Search all of my sites with Google

Monday, September 3, 2007

Design Goals -- Updated 09-13-07

Design Goals:
1-- Inexpensive. Recycled items where possible. Any purchase that must be made increases the likelihood it won't be completed.

2-- Electric-Assist: The motor isn't intended to do all the work, just enough of it to allow me to move more cargo and/or faster than I could by myself, and to have a longer range. It's Phoenix, Arizona--thus most of the time it's hot, and not much fun to bike 10+ miles to work, and having to essentially take a complete shower and change of clothes before I can start working, and again once I get home.

3-- K.I.S.S. It's already more complex than I'd like, but there are features I want that I can't get with the simpler ways I've thought of. Simple operation when completed is more important than simpler design/construction. However, the more complex the design, the less likely it is I'll finish it.

4-- "Automatic transmission": At some point, I want to have the whole motor assembly control itself to match the speed I set with the throttle in the most optimum way as possible. This is in direct violation of design goal #3, but unavoidable in order to get simpler actual operation, which is more important to the eventual goal of the whole thing.

5-- Weight/mass: Needs to be as light as possible, for all the obvious reasons. Batteries are probably going to be the heaviest add-on of all of this, with panniers second.

6-- Safety: Needs interlocks on the motor so it can't run:
a-- if the motor is too hot
b-- if my feet aren't pedalling
c-- if I'm not holding the throttle in position (needs spring return?)
d-- If I'm not on the bike
e-- if the batteries are too hot
f-- if ?? Need other possible failure modes that it should cut out on, and how to make it fail in a safe way.



These are new additions to this post on 9-13-07, after some discussions outside the blog that make it obvious that I haven't made my purposes and reasonings for this project clear enough.


7-- I won't use a kit.
a-- No challenge to me if I do so. Where's the fun in that? :-)
b-- Kits cost too much
c-- Many use inefficient hub motors
d-- Many hub motors in kits appear to be brushless and thus can't even use regen, AFAIK.
e-- I've not researched too many of the ones that use part of the pedal chain to transfer power, but they appear to have the problem that it's possible to get my feet knocked off the pedals by the motor if I am not careful to keep up with it, and that could cause an accident because I'll likely lose my balance, even if the motor includes a cut-off if it goes faster than I'm pedalling--it's too late once my feet are off the pedals. I'm not the most coordinated person in the world. :P
f-- A major goal is to use recycled parts wherever possible. Basically I'll take whatever I can find and make it work, if necessary. I'm a packrat and a pretty good scrounger. Plus I have all of you to help figure this out, right? ::crickets::


8-- Re: objections to using Electric power (vs gasoline, for instance) due to limited range vs heavy weight but cheaper batteries vs expensive lightweight batteries:
a-- Range is not as much of an issue as it might seem, because this is only an assist, and won't be running the whole bike, just adding power when I don't have enough of my own.
b-- It's a little bit of an issue, in that I do need to get at least 10 miles out of it, but once I get all the right parts in there, and learn how to shift for it, it should be efficient enough for at least that. If it's a workplace as my destination, I should be able to recharge there.
c-- Once I learn how to program and build the full version of the microcontroller for all this, it'll shift itself (motor only, not pedal drivetrain), and be even better, in theory (we'll have to wait and see on the practicality of that part).

Batteries and recharging are indeed a sticky point on this, and that will be the one thing that will cost. I have temporary testing solutions, but until I have a working design for the rest of it, the final choice of batteries will not be much of an issue. :-) I have some ideas about adapting various laptop and power tool batteries, potentially with adapters that allow usage of them in their original states, which includes their current, voltage, and thermal protections intact--if I can do this, then I can use old packs that might not give as much range as I'd get with new ones, but would be free to me as people often discard them when they replace them. I've collected many dozens of them over the years and reused the working cells in them to repair other packs for my own devices before. Most were still quite usable, just not necessarily for the device they came from. This is a topic I'll have to do practical research on after I have the rest of the design physically completed, and can actually test out theories on.


