Science, Physics, Math, & Myth

[markz]

No, I am serious....I have never hit a pothole in my life. However, around here.........few potholes!

 

[Rix]

John it really makes no sense to talk with you since you do not understand what i'm trying to point out.

For offroad, MX motos run a smaller wheel on the rear for better traction, so you've got some learning to do there too. Regarding potholes, sure I've hit a few, but frankly I ride with focus and avoid them, so maybe you should pay closer attention while riding.

Back and forth with you is exhausting. There's nothing I can learn from you, and you think a bit too highly of yourself to learn anything new, so I won't waste any more time.

Now guys, don't be bad mouthing my collections of 5403, 5404 and 5405 all laced to mc wheels :lol: And yah, they are heavy enough they would make great boat anchors.

 

[madin88]

John it really makes no sense to talk with you since you do not understand what i'm trying to point out.

There's not be a single thing you've posted that I haven't understood. Since you don't know how to tune your suspension, I'd suggest getting with MadRhino, who on one of his bikes runs an Xlyte 54xx which weighs about the same as HubMonster. For offroad, MX motos run a smaller wheel on the rear for better traction, so you've got some learning to do there too. Regarding potholes, sure I've hit a few, but frankly I ride with focus and avoid them, so maybe you should pay closer attention while riding.

Back and forth with you is exhausting. There's nothing I can learn from you, and you think a bit too highly of yourself to learn anything new, so I won't waste any more time.

If you have all understand what others and myself was going to bring up, why not agree with us at least with this part?

It is easy to take the highest kV motor and put it into the smallest wheel. what is not so easy is building a bike with good balance between all, performance, riding comfort, appearance
i do not have a frame in mind for high power e-bikes that is made for such very small wheel. if you know of one, please show us.
why you do not answer about the motor and controller? make mention of great stuff but than do not finishing speaking sounds like its fictive.

you say i do not know how to tune suspension and there is nothing you can learn from me?

seriously, you cannot "tune away" the disadvantage of unsprung mass with the settings on the damper. on top of that you have placed a big part of the battery onto the swingarm making the unsprung mass even higher. Thats terrifying! it would be not much different to take a hardtail frame with a damper in the seat post..
It is not only about potholes and yes we usually see them and can avoid them, but what we cannot always spot are small bumps or uneven surface especially in corners this can be very dangerous.

look what yourself wrote: "I've got the 40+lb motor and wheel on the end, and 74V10ah of battery in the swingarm, and the little Fox Float air shock doesn't control all that weight too well. It makes a nice cushy ride, though I have to be careful when bumps team up with curves or turns."
https://endless-sphere.com/forums/viewtopic.php?f=31&t=46898&hilit=hubmonster#p686834
why you mock at me if i bring up the same thing?

 

[John in CR]

Madin,

I was giving you the benefit of the doubt and didn't think you were clueless, but it looks like you get thrown in the bucket with Kingfish. Maybe it's just that you want to argue, and twist things around to keep it going.

Why on earth would I rehash motors that I ordered so cheaply 7 years ago this week? I tried to turn the forum on to them after my testing, but back then no one was interested in motors that for $125 plus shipping came with a controller, DC/DC converter, and motorcycle rim and spokes, since they required some simple mods to fit on a regular bike. If you can't understand how a motor with an 11% narrower stator and a KV of over 16rpm/v outperforms your generic parts 25-45% lower Kv motors running in much bigger wheels, then you don't understand how mid-drives can have high performance running tiny motors.

Everyone will have to be a little more patient about the controllers, but Teslanv forced the issue. I really wanted to get controllers in the US first, but I'll give you a taste. How does $117 sound for an 18fet with irfb4110 mosfets that is functionally better in significant ways than the Infineon / Xie Chang controllers, and is drastically better and easier in terms of programming? I'm still working on how to avoid the extra layer of international shipping for those outside N.America, so the price before shipping may be lower for you guys.

I told you to talk to MadRhino about your suspension issues. Many, if not most, motorcycles have greater unsprung weight, so of course it can be properly controlled. No other shock will fit on my SuperV, but it's a street bike and I have no problem with it riding like a 70's land barge Cadillac. It's only scary because it's so quick and fast, but in terms of handling like any 2 wheeler riding it requires using experience and common sense.

Stop wasting yours, mine, and everyone else's time with your pointless attacks and apples to oranges comparisons. If you want to talk about the effects of undersized wiring and controllers, then start a thread about it, since I'm sure some will benefit. BTW, one of the most elegant ebikes on the forum, Miles' Dahon, has small wheels, so get over your unfounded prejudice. If bigger is better then put your hubbies in 29ers and suffer the consequences of increased heat and decreased performance and efficiency.

