Help w/Ebike design, Q100, Mac, Bafang?

[cjr]

I've been reading and reading and reading... and learning some, have narrowed things down some, but still need some help. I'm looking to build a pedal assist system with the CA3, and a torque sensing bottom bracket to go on a fully suspended trike. I need it to take a 9 gear freewheel or cassette, and I understand that the Q100, Mac 500W, and the Bafang 36v/500W CST will all do one or the other. I am a women in my mid-50s, in decent shape. Seems that my situation is different than many here, I weigh a lot less than most of you, I always pedal, but I need both hill ability and speed. Performance needs are:

• Weight of trike and rider: 180lbs w/o motor, battery etc. Probably another 5-10lbs of stuff
  Provide some assist up to 17MPH on flat windless terrain w/moderate pedaling
  Provide some assist up to 25-26MPH with moderate pedaling on slight downhill/tailwind
  Able provide some assist and handle a speed of 5-10 mph on a 6% climb for 3 miles, after that it is rolling terrain for the remaining 10-20 miles of that ride.
  Normal Range would be 15-35 miles, with perhaps a spare battery for longer rides. Today's ride was 45 miles, but that was about the longest I think we'll be doing anymore. 15-30 miles is more likely.

I'm attracted to the light weight, stealth look of the Q100.. Just when I rule it out because I don't think it will handle the climb, I see a post making me think that maybe it could. Maybe someone with real world experience with this motor can shed some light. If I'm understanding all of this right, the 328RPM version @48V with the right controller would handle the speed requirements, but would probably burn out going 5-8MPH 6% grade, even with me pedaling? The slower 201RPM version might handle the hills, but not the top end speed I need, even at 48v? If that is the case, then it seems that I will have to go to the MAC or the Bafang. If the Q100 would handle the 3 mile hill, another option would be a dual setup. Get the Q100 for all those regular rides of 15-30mi. if it could handle the 3 mile 6% grade (or maybe the new Q140 which takes a cassette).. and get the Bafang which I could equip with the same cassette, and swap out the wheel when I need more umphf. How does the Cycle Analyst play into all of this? If it is controlling the output to provide the needed assist, would that allow the Q100 to handle the hill? I would surely appreciate any help I can get... I am frankly getting a little burned out and ready to just order a MAC kit from EM3EV, and would probably have done so if I didn't think the Q100 was so damm cool. Profiles of a couple of our rides below. Thanks in advance for any advice you can offer.

 

[spinningmagnets]

Geared hubmotors have a poor heat-path when shedding heat, so if you are set on one of these geared-hubs, I recommend getting one of the large ones (MAC/BPM). the CycleAnalyst has many great features, but I feel the most important one is its ability to apply an adjustable cap on the the amps you are sending to the motor (allowing it to run warm, but never so hot it becomes damaged). You are doing well to go right to 48V with your system design. Higher volts can add more power with very little addition of heat.

Choose the highest top-speed that you want and select the turn-count of motor that provides that at 48V. They are often listed at their rated RPMs using 36V, but at 48V they will spin approximately 25% faster. If the driven wheel is a 26-inch, I think the 10T MAC (code-10 BPM) will provide the 26-MPH (loaded) you mentioned.

 

[cjr]

Yes, I forgot to mention that, this is on a 26" wheel, and since it is a trike, Rear drive is certainly the easiest to go, though I did see something about someone doing dual front motors.. not going to go there, this is complicated enough as it is!

 

[dogman dan]

6% is not particularly a problem. But most hubmotors will hate doing it at 5 mph. If you plan to ride that slow, you want a very slow winding. Mac 12 t for example. A "normal" winding needs to stay well above 10 mph, generally 15 mph minimum climbing hills or they overheat. The same pedaling effort you use to go 5-6 mph up 6% generally does the trick to keep up the speed on 6%. Just wait for your friends at the top, then continue. Once back on the flatter, or less steep rolling terrain, you can ride slower no problem. It's just a problem to go too slow up hills at max throttle.

