How much interest in a 2.5kg 40kv sensored outrunner

[John in CR]

I'm getting one of these done up for myself for testing, but I'm trying to gauge if others might be interested, since it requires special made shafts for our use. Here's info I have so far.

3 phase brushless outrunner with hall sensors.
Recommended voltage 36V-84V
Maximum current 40-50A
Weight - 2.5kg approximate
Dimensions - 13x13x9cm for the main housing with attachment points and drive shaft extending past the 9cm length of the motor, which is relative to our bikes' width. To me it looks like there's some wasted width from another implementation that could be trimmed.
Kv - 40rpm/V
Slot count 18
Pole count 8

The motor is ventilated with a 50cfm claimed air flow, so I think the current ratings may be conservative for peak, since the same basic motor in a sealed housing runs a 20A controller, but the proof will be in the pudding. Personally I'd like to see it do 60A peaks and 30-40A continuous, since that's the lowest fun tuning of my main bike, though at 9kg less weight I might accept less.

If there's enough interest I'll look into quantity pricing of the shaft I'm having made for mine. What kind of attachment point and maximum diameter would be a good standard, because the bearing inside a hollow shaft means a much larger than typical diameter for RC type rigs?

John

 

[grindz145]

I'd be in man, a fairly simple single stage reduction and I have about 10 projects that would work beautifully with that.

 

[bzhwindtalker]

same here, it seems to be the motor we are all waiting for!

 

[John in CR]

same here, it seems to be the motor we are all waiting for!

Well actually it's been sitting in front of us all along. I've been so satisfied with the performance using my big hubbies, that when I looked around outside of that box it wasn't for long. What I saw was controller issues due to sensorless, and at the higher powers, adding sensors doesn't help because crazy expensive large controllers are required due to the high current. For those who accept mid range power, then a gear reduction is needed which takes up space and adds weight, negating much of the benefit. Plus it's all expensive from my POV.

The answer was always a motor with lower Kv and hall sensors. 2 years on I thought this would be available, but since those systems are designed a different use, there was no incentive for the manufacturers, because the increased weight made it useless for RC.

John

 

[McDesign]

Yeah, I'd be in - got an approximate cost, given a certain quantity made?

Forrest

 

[GGoodrum]

Yes, I think I'd be in for a couple as well. I too have been "waiting" for a sensored motor in this size range with a kV low enough to not require a lot of extra gearing. You can get an AstroFlight 32xx in pretty much any kV, but then the basic size will put a practical limit of about 75-90kV on it, from my experience. I've been watching the progress on the 14kW motor thread, but like most, I don't need that much power. My gut says this is sized right, and I agree, I think the power ratings are under-rated.

I've got a big pile of unfinished projects sitting here, but this would be a drop-in for a couple.

-- Gary

 

[Jeremy Harris]

Is it really practical to get 3-4kW from a motor with such a low Kv that is this small? I don't doubt that it can deliver this sort of peak power for a short time, it's just that the efficiency would seem to need to be very good for it to be able to deliver this much power continuously with such a low Kv, which implies a low operating rpm.

When we looked carefully at the Colossus 75Kv re-wind of the Turnigy 80-100 we discovered that it can't deliver 7kW at any reasonable supply voltage and rpm, unlike the higher Kv Turnigy versions, because the power loss in the copper was just too high at the high current needed to deliver this power.

Operated at the maximum voltage and around 3000 rpm it'd probably deliver this sort of power, but maybe not so much at at any lower voltage, I'd have thought, but I'd be happy to be proved wrong. I've looked at rewinding the similarly sized CA120 motor for a lower Kv, but was put off because of the power loss in the copper.

Jeremy

 

[mdd0127]

I'd most likely be interested in for one, possibly a few too. No idea what for yet but this is addictive and there are literally bicycles everywhere....

I think the size, weight, and low KV sounds great. It seems like a good mid range option for those that want some power but not crazy power.

Sorry if I missed it but is there an ETA on these things?