 

[Rix]

Cant' we all just get along? Okay, I am being a smart ass. Sorry guys. But, John in CR is right, no matter how you slice it, the smaller the diameter of the wheel is, the more work you get done under load with less amps.

 

[markz]

I'm sold on the idea or shall I say FACTS, John, Rix!

BTW cant wait to see what product you got coming for us, the electric bike community, John.

 

[speedmd]

You got me John. Time to spill the beans on the controller. When can I get one. Set up for 18s.

 

[Arlo1]

Cant' we all just get along? Okay, I am being a smart ass. Sorry guys. But, John in CR is right, no matter how you slice it, the smaller the diameter of the wheel is, the more work you get done under load with less amps.

That's fallacy. It depends on the wind...

 

[markz]

Cant' we all just get along? Okay, I am being a smart ass. Sorry guys. But, John in CR is right, no matter how you slice it, the smaller the diameter of the wheel is, the more work you get done under load with less amps.

That's fallacy. It depends on the wind...

What depends on the wind? Getting along?

 

[John in CR]

Cant' we all just get along? Okay, I am being a smart ass. Sorry guys. But, John in CR is right, no matter how you slice it, the smaller the diameter of the wheel is, the more work you get done under load with less amps.

That's fallacy. It depends on the wind...

What depends on the wind? Getting along?

I'm not sure exactly what Arlo1 meant, but while I understand and agree with what Rix was trying to say, I don't quite agree with the wording.

From the motor's point of view, a given reduction in wheel size is like lopping off that percentage of weight off of you and the bike, and reducing the load of wind resistance by the same percentage...ie the total load presented to a hubmotor. That means it requires less current for the same acceleration for a given motor, or the same current for greater acceleration so less time accelerating for less heat either way.

With our hubmotors, copper losses are the dominant heat source, so if the lower gearing of the smaller wheel makes you fall short in terms of top speed, you simply raise the voltage and there's very little increase in heat in the form of iron core losses that are rpm dependent. You can clearly see this playing around with the hubmotors on Miles' spreadsheet by increasing rpm and decreasing torque by the same percentage. The increase in core losses is far less than the decrease in copper losses for the same performance at hubmotor rpms, so you can actually crank performance up somewhat and still have greater efficiency along with better hill climbing and fun factor.

Of course there's limits, like with everything. You run into practical voltage limits to get the rpm you want, which is where speed wind motors help, and you have practical rpm limits due to the rpm based core losses, as well as visual appeal limits of smaller and smaller wheels (even to me).

There's always the shortcut route to the same conclusion. Why do you think no one is putting hubmotors on motorcycles, but there are literally millions of smaller wheeled escooters with hubmotors? The reason we see them on ebikes is because they can get away with it at the designed performance level due to the light loads and low power demands. Start pushing for performance and heat quickly becomes a problem. A smaller wheel isn't a cure-all, but it sure helps.

 

[Rix]

Cant' we all just get along? Okay, I am being a smart ass. Sorry guys. But, John in CR is right, no matter how you slice it, the smaller the diameter of the wheel is, the more work you get done under load with less amps.

That's fallacy. It depends on the wind...

I know winds have an affect, but allow me to share my real world experiences with 5403, 5404, and 5405 running 17, 18, and 19 MC wheels with OD of 23.4" 24." and 25.2" respectively. With 20s and 70amps DC controller setting, this is what I recorded. 5405 running 17MC wheel, I could only draw 61DC amps, with the 19MC, I drew 67amps. acceleration was the same, top speed with the 17mc was 36MPH and with the 19mc, was 40mph. Also, I did run he 18mc wheel on the 5405 and it drew 67amps, and top speed was about 38MPH. Keep in mind, acceleration felt the same all the way to 35MPH as I couldn't tell a difference at the wrist. Then I ran the 5404 with the same wheel combos. DC amps peak at 72, controller's mechanical limits, on all 3 different wheel sizes. However, the 17MC wheel was slighty quicker right off the line, then the 18, and 19 a tiny bit slower acclereration, as it should be due to the reduction in mechanical advantage. The 17" felt identical acceleration wise as the 5405 running all three wheels. Here's something else that's funny, speeds varied by only 1MPH between the 17 and 18 and 18 to 19 with 5404. Literally with the 17, I hit 44MPH, the 18 was 45MPH and the 19 was 46MPH. I didn't run the 5403 on the 18, just the 17 and 19. Up to about 35MPH, it was slower than the 5404 and 5405 but it kept pulling up passed 50mph (17MC) and 53MPH (19MC). Also, the 5403 drew 72 amps peak during acceleration.

