly guessing. And the got to be the right side up! Or you will make it slow and overheat. It actually increases efficiency if you do it correctly.

You use one magnet to find out which is the north south as ithey will repel or attract each other. Or safer to use a compass. Mark with black pen so you can see which is the same face. Stick them on magnetically and see if its faster. Do one side of the motor first. Then the other side of the motor. The chair will drive oddly and turn towards the slow side. Then do the opposite motor!

You could use these? Or any that you think are easier to get in there. Try to cover the whole magnet. Say 4 of these on each side of the motor can. https://www.ebay.co.uk/itm/5-x-Neodymiu ... 1195.m1851

The bigger the magnet, and the more of the magnet that you cover, the better it will work. The hobby motors I use in quadcopters etc are built with Neobdnium magnets and thats why they are so powerful.

Just bought 2 sets of those magnets, will experiment on an old set of motors. Will be fun! Are the magnets inside a motor located between the brushes or near them?

Steve

Check out this video. It shows where the magnets are at about the 7:05 point in the video. There isn't any electrical reason for the brushes to have any relation to where the magnets are located, so it is probably variable by manufacturer, and how they chose to wind the motors. I'd probably just start by putting a magnet halfway between the brush holders and the gearbox at a random spot, and then rotate it around the motor casing until you see something fun. Once you find a spot that gives the most performance change, try moving it closer or further from the gearbox too. It might not make any difference, or it might, I really don't know, but I would be curious to know.

https://youtu.be/taua7C8HqCM?t=420

[Burgerman]

So where will you put them on your 4 pole motors?

Repost vid link 1 https://www.youtube.com/watch?time_cont ... e=emb_logo

Vid 2 https://www.youtube.com/watch?v=Uks_modsd3M

vid 3 https://www.youtube.com/watch?v=EVyEsdWtGTA

vid 3 https://www.youtube.com/watch?v=6GzM56ejfjY

[Frank]

The 8 mph motors for the quickie 700 R are 80 sterling then the 6 mph motors. Are the 8 mph motors stronger than the 6 mph ones or they are just the same?

[Burgerman]

They are rated a few watts higher on the sticker. But who knows why. They look and measure identically and both are the same measured impedance. And same 45mOhm in settings. As far as I can detect only the gear ratio is different. The sunparts site says the 6mph ones are recommended for high torque, heavy users and presumably heavy rehab chairs as that can add 100kg... It advises that torque is lower on the 8mph ones. Why the 8mph ones are cheaper is anyones guess. They should be identical. The only difference seems to be 2 teeth in the gearbox!

[Burgerman]

There are HD 5mph 8kph motors on bariatric chairs. So they will move/turn. Dont worry about watts. Its pretty meaningless to us as it says nothing useful about how a motor performs. Other than heat/best efficiency/cont rating. Not torque or power.

The 8mph motors are 100 cheaper retail. The 6mph motors are much better on rear drive chairs where more torque is needed anyway to turn in place. Better linear reliable control. This torque thing applies to every brand or chair.

13kph is 8 or 8.5mph
10kph is 6.2 or 6.5mph
8kph is 5mph
6kph is 4mph. On a Q700 there are no 2 pole or 4mph motors. You will get 6mph motors on an NHS supplied one that are programmed to 4... Without lights.

[Yennek]

Repost vid link 1 https://www.youtube.com/watch?time_cont ... e=emb_logo

Vid 2 https://www.youtube.com/watch?v=Uks_modsd3M

vid 3 https://www.youtube.com/watch?v=EVyEsdWtGTA

vid 3 https://www.youtube.com/watch?v=6GzM56ejfjY

Video 1:
0:45 - He should use something like a compass here. It would also help him identify how the field lines are oriented on his neodymium magnet too.