 

[John in CR]

Ok, so there's at least enough interest for a group buy kind of thing and I front the costs unless it gets too big. Price for any kind of initial run(s) is going to be so low that you might be ill thinking about other stuff you've put on your bike. Let me get the a prototype running first.

I'm looking at a gear reduction of about 5:1 for lower power use at 12s, and 6:1 or more for higher voltage and real hill climbing, so what kind of sprockets and chain combo do you guys think? Can a belt work? Bearing requirements are going to push the shaft out past a 30mm diameter unless more expensive machining is done starting with more expensive material, so suggestions are welcome.

I want to end up very low noise and regen braking, so ideally I want to use either a belt or silent chain in an economical manner. This first one will just be a whatever works deal, probably just bike chain.

John

 

[John in CR]

Is it really practical to get 3-4kW from a motor with such a low Kv that is this small? I don't doubt that it can deliver this sort of peak power for a short time, it's just that the efficiency would seem to need to be very good for it to be able to deliver this much power continuously with such a low Kv, which implies a low operating rpm.

When we looked carefully at the Colossus 75Kv re-wind of the Turnigy 80-100 we discovered that it can't deliver 7kW at any reasonable supply voltage and rpm, unlike the higher Kv Turnigy versions, because the power loss in the copper was just too high at the high current needed to deliver this power.

Operated at the maximum voltage and around 3000 rpm it'd probably deliver this sort of power, but maybe not so much at at any lower voltage, I'd have thought, but I'd be happy to be proved wrong. I've looked at rewinding the similarly sized CA120 motor for a lower Kv, but was put off because of the power loss in the copper.

Jeremy

Definitely question marks at this point. By how much does putting a blower on a motor typically increase it's rating? Is a 50cfm flow enough? I haven't seen any blowers on RC motors, but does 80-100 do 7kw mounting on a plane, since I think that would be more similar?

Is stator saturation something I can easily measure, because to me that may be the practical current limit? With it this small and light, I think I can live with 2kw continuous and 3-4kw peaks. With my big hubbie I pull some continuous type duty above 3kw, but that's really only up hills. Being non-hub this one will be geared lower, because I don't want or need 50mph plus on a very light short wheelbase.

EDIT
I changed the title, because I don't know if those current numbers are just peak, which it sounds like they probably are. I came up with the power numbers myself just based on current and voltage, and I don't want to lead anyone down the garden path. The proof is all in the pudding. I have bikes, and I have mountains, and bikes with known performance for comparison. Load my girth on a e-bike and send it up hills, and the truth comes out pretty quick, just like Method's indestructible motor melting controller back in the days before everyone had a true appreciation of what it takes to drive low turn count hubbies.

 

[StudEbiker]

Sounds very interesting for a project I have in my head right now. If the price is right I will be interested as well.

 

[Jeremy Harris]

My best guess is that stator flux saturation may not be the the biggest problem, and that copper loss will be the limiting factor. Generally, high efficiency low Kv motors need a pretty large diameter and a high pole pair count, to keep the current density in the windings within reasonable limits.

The very general rule of thumb is that big diameter = good efficiency at low Kv, small diameter = high efficiency at high Kv

Adding a blower will help cool an inefficient motor down, but there is still the basic problem that the motor will be wasting power from not being as efficient as it could be.

Jeremy

 

[John in CR]

The very general rule of thumb is that big diameter = good efficiency at low Kv, small diameter = high efficiency at high Kv

Adding a blower will help cool an inefficient motor down, but there is still the basic problem that the motor will be wasting power from not being as efficient as it could be.

Jeremy

What's the rule of thumb for medium diameter? I was looking for a pancake motor in the range between RC and hubmotors, specifically looking for moderate Kv. I would have preferred 30 but 40 seems like it will do.

Regarding efficiency, I can only leave it to you experts to help design better motors. I understand that motors like the Astros have to be very efficient, due to fact that they are sealed and their small size gives them limited surface area for heat transfer. I also appreciate that all things equal, of course we want better efficiency, but if there's only one game in down and it's cheap, then I only really care about efficiency from the standpoint that it doesn't burn itself up at the power I want. I don't really care if I waste a few pennies a week more in electricity, while I wait for a better motor. When it does come, I'll just run it to higher power and waste even more giving a higher percentage back to the wind anyway.