Based on my real world findings, load wise, the 5405 was getting more work done with 17MC wheel over the 18 or 19. Ironically the 5405 was the only motor that didn't pin the DC AMP draw to the limits of the controller. Load inductance factor coming into play. The 5404 did, and the 5403 did max out my controller. The 5404 running the 17MC had the same perceived acceleration as the 5405 with the 17, 18 and 19MC wheels, and the 5403 was noticeably slower off the line over the 5404 and 5405. Infact I would go as far as to say the 5403 felt slower with the 17 then the 5404 did with the 19. Anyway, I contacted Kenny at Crystalyte and asked him why the 5404 and 5405 had close to the same acceleration and the 5403 was a bit slower, his reply was this. The 5403 uses .083 wire and has 3 turn winds, the 5404 uses .083 wire and has 4t winds, and the 5405 uses .065 wire and has 5t winds. So the copper fill between the 5404 and 5405 are close even though the 5404 has one less turn, and both motors have quite a bit more copper fill than the 5403. I know in the grand scheme of recent conversations, I have stepped back wards from the topic of discussion between Arlo, Markz, and JohnCR, but thought some folks would be interested in my findings.

 

[madin88]

Based on my real world findings, load wise, the 5405 was getting more work done with 17MC wheel over the 18 or 19. Ironically the 5405 was the only motor that didn't pin the DC AMP draw to the limits of the controller. Load inductance factor coming into play. The 5404 did, and the 5403 did max out my controller. The 5404 running the 17MC had the same perceived acceleration as the 5405 with the 17, 18 and 19MC wheels, and the 5403 was noticeably slower off the line over the 5404 and 5405. Infact I would go as far as to say the 5403 felt slower with the 17 then the 5404 did with the 19. Anyway, I contacted Kenny at Crystalyte and asked him why the 5404 and 5405 had close to the same acceleration and the 5403 was a bit slower, his reply was this. The 5403 uses .083 wire and has 3 turn winds, the 5404 uses .083 wire and has 4t winds, and the 5405 uses .065 wire and has 5t winds. So the copper fill between the 5404 and 5405 are close even though the 5404 has one less turn, and both motors have quite a bit more copper fill than the 5403. I know in the grand scheme of recent conversations, I have stepped back wards from the topic of discussion between Arlo, Markz, and JohnCR, but thought some folks would be interested in my findings.

the most important question: did you adjust voltage and current settings?

referrting to 4T setup:
on 5T it would be 4/5 x current and 5/4 x voltage
on 3t: 4/3 x current and 3/4 x voltage

would not be easy because you need 3 different batteries additionally you would need to consider that 3T motor will have more losses outside the motor if phase wires are the same.
in the simulator it would be come quite close if you play with controller resistance to see the changes at same battery power. acceleration will change. ok, not much but it will.

about the windings and different cross section of strands: did they have all the same number of parallel strands? i guess it is so otherwise the different cross section of single strands would not make sense..

Many, if not most, motorcycles have greater unsprung weight, so of course it can be properly controlled. No other shock will fit on my SuperV, but it's a street bike and I have no problem with it riding like a 70's land barge Cadillac. It's only scary because it's so quick and fast, but in terms of handling like any 2 wheeler riding it requires using experience and common sense.

thats not true.
almost no motorcycle will have bigger unsprung weight as the 30kg total mass you have behind your damper or generally as most of our ebikes with very large hub.
and again, if you would read and not only write your stuff you would know that the most important thing is the ratio: sprung / unsprung weight. the higher it is the more the wheel stays on the ground at uneven road. a heavy rider with a light hub is best and a lightweight rider with heavy hub is worst scenario.
And now check the ratio on your bike and compare it with a motorcylce. As i mentioned it would not be much difference in suspension if you would have built a hardtail with a damper in the seatpost.
Ohh but i do not know anything about suspension and you nothing can learn from me. Man, you really should not act so ignorantly.

 

[John in CR]

Rix,

The difference was your controller not delivering the enough phase amps for the 5303. The wire size mentioned didn't include strand count and using Justin's simulator the 5303 and 5304 have exactly the same copper fill, because it's easy with the right combo of voltage and current to make their performance and efficiency curves match exactly, which can't happen happen with unequal fill.

Also, using the same controller and wiring there's some the effect that Madin88 harps about endlessly and needlessly.

 

[John in CR]

Madin88,

I really won't waste my time further.