1:35 - I actually like the test setup here. Without having some sort of optical tachometer to measure the speed, the scale provides a nice analog. There are some limitations with this method. At very slow speeds, it probably won't have the accuracy to detect the motion, and it also probably gets inaccurate if the tips of the fan blades approach the speed of sound or start cavitation, or if the fan blades start to distort from speed. But I have a feeling it will probably provide a pretty linear relationship of speed to weight during his testing. No complaints here.

1:55 - 3gm of thurst. Okay. Good baseline. Since he's not going to do something like change the air density, any time he has 3gm of thrust, his motor going to be going this speed.

2:19 - Here he still get 3gm of thrust, because the fan is still going the same speed. Why? Probably because the magnetic field lines from the neodymium are mostly perpendicular to the ones in the motor, and basically they don't interact. Again, compass to the rescue.

2:36 - As one would expect, no change, because they are still perpendicular.

3:05 - Same deal here, without knowing how the fields are coming out of the ferrite magnets, they could also be oriented in a way that it just isn't interacting with the fields in the motor. (They also could just be too weak to do anything observable.)

3:50 - Note that here he has oriented the magnets 90deg from how he had them before. This change should bring the field lines in parallel with those of the motor, and an effect should be seen.

4:25 - Why is it rotating the opposite direction? Well, the flux from the neodymium magnets, when coupled into the motor are both stronger and in the opposite direction than the magnets build into the motor. So the motor basically went from positive flux to negative flux. Unfortunately I that plot I was sharing doesn't cover the cases of either negative flux, nor negative speed. The electrical model however still works, and we can do some math with this if we want. He doesn't actually say it in the video, but the reverse thrust his scale is reading, shows a thrust of between -2 and -3 grams. I'm going to call it three for the moment.

So, we can see the effect of Ve still. Ve=Ke*speed, or Ve=Ka*phi*speed.  It's spinning backwards at about the same speed as it was spinning forwards, and we know he had it hooked to the same battery. So, Vt is the same, R is the same, speed is the same but now negative, and the load is the same, I has to be the same. So in our model, that means Ve also has to be the same before and after the magnets were added. Therefore: (note phi(i) is the contribution of flux from the internal magnets, phi(e) is the contribution of flux from the external magnets).

Ve = Ka*phi(i)*3 = Ka*(phi(i)+phi(e))*(-3)

So we can group our terms in that equation together a bit:

Ka*phi(i)*3 =Ka*(phi(i)+phi(e))*(-3)  ---> divide both sides by Ka and shuffle the three around to the front.
3*phi(i) = (-3)*(phi(i)+phi(e)) ---> divide both sides by 3
phi(i) = -(phi(i)+phi(e))  ---> Distribute the negative into the parentheses
phi(i) = -phi(e)-phi(i)  ---> Add phi(i) to both sides
2*phi(i) = -phi(e)

And thus we can see that the model shows that the external magnets have about twice the effect of internal magnets. 

4:44 - He says "they were so powerful they were overriding the the magents in the motor". Looks to be a correct statement.

4:58 - Okay so it goes faster. This tells us a lot about where on the plot we are. I've attached an update version of the plot. New orange line shows a flux of "3", compared to the green (because phi(i)+phi(e)=3 if we put them on in the correct orientation). Because the motor sped up when the flux was increased we know that we started on the green line somewhere between points C and D with the unmodified motor. The magenta line represents the amount of torque needed to spin the fan at a given speed. This is strictly a property of the fan, and the medium in which it is spinning, it doesn't depend on the motor at all. And it's probably not a straight linear line either, it's probably some sort of exponential function. But this drawing tool doesn't have a good curve function, and it really doesn't matter much, since we don't have many real numbers here anyway.

Where the magenta and green lines cross is our starting point (point A), because that's where the torque and the speed of the motor, equal the torque and speed of the fan. The ending point is point B, where the magenta crosses the orange. As you can see, this shows the motor spinning faster, and having more torque, just as shown on the video. But it's only spinning faster because that fan is putting a heavy load on the motor, compared to the capacity of the motor. Percentage wise, it's probably even more load than a powerchair on most hills. Unfortunately the guy doesn't provide us with any other useful data. If he had also had a way to measure the motor speed without a fan attached, we could have confirmed whether the no-load speed actually decreased like the model says or not. And confirmed the observation about how heavily loaded the motor is. Current measurements would have also been interesting to analyze with regards to the speed. But no such luck.