John

 

[Jeremy Harris]

I'm no expert, but the mechanics are fairly straightforward and may help explain things, at least as I see them.

Motors are a bit like levers, in that the force that makes them turn comes from the combination of the strength of the magnets (pretty much fixed) and the strength of the magnetic field from the stator (dependent on current and number of turns).

If the motor diameter is increased, then the torque will increase for a given magnet strength, number of turns and current.

As there is a finite limit to the magnet and stator field strength (determined largely by the maximum current that the windings can take without overheating), torque is pretty much proportional to the diameter and length of the motor, as it depends on the total force that can be applied at the radius of the motor. There is therefore a maximum torque that a given diameter and length of motor can produce, which means that for a given diameter of motor power can only be increased by increasing rpm. Limiting rpm will limit power, and no amount of cooling will ameliorate this, particularly as the power loss is proportional to the square of the winding current.

Jeremy

 

[LI-ghtcycle]

Ok, so there's at least enough interest for a group buy kind of thing and I front the costs unless it gets too big. Price for any kind of initial run(s) is going to be so low that you might be ill thinking about other stuff you've put on your bike. Let me get the a prototype running first.

I'm looking at a gear reduction of about 5:1 for lower power use at 12s, and 6:1 or more for higher voltage and real hill climbing, so what kind of sprockets and chain combo do you guys think? Can a belt work? Bearing requirements are going to push the shaft out past a 30mm diameter unless more expensive machining is done starting with more expensive material, so suggestions are welcome.

I want to end up very low noise and regen braking, so ideally I want to use either a belt or silent chain in an economical manner. This first one will just be a whatever works deal, probably just bike chain.

John

Yes! Do this with some form of simple machined adapter that will attach to the outrunner body (much like where a prop shaft would) with bolts that is made from a cylinder of aluminum turned down and threaded for 1 3/8 in standard bicycle freewheel thread, and hollow to allow the shaft of the motor to pass through into a bearing block, so that one could use either 1/8in fixed cogs or FW's with 1/8 bicycle chain for a very strong easy to build and buy parts for drive, and if you do have good luck with this prototype, I would definitely join in a group buy!

Is this something you would be marketing to the public at large eventually or just for us experimenters?

 

[John in CR]

Yes! Do this with some form of simple machined adapter that will attach to the outrunner body (much like where a prop shaft would) with bolts that is made from a cylinder of aluminum turned down and threaded for 1 3/8 in standard bicycle freewheel thread, and hollow to allow the shaft of the motor to pass through into a bearing block, so that one could use either 1/8in fixed cogs or FW's with 1/8 bicycle chain for a very strong easy to build and buy parts for drive, and if you do have good luck with this prototype, I would definitely join in a group buy!

Is this something you would be marketing to the public at large eventually or just for us experimenters?

These motors will end up drive side specific, but I don't really like threaded on a drive anyway, because right side means the force will be to unscrew. Getting a 5 or 6:1 reduction with a FW as the drive sprocket seems near impossible anyway, or are you talking about for the pedal input and using the motor as a jackshaft? Bike threading is a real pet peeve of mine, because it's not standard threading at all, and taps and dies or nuts to fit are nearly impossible to find, and bike specific has caused me some real headaches and expense. Just about every thing on bikes is non-standard and it took a concerted effort by manufacturers to get it that way so you'd be forced to buy from them at greatly inflated prices. As ebikers we should push to end this monopoly and make our DIY lives easier in the process. My attitude about freewheeling in the drive system is similar, because lets put those bike brake pad manufacturers out of business too by using regen braking. Over $10 for those tiny pads that wear all to quick is absurd. Look at the car brake pads you can get for less than $20.