 

[Rix]

Based on my real world findings, load wise, the 5405 was getting more work done with 17MC wheel over the 18 or 19. Ironically the 5405 was the only motor that didn't pin the DC AMP draw to the limits of the controller. Load inductance factor coming into play. The 5404 did, and the 5403 did max out my controller. The 5404 running the 17MC had the same perceived acceleration as the 5405 with the 17, 18 and 19MC wheels, and the 5403 was noticeably slower off the line over the 5404 and 5405. Infact I would go as far as to say the 5403 felt slower with the 17 then the 5404 did with the 19. Anyway, I contacted Kenny at Crystalyte and asked him why the 5404 and 5405 had close to the same acceleration and the 5403 was a bit slower, his reply was this. The 5403 uses .083 wire and has 3 turn winds, the 5404 uses .083 wire and has 4t winds, and the 5405 uses .065 wire and has 5t winds. So the copper fill between the 5404 and 5405 are close even though the 5404 has one less turn, and both motors have quite a bit more copper fill than the 5403. I know in the grand scheme of recent conversations, I have stepped back wards from the topic of discussion between Arlo, Markz, and JohnCR, but thought some folks would be interested in my findings.

the most important question: did you adjust voltage and current settings?

referrting to 4T setup:
on 5T it would be 4/5 x current and 5/4 x voltage
on 3t: 4/3 x current and 3/4 x voltage

would not be easy because you need 3 different batteries additionally you would need to consider that 3T motor will have more losses outside the motor if phase wires are the same.
in the simulator it would be come quite close if you play with controller resistance to see the changes at same battery power. acceleration will change. ok, not much but it will.

.

No I didn't adjust current. But I get what you are saying, I could make the motors perform almost identically by adjusting amps and volts. If copper fill was equal the 3T running 50volts @ 100 amps would perform like a 5T running 80 volts @ 62 amps. I forgot to note that phase wires were all PTFE 11g wires same length. 19.5 inches from the axle to the XT150 plugs.

 

[markz]

So copper fill of 3, 5, 7 (odd) and 2, 4, 6, 8 (even) are all the same, right.

What about the characteristics of going from a 4T to a 5T, or 4T to a 3T?

 

[Rix]

So copper fill of 3, 5, 7 (odd) and 2, 4, 6, 8 (even) are all the same, right.

What about the characteristics of going from a 4T to a 5T, or 4T to a 3T?

On the 54xx series, copper fill isn't the same. I am assuming with the number above you are asking about 9C motor? Going from a 4t to 5t on 20s is really disappointing. the 5T doesn't pull any harder off the line then the 4t, and has a slower top speed, using 17mc wheel. Using 19MC, I can just barely feel the difference off line with the 5T over the 4T launching from dead stop. Its not much either. And 5T top speed is 6MPH then the 4T. The 3 by comparison is disappointing. Its noticeably slower launching from dead stop, but then pull about 13-14MPH faster than the 5T top speed with the 19mc wheel. With fixed voltages input from 16S-20S, IMO the best overall performance comes from the 5404 in terms of acceleration and minimal and compromise for top speed loss.

But its like JohnCR and other have noted, when you factor in that motor characteristics can be manipulated by volts and amps input adjustment, such as if copper fill being equal, the 5403T running 50volts @ 100 amps would perform almost identical to 5405T running 80 volts @ 62 amps. With this example, there really is no high speed, high torque motor. The only notable advantage would be the higher volt set up being slightly more efficient.

 

[markz]

Sure in whatever motor you want Rix, but I got the mxus 3000w.
I guess I was trying to get the difference in copper fill 4T and 5T. And the more the better right?
I remember people chatting about that. Cuz the strand # in front is different. 21x4T and so on. IIRC its not much, 5% to 8%.

 

[Rix]

Sure in whatever motor you want Rix, but I got the mxus 3000w.
I guess I was trying to get the difference in copper fill 4T and 5T. And the more the better right?
I remember people chatting about that. Cuz the strand # in front is different. 21x4T and so on. IIRC its not much, 5% to 8%.

I'm sorry amigo, I see what you mean. I can't answer that question for any motor. I just know that the 5404 uses .083 wire for that wind and the 5405 uses .065 wire. Copper fill based on wire weight is really close to the same per Kenny from Crystalyte.

 

[speedmd]

John brings a very good point to the table on the smaller wheel / most hub motors being geared too tall for their own survival if pushed too hard. Looking at the simulator, it is easy to see that reducing the wheel size will increase the overheat times allowing some motors to do continuous work at high loads. These same motors fed identical controller amperage limits see overheating in a larger wheel at near identical speed effectively reducing their ability to do work or survive.