A wattmeter is really helpful for getting up steep hills, you quickly learn how to get up the hill with the least watts, while keeping up a reasonable speed for the motor. Too slow under heavy load just makes heat. Somewhere there is a sweet spot for each hill. For some it's just full throttle, others it might be 3/4 throttle and pedal briskly. The basic idea, is to get up the hill with the least load, yet stay above 12-15 mph. The mac 12t would be able to tolerate 8-12 mph up the hills. If the Mac 12t will be too slow for your other needs, just give it more volts. 48v 20 amps should be plenty, then you will have 18- 20 mph on the flats I think, yet still climb hills cool. If you need above 20 mph on the flat, then get the 10t motor and run 48v. If downhill, the freewheel will allow the bike to go as fast as you have gears to pedal it without the motor.

 

[TheBeastie]

I'm looking to build a pedal assist system with the CA3, and a torque sensing bottom bracket to go on a fully suspended trike. I need it to take a 9 gear freewheel or cassette, and I understand that the Q100, Mac 500W, and the Bafang 36v/500W CST will all do one or the other.

Only the Bafang 500W "CST" will take a 9 speed, unless you also want to swap out your Derailleur, chain and chankset I believe. Also if you want to use a CA you will need to source a quality controller, the basic Chinese stuff like the KU.. controllers are not compatible.

 

[cjr]

Thanks for the help... swapping out the crankset etc. is not an option... so I'll have to go w/MAC or Bafang CST..

• Need a better controller to go w/CA than the typical, will the Infinion do it? He sell it w/his kits and markets the CA to go with them...
  Any set up that gives me the speed I need on the flats is going to be unhappy going 5mph up a hill
  An over-volted slower wind will handle 8-12mph, and help w/top end, but may not get me the top end that I am looking for

What about short periods of time at the slower speeds... i.e. rolling terrain? Looking at the stats from our ride yesterday, most of it was above 10mph, average speed was 15mph over 45 miles. On that ride there are just a few short sections that slow down.. but I guess that keeping up the speed and waiting for hubby in that situation isn't a problem. The whole idea of this has been to ride with my husband, who is also riding a trike... hence the 5-6MPH on the steepest sections of the long hills.. I guess if I have to choose, I'll take speed over the hills. First, seems that nothing in a geared hub motor is going to handle 5-6mph for any extended time anyway, so that leaves me riding to the top and waiting for him, or letting him head out first and I catch up. I guess I am leaning to a MAC setup.. it's seems more complete/plug/play, I can get the CA3 there, lots of choices in the winds and controller... I don't see a kit at BMS that uses the Bafang CST...

 

[cjr]

I relayed all of this to hubby.. his comment was "there are a lot of hills we ride that I can't maintain 10-12 mph"... He is riding a 45lb trike.... But, these sections of rides are not 6% for 3 miles. I'm guessing that we drop into that 8-10mp range at 2%-3% grades. I pretty much get how to figure out the top end speed, it's the hill situation that complicates things...Remembering that we ride for fitness, and therefore I am always pedaling, and do not want the motor to do all the work.. I just want it to give me a boost.. Couple of questions:

• 
  What I am hearing is that these motors can handle the hills (depending on configuration), but they don't like doing them slow. They need to go fast or the motor overheats.
  
  Does my pedaling contribution affect the motor from the standpoint of burnout on hills - If we are on a 3% grade, going at 10mph, and I am putting in 50% does that mitigate the overheating problem? Or is it just a function of speed regardless of what I am putting in? I can put in 50% and go 15mph, and I'm OK, but putting in 50% and going 7mph, and my motor is not a happy camper... ?
  
  How does the CA play into this? If the CA w/torque sensor is adjusting and putting a cap on the Amps sent to the motor, does this expand the ability of the motor to handle hills?