Video 2: No numbers, no data, nothing to analyze, but based on sound, the motor sped up with the magnets in one orientation, and slowed down in the other, but I don't think that was ever a disagreement point.

Video 3: Also no numbers, so same as Video 2.

Video 4: Well there is at least 1 number here but I don't trust it. He has is multimeter set to volts, but is isn't using the common port of the mulitmeter. It's a little hard to read in the video but it looks like he has the red probe in the correct location, but instead of putting the black probe in common like he should to do a voltage reading, he has it in what is probably the high current current measurement port. Now depending on the design of the multimeter, it might still be giving a close to real value, it's hard to know. But it's also only voltage, and because it's constant here, it really isn't useful for any calculations, and this video is no different than video 2 or 3. 

[Burgerman]

OK. But it definitely had significant increase in fan speed witH greater magnetic strength. Once he got them correctly positioned. And a relatively light load like that tiny fan. And thats what I see on my RC truck. Its also what I would expect on a wheelchair motor. The load is quite high on the flat maybe 20A. And like my truck I would expect a significant increase in speed. Very high load on a minor hill, and irrelivant going downhill...

I have one of these. What can I say I am a child. With FPV and long range radio and video digital transmitter. I swapped the can on the 550 motor for an aftermarket one with neobdinium magnets. Its crawls better more torque. And its around 20% faster visibly. Which is why I am confused. Your explanation sounds correct. But doesent appear to map to reality.

https://www.youtube.com/watch?v=q1lJT6OyiGk

[Yennek]

OK. But it definitely had significant increase in fan speed witH greater magnetic strength. Once he got them correctly positioned. And a relatively light load like that tiny fan.

But is it a light load compared to that motor? Based on the information available, I'm leaning towards, no it's not. But clearly that experiment needed better data collection.

And thats what I see on my RC truck. Its also what I would expect on a wheelchair motor. The load is quite high on the flat maybe 20A. And like my truck I would expect a significant increase in speed. Very high load on a minor hill, and irrelivant going downhill...

I don't know what your definition of lightly loaded is, but mine is, a load that allows a motor to run at >80% of the no load speed. Which is also the same as operating at 20% of the zero speed (stall) torque (or stall current). So if your powerchair on blocks say, spins the tires at 6.5mph when at full throttle, that means I'd consider it lightly loaded until it hit a hill steep enough to lug it down to 5.2 mph. So if you have say 45mOhm motors, your stall current should be about 490Amps if your controller is set to a 22V max (it doesn't matter if the controller limits it when actually at stall, it just matters what the theoretical value is if one were able to). 20% of that is still 98Amps. So you may consider 20 amps a high load on the flat, but it's not that high, in this example it's only about 4% loaded.

And there is a good reason why a powerchair is probably operating at around 5% load most of the time, is because in most PMDC motors, peak efficiency is usually found somewhere above 90% (93%, 95% ish) of the no load speed ie. less than 10% of the stall torque. The model I've been using has mostly been mostly ignoring inefficiencies, because stall torque is always 0% efficient, and the near the no load speed are low enough to not have an effect on the basic physical relationships we have been covering (Because really, if you halve the flux, do you really care that the no load speed went to up 2.0x or 1.9x? It went up a lot).

I have one of these. What can I say I am a child. With FPV and long range radio and video digital transmitter. I swapped the can on the 550 motor for an aftermarket one with neobdinium magnets. Its crawls better more torque. And its around 20% faster visibly. Which is why I am confused. Your explanation sounds correct. But doesent appear to map to reality.