9-- Why Gas Won't Work:
a-- I can't stand the smell of it
b-- It costs too much
c-- it pollutes too much directly (yeah, I know, the technology that makes the batteries and other parts for my electric will also pollute, both during manufacturing and after the parts are used up and worn out and "discarded").
d-- I'm not convinced I can repair an ICE correctly, as I've never had much luck even with gas lawnmowers, while with electrical stuff I can even re-wind a motor if I have to (hate doing it, though), and I can design and build my own electronics, etc, where necessary.
e-- Gas and oil present spillage and fire hazards, as well, in case of an accident (yes, so do some battery solutions, especially Li-Ion and Li-Poly).
f-- Too noisy. I'd like a fairly quiet ride, when I just want to go and think on the way somewhere that I will need the assist to get to or from.
g-- ICE appear to be forbidden on some, perhaps all, bike trails and parks, etc, here in the Phoenix, AZ area (valley of the sun).
10-- Not sure of the DMV status exactly, but appears to require registration, licensing, insurance, etc once a vehicle uses an ICE. Don't want to deal with all that. Plus any emissions testing, if any.
11-- I also can't easily design starters and cutoffs and such for ICEs, whereas they're super easy for electrical stuff. Since I intend the entire system to only run when I am pedalling, and cut itself off entirely when I stop pedalling, mainly for safety reasons, then those starters and cutoffs are necessary.
12-- Since the only place I can keep the bike safely is in my livingroom, well, would you want to have that smell in your house, or stains on your carpet from it?


13-- Practicality of Electric-assist bike vs Car:
a-- Carrying power: almost every thing I need to carry, such as groceries, will easily fit on the bike.
b-- Range: Almost every place I need to go is easily within my pedal-only range.
c-- Time: Almost every place I need to go is *at least as fast* to get to on the bike as by car, and often enough is *faster* on a bike during heavy traffic times, even though I am obeying all traffic control points just like cars are, because there are alternate streets and routes that allow easier movement for bikes, without me disturbing the neighborhoods (which would happen if all those cars were driving thru them). For longer distances where freeways are involved, cars will win out time-wise, as well as when they use alternate routes with low or no traffic.
d-- Special arrangements are normally very easy to make for when a, b, or c make a bike impractical or impossible, but those are not usually necessary.
e-- For those who have suggested Motorcycle or a gas-powered bike instead of electric, I would still have to use special arrangements in case a-- was the issue, as the motorcycle or gas-powered bike wouldn't move the things that won't fit on the pedal bike, anyway. Range would be increased to as far as the tank would carry me plus any gas stations, but as per point 12 above there are many reasons gas itself is highly undesirable.


14-- Weight of Electric vs gas: Batteries are good enough even now for me not to have to have the weight of the rest of the system plus rider in batteries, just to get it moving. That's an improvement over when I very briefly considered this once before, about 11 or 12 years ago, I think--the batteries were what stopped me then, as the only alternatives I knew of were NiCad or LeadAcid, neither of which is all that light or energy-dense. Li-Ion came out, but then was so incredibly expensive (some ~1Ah laptop batteries were $350-$500 or more!) that I could not seriously consider even attempting it. But now, it's far far cheaper, and better technology. Good enough for what I want to do, even if not yet optimal. Since I am already needing the assist sometimes, I don't think my aging body will wait for the batteries to get better. :P

4 comments:

  1. Hi there,

    I found your blog because I was searching for information on how to go about putting together home made electric motors and generators, for my own grand 2-wheeled schemes.

    I've really enjoyed your story, and drawn a few good ideas from it too.

    I'm still learning all the technical details about how electric motors function, and I'm surprised to note that you say that brushless motors can recapture braking power. Up until now, I'd been concentrating on brushless, as it appeared a better technology, but without regen, I'm no longer interested. I also note that you /appear/ to be using brushless motors later on (from the ceiling fans). Do you have any more comments on this subject?

    Secondly, whilst I'm now confident on the basics of motors, I'm trying to get more information on what factors affect the characteristics of the motor, and how to effectively control them. I am referring to how one translates from Voltage, Current, Winding properties, and so on, to speed and torque. Do you know of any good texts available on the 'net that could help me?

    Thanks, and keep up the good work.
    Matt.

    ReplyDelete
  2. > I found your blog because I was searching for information on how to go about
    > putting together home made electric motors and generators, for my own grand
    > 2-wheeled schemes.