I already sent a request to Ramses. Hopefully they come back with a bargain solution for their silent chain and sprockets. Ebikes represent and incredible market for them. Who wants to hear chain noise as the loudest noise their bike makes? I'm trying to find silent tires for my hubmotor bikes, because that noise is starting to bug me now that my FW is so worn I can't hear it at all. I guess my tires aren't too bad though. I zipped by a cyclist today at about 40mph, and he was coasting waiting for someone, and his fancy carbon bike had the god awful loudest FW I've ever heard, which was far louder my bike at speed and wind noise across my ears. Is that the new thing instead of a horn or bell, you just stop pedaling and a loud ratcheting FW is the "bicycle coming" warning device?

Sorry, couldn't resist the soapbox.

John

 

[LI-ghtcycle]

I understand if you use a LH drive, and I understand that you don't like the propriety parts that makes bikes what they are, but isn't that what we are starting with? I have no real desire to put a FW on any drive chain connected to a motor, but I know that some like that option, if there is just as readily available off the shelf parts around that would get us bicycle 1/8in chain compatible sprockets from just as common, I'm all for it. From what I am seeing especially if there is a bicycle involved in the vehicle, it makes sense to use parts that you can get easily from common bike part sources.

I'm of course biased towards something that would work well with my dual RH drive set-up with the Nuvinci, but there are LH threaded FW's so I imagine there are LH threaded cogs too, but there again, nothing that would be off the shelf that I know of.

I'd also be find with #25 chain sprockets since they are so common on scooters, and I could get some in a local scooter shop, but IMHO, better to visit one shop for all your needs than sourcing things from several.

 

[John in CR]

Using bike parts is fine until you have to make something that doesn't exist. It's just ridiculous that they don't use standard threading and you know they did it on purpose.

I'm sure I'll end up going #25 chain to start, since I have a variety of sprockets and chain, and only need few different sizes of the larger drive sprockets to be able to try any type of gearing combo.

Almost anything is possible, but my approach to a group thing would be minimize the cost of the single specialty item, that output shaft and make it good fit for everyone as a standard. I really think bike type sprockets are out, because on the wheel side it would be big enough to be a major cost and hard to find. That's because the drive side is too big. At the other end of the spectrum the hollow shaft must have at minimum a 12mm ID hole, so I think that means we're looking at a .75" minimum and 1" or larger better. Is there some standard other than FW threading that would be a good generic size, or just include sprockets or pulleys ready to go?

Hopefully Ramsey will get back to me with pricing that makes it too hard to pass up and everyone gets silent chain that is also better than roller chain for a make John smile price. I feel lucky. 8)

John

 

[LI-ghtcycle]

Using bike parts is fine until you have to make something that doesn't exist. It's just ridiculous that they don't use standard threading and you know they did it on purpose.

I'm sure I'll end up going #25 chain to start, since I have a variety of sprockets and chain, and only need few different sizes of the larger drive sprockets to be able to try any type of gearing combo.

Almost anything is possible, but my approach to a group thing would be minimize the cost of the single specialty item, that output shaft and make it good fit for everyone as a standard. I really think bike type sprockets are out, because on the wheel side it would be big enough to be a major cost and hard to find. That's because the drive side is too big. At the other end of the spectrum the hollow shaft must have at minimum a 12mm ID hole, so I think that means we're looking at a .75" minimum and 1" or larger better. Is there some standard other than FW threading that would be a good generic size, or just include sprockets or pulleys ready to go?

Hopefully Ramsey will get back to me with pricing that makes it too hard to pass up and everyone gets silent chain that is also better than roller chain for a make John smile price. I feel lucky. 8)

John

I'm very intrigued by this, what kind of chain are you thinking? Is this something similar to what you find used as a timing chain in a car?

 

[John in CR]

I don't know. I've never seen one. I've never actually held a timing chain either, but we obviously don't need that wide. It sure sounds good to me...quiet...very high efficiency...very low stretch. My only problem with it is that I'd have to actually take care of it to some extent...probably deserves a protective shield.

http://www.ramseychain.com/index.asp

 

[Hillhater]

John, dont waste time with the "silent chain" route.
If you want a proven , quiet, low weight, long life, power transmission system.... toothed belt is the answer.
Chain is cheaper and effective with the correct type and sprocket sizes, but belt drive is the current leader !