I'm guessing on a 6% grade, I might be adding 50%, maybe less. Riding that 6% hill unassisted on a 25lb bike, I'm working real hard (HR of 150-160) in spots to get to 4mph. With the BionX, at geared down to stay at 5mph, my HR is 120. On the "flats", I'm doing more, I'm guessing 75% based on the BionX gauge, and when we have that downhill tail wind, I'm doing 90%-100% of the work. Looking at the meter on the console, on that 3 mile hill the BionX is at 50-75% peaking to 100% for a few minutes in the very, very steepest spots.

Unfortunately, I am not an electrical or mechanical engineer, just a middle age wanna-be women who loves to tinker with stuff like this, and is always getting in over my head, and wants an etrike to replace my failing BionX so I can ride with my husband. But, if the best solution I can come up with means either 1) In order to stay with him at 20-25mph, I have to ride ahead every time he slows below 10MPH, or 2) I can stay with him at 10mph, but whenever he goes over 20MPH, he is going to have to slow down to wait for me.. than... well, it isn't too promising... Again, I appreciate very much all the advice and input... Thanks -cjr-

---------------------------------------------------------

6% is not particularly a problem. But most hubmotors will hate doing it at 5 mph. If you plan to ride that slow, you want a very slow winding. Mac 12 t for example. A "normal" winding needs to stay well above 10 mph, generally 15 mph minimum climbing hills or they overheat. The same pedaling effort you use to go 5-6 mph up 6% generally does the trick to keep up the speed on 6%. Just wait for your friends at the top, then continue. Once back on the flatter, or less steep rolling terrain, you can ride slower no problem. It's just a problem to go too slow up hills at max throttle.

A wattmeter is really helpful for getting up steep hills, you quickly learn how to get up the hill with the least watts, while keeping up a reasonable speed for the motor. Too slow under heavy load just makes heat. Somewhere there is a sweet spot for each hill. For some it's just full throttle, others it might be 3/4 throttle and pedal briskly. The basic idea, is to get up the hill with the least load, yet stay above 12-15 mph. The mac 12t would be able to tolerate 8-12 mph up the hills. If the Mac 12t will be too slow for your other needs, just give it more volts. 48v 20 amps should be plenty, then you will have 18- 20 mph on the flats I think, yet still climb hills cool. If you need above 20 mph on the flat, then get the 10t motor and run 48v. If downhill, the freewheel will allow the bike to go as fast as you have gears to pedal it without the motor.

 

[spinningmagnets]

If you have some way to monitor the heat in the motor and in the controller, you can cap the amps at 25A max, and then go for your usual ride. If the motor and controller had warmth to spare, you can raise the amps to 30A, test again to see how they handle it. Keep going up until your normal ride is getting the components close to too hot, and that would let you know.

On flat land, you can use a 45A limit to get some really awesome acceleration and torque, but after a few seconds of acceleration, your speed would stabilize and the amps drawn would go down, giving the motor/controller time to shed heat and cool off some. If you put a 45A limit on a bike that has a long and steep uphill...it will keep getting hotter until something fries.

A CycleAnalyst is an easy way to adjust the amp limit.

 

[ddk]

gonna state the obvious:
your situation calls for a non-hubmotor solution, also known as a mid-drive motor, where the motor drives "through the gears".
Using the trike's gearing allows the motor to run at it's preferred RPM range while you control your top speed with the gear selector.
While it seems you might not be able to handle such a project, luckily you live in an area where a myriad of e-bike builders resides.
(a slight over-exaggeration, but myriad is my "word-o-d-day")

 

[cjr]

Hi DDK, yup.. that's what my brother told me.. and I quote:

If I were to design an ideal solution for your problem, it would be an adaptation of an rc motor drive system. It would probably have a 2 speed transmission, and a liquid cooled outrunner inside a sealed housing to minimize noise, with a belt final drive. I would figure out a way for it to channel power in every direction, regen while pedaling and driving, regen while pedaling stationary, pedal assist or all motor. It would be made from carbon fiber, titanium and forged 7068-T6 aluminum. Nothing like that exists today. It would cost about 5 million dollars to engineer, and end user price would be 3-4k for just the motor and drive (in quantities of 100,000 or more).