I wouldn't be surprised if your RC motors are running very heavily loaded. There is a very obvious size and weight penalty to going to bigger motors that can be more lightly loaded in an RC application. Also, from what I have been seeing, now that you have my youtube suggesting all sorts of RC cars motor things on it, they are getting quite hot! That heat is coming from inefficiencies. So if it is more heavily loaded and are already operating at say greater than 50% load, then yes you would see improvement pretty much everywhere, because you working those motor HARD! (Which is just like the data suggests for the fan video.)

By comparison, the powerchair motors aren't working nearly as hard.

https://www.youtube.com/watch?v=q1lJT6OyiGk

That is pretty darn cool. There are lots of things that I find cool, and there are plenty of things that I'd love to pick up as hobbies, including RC stuff. But there just isn't enough time in the day for all of them..... But I love the portal axles on it. I have a VW with portal axles and it makes the ground clearance so much better when crawling over rough terrain. I've never understood the people who pull them off and slam them to the ground. I'd much prefer to be crawling over the mountains in it than "looking cool" cruising main street. I actually have better ground clearance than some of my friends pickups. Granted they have 4wd......

[Frank]

On the Sunrise form if you choose the 8 mph motors you have to pay 81 sterling more, so they are more expensive than the 6 mph motors. I wonder why they don’t do the 8 mph motors stronger, more powerful?

[Burgerman]

When you have a 6mph geared motor, you reach maximum torque when the 120A R-net controller says you reached its limit. Since torque = Amps proportionally then you would limit the torque to exactly the same level no matter what motor they used that reached 8mph. But the taller gearing means you lose torque in proportion to that speed change.

Most wheelchair motors have a true stall torque that is around 3x or more greater, at 3 or 4 hundred amps. Way more than we ever allow. The limit is 120A from powerchair controller. Many are less. So as soon as this 120A is reached, theres no more capability left.

Take a look here.
This is my R-Net system reporting M1 and M2 (motors) amps reading 113.9A and 102A and all I am doing is turning in place in my hallway on a heavy carpet in the Salsa rwd 6mph chair. Note that if I had fitted the 8mph motors, this current would try to increase by the same amount. So 6.2mph + 30% equals 8mph. And an increase in current by the same 30% on every turn or movement. So that 113.9A would have required 148Amps. The 102A would need 132.2A. Since its not available its turn in place would have either stalled or more likely just slowed down, and the opposite motor would have moved faster as it wanted less current. You feel this as a soggy non linear powerchair that doesent do as its told. Doesent follow the stick directly. Or does but not quite as expected.

This is why I choose 6.2mph motors. Having a more powerful 8mph motor would not increase torque, just runs up against the 120A limit sooner.

[expresso]

but also - you need to mention that if a person is light - it wont be as affected - that person may not feel it as much as someone else - so it depends - for me it feels fine - not a big deal - i dont find myself wishing i needed more bottom end on the 8.5 mph motors with the sunrise chair -

at least at that speed - i dont feel i am missing anything -

[Burgerman]

Its all a big compromise. A chair at double the weight needs double the amps to achieve the same manoever. A 90KG non rehab chair is half the weight of a 190KG Full rehab one with lift, tilt, legs, recline etc. And a 130KG user needs again, double the Amps to turn in place or climb a threshold compared to a 65KG one. My ex was 50KG. She would need way way less motor amps than I do for the same manoever.

And it also depends on programming. If you have the typical stock low turn acceleration of most chairs its still happening but the chair is already set up never to respond properly so you wont feel it. You are stirring a pudding rather than slicing a cake.

All of these things matter. All of them make a difference. I like my chairs to do as I tell them. When I ell them. To me it matters!