    You should blog yours as well; no matter how it turns out at any step, it's useful information for others to either know what *not* to do or learn how to do something! :-) Blogspot is free and hosts all the pics and videos, etc, and is easy to use.


    > I've really enjoyed your story, and drawn a few good ideas from it too.

    Thanks! I'm hoping it is helpful to anyone that comes across it. :-) I really ought to put more detail into it, but so much of what I do is guesswork that I don't have a specific recipe for--just trial and error.

    Feel free to (read: Please do!) comment on any and all blog entries with specific feedback or questions, at any time! I'll reply to them as I get them (blogspot sends me copies of the comments so I can approve them, which keeps pure spam off my blog).



    > I'm still learning all the technical details about how electric motors
    > function, and I'm surprised to note that you say that brushless motors can
    > recapture braking power. Up until now, I'd been concentrating on brushless,
    > as it appeared a better technology, but without regen, I'm no longer
    > interested.

    Actually, I think I only said that AFAIK it's *not* possible to do regen with BLDC (BrushLess), only with Brushed DC. If I did say the opposite, please point out where so I can either fix it or find out what I was thinking at the time. :-)

    At first, my attitude was also that I wasn't interested in BLDC if I couldn't use regen, but there's too few things I need regen to do that I can't live without or find alternate means of doing to discard the BLDC idea entirely, even if it can't do regen.

    Now, I think it might actually be possible (though I'm not sure of the electronics needed to do it yet) to recapture power from *any* permanent-magnet motor; it's just probably a lot more complicated with BLDC than regular Brushed DC (which is pretty straightforward, and a standard part of some controllers, like the better ones from 4QD). It may not be possible to do it efficiently enough to make it worth doing with BLDC, and that might be why it's not being done.

    However, BLDC motors are more efficient than Brushed, both because of friction and because they don't have the internal arcing from the brushes to the commutator segments that wastes energy as RF bursts and heat. *Lots* quieter, too. Controllers are more complicated (and at least a little more expensive because of that if you're buying off-the-shelf).


    Also, keep in mind that the recovery of the energy may actually not be such a great thing in a bicycle. I've been pondering back and forth for months on this, and here's some disadvantages to doing it:

    1) Regen will act as a brake, slowing you a lot faster than just cruising to a stop. That may be desirable depending on the style of your ride, but you can't *depend* on it to work, since if your batteries are fully charged already when you try to use it, it may not work as you expect. Or something in the motor or electronics or battery could fail, and you'd be unable to stop. So you have to keep your regular brakes too.

    2) You won't get all that much energy back out of the motor, depending on how efficient your motor design is, the friction of the bearings/etc, and your electronics. Remember, in order to get the energy back, it's going to act as a brake, and slow you down/stop you, as in 1) above.

    3) If you set your motor up to do regen, then you cannot use the regular rightside chain & sprocket drivetrain of the bike with the motor to drive the rear wheel--if you do, then the rearwheel's freewheel will prevent any energy from going back into the chain and thus into the motor. You have to either make a separate drivetrain to directly connect the motor to the rightside of the wheel, or use friction (really lossy and bad in rain), or use a hub motor, all of which mean that the motor and/or drivetrain is causing drag/friction on your pedalling when it is not being used, forcing you to *always* use more energy than you otherwise would personally.

    3a) Adding a separate drivetrain adds complication to your bike, and makes it heavier, thus also taking more energy from both you *and* the motor to move it.

    4) The electronics are more complicated, and thus at least a bit more expensive.


    Some advantages of doing it are:

    1) You get some of the electricity back as you slow down or stop. This might extend your range a bit, but probably not by much; it totally depends on your regen electronics, friction losses, speed you were going (faster is better for regen to work, AFAIK), motor design, etc.

    2) You get "free" braking ability from it. (this might be bad, though, if you don't want to stop that fast).

    3) It's nice to be able to "recycle" the energy you're using up. :-)

    4) In theory, you could charge your batteries back up by riding in regen mode and pedalling, but it will be a lot harder than just riding normally with a typical generator that runs lights; it's a much bigger load physically and electrically and you likely won't be able to generate pedal power very long trying to do this.