Any way of doing the mid-drive w/o the noise? We live in a rural area most of our rides are in the quiet back country.. and I know me, the noise of the mid drive is going make me nuts. I realize I have a set of parameters that is making it difficult to find a solution that gives me the performance that I want, but at least I gave up the regen, so I've moved a little bit...... *sigh*...

 

[ddk]

first off.
Don't use an R/C motor!

Use either a geared or direct-drive hubmotor for the drive.
This already reduced the gear ratio between motor and chain to something more manageable
The only sound you'll hear is the extra chain.
I successfully use geared-hubmotors for my mid-drives. They're quiet and reliable.
I have a Bafang on one trike and a MAC on the other. However, for redundancy and increased hill-climbing ability I also use front hubmotors. But I lack pedal ability these days whereas "You Can Do It"

While you 'can' develop regen with a mid-drive system, it adds so much stress to the chain and gears, so as to be inadvisable for most systems.
I only see regen useful as a braking device.
On a trike if you need more range, carrying more batteries is not an issue. (except cost of batteries sometimes)

about your brother's suggestions:
While I suppose making a human hybrid vehicle the most efficient machine evah might be something to strive for, without all the suggested razzle-dazzle high-tech unobtainium crap, an e-bike (err, trike) is ALREADY pretty darned remarkably efficient.
(imagine Sarah Palin saying that...)

 

[spinningmagnets]

Since this is for s trike, you have some options that conventional bicycles do not. If quiet is a desireable feature, you can use a direct-drive hub-motor as a non-hub. The 9C motors are a little on the large side, as far as the diameter, but there are other direct-drive hub-motors that are smaller, available from Conhis (if you need a smaller size to fit). http://www.conhismotor.com/ProductShow.asp?id=93 Don't let the smaller size fool you, by giving the motor some gears to downshift to, motor-RPMs can be kept up in the efficient range while the bike has slowed some on a steep uphill. Plus, DD hubs have responded well to air-ventilation holes, or.... also oil-cooling (filling 1/3 of the internal volume with ATF oil).

A MAC or BPM geared-hub would also work great at this, but the internal gear reduction makes a slight cordless drill sound that some find to be undesireable. I would aim for 48V and then adjust the amps to find the minimum required to give you acceptable performance when tried on your typical uphill coommute.

By giving a non-hub motor some gears, the lower amp-draw can provide exceptional performance from a small 6-FET controller (perhaps using the cool-running 3077 FET, limited to a 60V battery max. 4110's are popular because they are limited to 100V, but run a little warmer than 3077's, per a given wattage drawn). If you have the room on your frame and the budget, a 9-FET or a 12-FET will run even cooler per a given average wattage.

So few of us here have trikes, that there's not a lot of chat about them, but Rassy is getting incredible performance from a 9C mounted behind his seat.

Please note that Rassy has the mid-drive motor use the BOTTOM of the freewheel to add power to the chain...

http://endless-sphere.com/forums/viewtopic.php?f=28&t=18606

Same guy, different trike:

http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=47043

 

[23skidoo]

I'll be the contrarian here:

You need something as simple as possible. Use a Crystalyte 3525 with a sensor less controller. Ilia at ebikessf can help you locally. Use a simple throttle with a kill button. It's big enough to tolerate lower efficiency states like riding slowly without overheating. It's quiet.

Fussy connectors can be a real hassle and this minimizes the number. I've had problems with the gears and clutch on my wife's bike too.

This is the most reliable system.

The main drawback is the weight, which is important if you need to lift the bike on the roof of your car, but isn't important while riding. It doesn't freewheel, but this is only important if you loose power for some reason and even then isn't that different (my Crystalyte coasts about the same as my wife's BMC/MAC).

Don't worry; there's plenty of tinkering involved even with the simplest of options.