Wondering if the existing constants can explain this mathematically via several of the existing motor constants or are we dealing with some sort of additional thermal /torque constant for each motor to best explain the motors wind /constructions robustness to stalling /running a taller gear / driving higher torque loads with a certain duty cycle. Would be nice to be able to identify which setup is less sensitive to lugging around if that is part of your daily run.

 

[John in CR]

Wondering if the existing constants can explain this mathematically via several of the existing motor constants or are we dealing with some sort of additional thermal /torque constant for each motor to best explain the motors wind /constructions robustness to stalling /running a taller gear / driving higher torque loads with a certain duty cycle.

That's exactly the kind of stuff that is possible once you get familiar with using Miles motor comparison spreadsheet and those variable input columns. We need to get it populated with every motor we can get our hands on.

2 things to get a firm grasp on are:
-We turn the throttle to ask for torque from the motor and you make copper heat at the rate of phase current squared times X winding resistance. Efficiency goes up as you turn more rpm (until peak efficiency), primarily because the motor is producing more power (torque X rpm = power). Don't forget resistance goes up with temp by 30% or more as we push our motors harder. The torque per amp (Kt) of a motor is directly related to Kv, 9.549/Kv=Torque per amp (Kv in rpm/volt and Kt in Nm/amp), so it's simple to put motors on the same footing if you're comparing their resistance. AFAIC phase-to-phase resistance (the current goes to the motor on one phase and back to the controller on another) is the most important bit of info for comparing motors, and we should all push for that along with Kv as a standard technical detail. Teslanv is one the first vendors to give us that for all the different windings, and I applaud him for doing so.

-Heat from iron core losses go up with rpm, and even occur while coasting. At moderate rpm they're really low as you can see on the spreadsheet, which splits out the 2 components. Where it gets important to look at with hubmotors is getting to high speed, or even moderately high speed if you're after the ultimate efficiency. I say rpm, but it's really the operating frequency, how fast the stator teeth change polarity as the alternating permanent magnets' poles pass by. Thinner stator lams reduce the eddy current losses for a given quality of the steel. Reduced slot and pole count also reduces these losses, which occur as heat. Since copper losses are so minimal at no-load speed, the power used at max no load speed of a raw motor gives us the maximum iron core losses of the motor while under power, which is why no-load current is such vital info. Miles spreadsheet calculates the core losses at any rpm, so we can see what happens if we increase voltage.

 

[Punx0r]

This thread has descended into an straw-man argument of "the higher the KV of a motor (and/or the smaller the wheel), the better". A reduction ad absurdum is then being applied of "Ah, so you claim a 1V/1000A system is better, hmm?"

It's kinda sad.

 

[speedmd]

That is not how I see it Punx. It is more copper / lower resistance is always better than higher resistance in the coils. Certainly you need a reasonable KV that can be handled. The question is, what exactly allows a motor to effectively drive a bigger wheel with out risk of harm to itself? What other things need to be looked at?

Do we sacrifice the amount of steel, making more room for copper? This trading ultimate magnet strength for resistive losses? It would also lower the iron losses at higher speeds. Do we go to fewer poles? Or is it simply a larger diameter stator? Lots to learn here without taking a stand at a very low level of understanding (at least in my case) the real world trade offs.

 

[Rix]

That is not how I see it Punx. It is more copper / lower resistance is always better than higher resistance in the coils. Certainly you need a reasonable KV that can be handled. The question is, what exactly allows a motor to effectively drive a bigger wheel with out risk of harm to itself? What other things need to be looked at?

Do we sacrifice the amount of steel, making more room for copper? This trading ultimate magnet strength for resistive losses? It would also lower the iron losses at higher speeds. Do we go to fewer poles? Or is it simply a larger diameter stator? Lots to learn here without taking a stand at a very low level of understanding (at least in my case) the real world trade offs.

So many variables. One example in what appears to be a bunch of variables is a thinner stator with the same turn and larger diameter hub can act like a wider stator with a smaller diameter hub. One of my friends noted this exact circumstance with V1 Cromotor and the 5404. The V1 cro has a 4T wind with 50mm wide stator and the hub flange of 233mm. The 5404 has a 4T wind, a 42mm wide stator and a 244 hubflange. laced in the same wheel running the same volts amps, my buddy couldn't tell much difference in performance, and the cromotor overheated up quicker at 8kw. Of course that was the V1 which had a stamped steel stator armature instead of aluminum, so the V2 and V3 probably perform better than the 5404 in terms of acceleration and over heating resistance. .

 

[Punx0r]

Adding stator width increases torque linearly, adding radius squares it. Larger diameter = winning.