[expresso]

i agree - its good to point out the PROS and CONS - but also its different for everyone - programming etc, - basic changes can help every user - but not every user needs extreme programming or even wants it - but what ever works -

i dont feel any real Con having the 8.5 - feels great to me - and i go up hills just as fast as the 6 mph chair so i dont loose much at all overall - even if i drop a bit going up hills its still as fast or bit faster than the 6 going up same hill - yes i may be pulling more amps and using more battery all the time

but who cares - we have enough lithium that its an non issue - with a 6 would last even longer - but who wears out a 200ah pack daily - ? i tried it once with 180ah got 68 miles - just to see - so if you have 6 - maybe you get 80 miles out of it - its pointless - your never going to ride that much but good to know you can

[ex-Gooserider]

Interesting discussion... What I got from the first video was that with the magnets in the right place, the fan gave more lift, and since the blades didn't change, that implies that it was spinning faster... (it certainly sounded faster....)

The power source seemed to be a constant voltage (or current?) power supply, so if a CV supply, the motor needs to be drawing more current (increasing torque so it can spin the fan faster at a given voltage) or if a CC supply, drawing a higher voltage and spinning faster because of the voltage increase at (presumably) the same torque.... If the supply was unregulated, it might have been a combination of both... Either way it had to be due to a lower effective resistance in the motor windings.

It would have been better if the test setup had specified the type of supply and had a meter to see what the non-fixed output was....

Another thought is that the magnets used in the video were pretty close to being the same physical size as the motor magnets.... On a chair motor the magnets are MUCH bigger - at least on the motors I've opened up... This makes me wonder how big a magnet one would need to make a meaningful difference on a chair size motor? (also how much would one need to worry about matching the curve of the stock can?)

A second question is about practicality if trying to do this with a motor on a chair.... The stock motors are inside the motor can, which limits how 'magnetic' the outside of the motor is, so you don't pick up much random debris rolling down the street... If you stuck a bunch of magnets on the OUTSIDE of the motor, would they have a tendency to pick up ferrous debris (I hate to think about rolling around the Asylum welding shop if that was the case....)

ex-Gooserider

[Yennek]

Interesting discussion... What I got from the first video was that with the magnets in the right place, the fan gave more lift, and since the blades didn't change, that implies that it was spinning faster... (it certainly sounded faster....)

The power source seemed to be a constant voltage (or current?) power supply, so if a CV supply, the motor needs to be drawing more current (increasing torque so it can spin the fan faster at a given voltage) or if a CC supply, drawing a higher voltage and spinning faster because of the voltage increase at (presumably) the same torque.... If the supply was unregulated, it might have been a combination of both... Either way it had to be due to a lower effective resistance in the motor windings.

It would have been better if the test setup had specified the type of supply and had a meter to see what the non-fixed output was....

Another thought is that the magnets used in the video were pretty close to being the same physical size as the motor magnets.... On a chair motor the magnets are MUCH bigger - at least on the motors I've opened up... This makes me wonder how big a magnet one would need to make a meaningful difference on a chair size motor? (also how much would one need to worry about matching the curve of the stock can?)

Depending on the quality of the magnets involved, a Neodymium can be up to 20-ish times stronger than a similar size ceramic magnet (but of course could be a lot less than that too). So, you might not need that big of a magnet to have an effect. And of course the metal can of the motor will block some of the effect as well, since, at least on the motors I have here, the case is steel. If you made a motor case out of aluminum, the case itself wouldn't really have an effect on the magnetic flux. As far as curve, ehh, glue it in place if it moves.

A second question is about practicality if trying to do this with a motor on a chair.... The stock motors are inside the motor can, which limits how 'magnetic' the outside of the motor is, so you don't pick up much random debris rolling down the street... If you stuck a bunch of magnets on the OUTSIDE of the motor, would they have a tendency to pick up ferrous debris (I hate to think about rolling around the Asylum welding shop if that was the case....)

Yes, debris could certainly be an issue.

My current pondering on this is how practical it would be to make a hybrid motor. A motor that has both permanent magnets and a set of field coils. Energize the coils adding flux to when starting off to give more torque, and operate most of the time with out the coils energized, because energizing the coils would likely reduce efficiency overall. And then when going for max speed under low-load situation, energize the coil the other direction to reduce the flux and increase the max speed (again with probably some efficiency hit). Would work sort of like a 3-speed gearbox.