 

[ddk]

OPINIONS TIME :lol:
Rassy's trikes are extremely clever single-chain drive designs... but they were NOT designed for 'real' pedaling
(at least REAL efficient pedaling)
he even mentions this in his thread... and we've discussed it a couple of times (he lives up the road a bit from me)
My opinion is: a three chain or two chain system allows for more effective "human" input

You might research "Cyclone" or "GNG" drives to get an idea.
I use delta trikes so my pictures aren't really applicable to you but here's a pic of one of my drive setups, using a NuVinci CVT hub transmission in loo of derailleur gears.

The geared-hub Bafang motor freewheels when not powered, so pedaling the trike with no motor-assist is no different than pedaling an un-powered trike.
Rassy's design changes the chain direction 180 degrees, which adds significant friction to the chain drive.
but as always:
YMMV

 

[cjr]

Appreciate all the imput... OK, will look at these mid-drive options some more. I'll explore all the options. That Conhis hub motor is very cool looking! Had not run across that one yet. On the hub motor approach, my husband and I talked last night and thinking that we could give some on the top end. Without the weight and drag of the BionX, we are thinking that maybe 20-23mph with assist would be OK, and after that I can stay with him in his draft. As for a mid-drive option. My trike is fully suspended, with all the suspension stuff, there are some room limitations, and it would have to be mounted on a non-suspended part of the bike.... See where the BionX battery is located... I tend to end up sitting on the battery. I had started at one point to look at the friction drive setup, (loved what this guy did - so totally stealth) http://endless-sphere.com/forums/viewtopic.php?f=6&t=47139&hilit=new+build+full+carbon But he ditched it for a hub motor because of the noise. Plus in my case it wouldn't work with the suspended rear anyway. Weight has been one of my driving issues for a couple of reason. One is so that when it fails, or I run out of battery unexpectedly, I'm not pushing a 65lb bike myself. The BionX makes it impossible for me to lift the bike into my truck myself. So, if I need it to take to the shop, or drive somewhere to ride... well, it's just impossible. I have to wait for my husband to help me. We rigged up a couple of 2x6's ramps, but it's even hard for me to push the bike up those ramps. It's a bitch getting old... Anyway.. that is why I started looking at the Bafang, Q, etc. in the first place, to come up with a lower weight, more stealth option...

 

[spinningmagnets]

I don't often recommend the BionX, but they do have some features of note. They run very quiet, but since they are a DD motor that is fairly small in diameter, they have had less interest in them these last few years. If you do decide to go with a small non-hub build (like the Conhis motor), the BionX you already have would be a good starting point.

By giving it some gears, there is a marked improvement in performance, while also allowing whatever motor you choose to run cooler than it would when mounted in the wheel.

Rassy's trike definitely has more room behind the seat than yours. so...as much as I like non-hub builds (especially for steep hills), I have to admit you may be stuck with a hubwheel. A MAC/BPM would be lighter and would also freewheel very well compared to a DD hub. I have heard that sinusoidal controllers run quieter than the common E-bike controllers (common controllers have an abrupt on/off power to the motor phases that contributes to the motor "buzz"). They cost slightly more, and there are fewer choices, but one of those might make a geared hub run quieter enough that it would become the better choice, compared to a (quiet but heavy) DD hub.

 

[ddk]

while you're investigating stuff, think about an easily removable battery.
That helps solve the lifting issue as most the weight of an e-conversion resides in the batteries.
If you go with a hub motor, I'd suggest a geared-hub motor over the conhis, mainly because of their weight and secondly because of a geared motor's freewheeling characteristics.
-and yes-
getting old sux

these are my personal opinions and are subject to manipulation as more facts are presented.

 

[cjr]

OK, I should mention that the reason we are ditching the BionX is because it is dying. It's overheating after about 20 minutes headed up "the hill".. I have to stop and let it cool down to finish the last 10 minutes of the climb. Sent it in for repairs, they sent back still not working... charged us $180 ( not including having the wheel rebuilt so I could ride while it was gone)... turns out they did nothing to the motor.. bench tested and said it was fine and sent it back (they told me this themselves)... Anyway, long story short after 2 months of emails trying to get them to fix it, things turned ugly and they terminated our business relationship and are refusing to service it.. they did refund us the money... And that started the search for a replacement.