 

[MitchJi]

Hi John,

At the other end of the spectrum the hollow shaft must have at minimum a 12mm ID hole, so I think that means we're looking at a .75" minimum and 1" or larger better. Is there some standard other than FW threading that would be a good generic size, or just include sprockets or pulleys ready to go?

1/2" or 5/8" keyed shafts. Can be used with off-the shelf FW adaptors which can be mounted backwards for right side drive FW's or used with any sprocket that takes a 1/2" or 5/8" keyed shaft.

Hopefully Ramsey will get back to me with pricing that makes it too hard to pass up and everyone gets silent chain that is also better than roller chain for a make John smile price. I feel lucky.

Great idea ! Stronger than belts, probably smaller sprockets/greater reduction than belts (14t chain vs 20t belt drive sprockets). Quieter than #25 chain which Matt says is fairly close in noise level to HTD belts. With a housing probably as quiet or quieter than belts.

 

[John in CR]

Hi John,

At the other end of the spectrum the hollow shaft must have at minimum a 12mm ID hole, so I think that means we're looking at a .75" minimum and 1" or larger better. Is there some standard other than FW threading that would be a good generic size, or just include sprockets or pulleys ready to go?

1/2" or 5/8" keyed shafts. Can be used with off-the shelf FW adaptors which can be mounted backwards for right side drive FW's or used with any sprocket that takes a 1/2" or 5/8" keyed shaft.

This is why I brought it up, but I fumbled over the wording. The output shaft is hollow and the material would be too thin for a pipe 12mm ID and 1/2" or 5/8" OD. Those #25 sprockets might work inside the shaft, but I have to talk to the machinist about making that work. It could make a straight pipe work as the shaft, and then maybe the assembly gets cheap enough for the shaft and permanently attached sprocket to be a replaceable part.

Great idea ! Stronger than belts, probably smaller sprockets/greater reduction than belts (14t chain vs 20t belt drive sprockets). Quieter than #25 chain which Matt says is fairly close in noise level to HTD belts. With a housing probably as quiet or quieter than belts.

If #25 is that quiet, there's our easy solution, unless Ramsey comes back with some joy. I'm not as worried about their chain as the deal killer, but instead the sprockets, since those chains will run on ASME standard tooth gears, which sounds both heavy and expensive.

John

 

[MitchJi]

Hi John,

At the other end of the spectrum the hollow shaft must have at minimum a 12mm ID hole, so I think that means we're looking at a .75" minimum and 1" or larger better. Is there some standard other than FW threading that would be a good generic size, or just include sprockets or pulleys ready to go?

1/2" or 5/8" keyed shafts. Can be used with off-the shelf FW adaptors which can be mounted backwards for right side drive FW's or used with any sprocket that takes a 1/2" or 5/8" keyed shaft.

This is why I brought it up, but I fumbled over the wording. The output shaft is hollow and the material would be too thin for a pipe 12mm ID and 1/2" or 5/8" OD. Those #25 sprockets might work inside the shaft, but I have to talk to the machinist about making that work. It could make a straight pipe work as the shaft, and then maybe the assembly gets cheap enough for the shaft and permanently attached sprocket to be a replaceable part.

Sorry if I'm still not getting it. The hollow output shaft has an ID of 12 mm and an OD of either 12.7 mm or ~15.9 mm? A maximum thickness of .9 mm and its going to need one of the following?:

• a slot for a key
  threads
  dimples for set screws

Sounds like a problem.

If #25 is that quiet...

According to Matt noisier than a belt but close enough with 14t drive sprockets that its ok. He also said its quieter than #219. I think if you use #25 it should be restricted to short lengths.

 

[John in CR]

Ramsey silent chain is out. It requires an oil bath.

Why does length matter for #25? Would it bounce more, or just because more links means more stretch? We have to go to the wheel, so the only way to get it really short is to put the motor outside the chain down by the wheel. That could be a great option, USPD style, but too much power to mount on just a single side pair of stays.