    Now, another thing you can do if you use BLDC as the main motor is put a regular Brushed DC motor on there as a generator only, or even as a motor/generator (like secondary boost power), so you're using the most efficient one (BLDC) most of the time, and only physically connect the Brushed one when braking, or when needing an extra boost (long uphill climbs, for instance). It's more weight, but you can use a much smaller Brushed motor (100-200watt, maybe less) than otherwise, and still easily get enough power back out of regen that you could justify it's being there.

    One thought I have had along those lines was using the BLDC as a hub motor (in the rear, probably, due to risk of fork destruction in front, as well as steering problems from weight), and using a physically-disengagable Brushed motor for regen and braking; it would be cable-actuated with the brake lever to friction-engage the rearwheel treads during braking. I have a small Toyota Tercel radiator fan motor that I experimented with friction drive with for a short time, but it was too small to use for a drive motor--it would however make a nice regen-only generator. This will probably be in one of my tests after I get magnets for the cieling-fan motor conversion.


    > I also note that you /appear/ to be using brushless motors later
    > on (from the ceiling fans). Do you have any more comments on this subject?

    The cieling fan motor is indeed being turned into a BLDC motor; I haven't yet gotten better magnets for it than those salvaged HDD magnets you see in the video, and am still working out a controller design (trying to do one that doesn't require a microcontroller, as I have no real skills at programming them yet). Whatever I build for the project I'm trying to use only recycled parts, so also trying to work within the constraints of the parts I salvage off of old or dead electronic items (I have many smaller FETs from motherboards, for instance), though I will probably have to use some new MOSFETs I got as sample parts from various electronics component manufacturers.


    The entire design of the cieling fan motor aches to be a hub motor; it's probably the first thing I'll try to do with it once I have magnets/controller for it, and rewind it's coils to match the magnets.


    > Secondly, whilst I'm now confident on the basics of motors, I'm trying to
    > get more information on what factors affect the characteristics of the
    > motor, and how to effectively control them. I am referring to how one
    > translates from Voltage, Current, Winding properties, and so on, to speed
    > and torque. Do you know of any good texts available on the 'net that could
    > help me?


    I mostly got the idea to do it after accidentally running across information on people who have been converting old CDROM motors (already BLDC, but very low power/speed) into motors powerful enough to fly RC planes with; then finding some RC enthusiast forums (can't remember the names right now) where there are posts, tools, and other info that I can adapt to my purposes.

    http://www.bavaria-direct.co.za/models/motor_info.htm is a good place to start for the info on how to calculate these things out. What you don't see there you might find links in the http://www.rcgroups.com/forums/index.php forums to; I think that's where I found what info I have (I didn't write down where I got the info from, just saved the info locally as I found it).

    There's also a specific way of winding a BLDC motor called LRK, and has a mailing list dedicated to it: http://groups.yahoo.com/group/lrk-torquemax/


    Since RC planes are looking for high-speed and low-torque motors, I have to swap factors around to get the motors to work for my purposes, which is low-speed and high-torque, to move the bike.

    If I ever remember to do it, I keep meaning to make a post with all the stuff I did find out there. I just never have the time when I have the round tuit.

    ReplyDelete
  3. Hi Again,

    Sorry, I meant to type "I'm surprised to note that you say that brushless motors CAN'T recapture braking power", which means that everything should make slightly more sense now.

    I will indeed be blogging my build, when I start, however matters are currently totally in the land of scientific theory at the moment as other ongoing hobbies and projects take priority.

    My plan was to construct a 3-wheeled contraption, probably with a custom-welded frame.

    The whole idea came from a set of old motorcycle wheels I saw laying about. They had really over-sized hubs to hold chunky old drum brakes (rather larger than most drum hubs I've seen) and I decided that it looked an absolutely ideal space to build hub motors.

    The design would call for what I think is is called an "outrunner" motor, where the rotor is built to the wall of the hub (where the brakes used to push against), and the stator is a fixed roughly in the place of the shoes (attached to the non-spinning side cover).

    This design appealed because it takes care of the casings for me, as well as suitable bearings (I'm sure motorcycle wheel bearings are up to the job). The wheels are also super-tough, and having a motor in each wheel would give slick handling and nice off-road potential.

    My ethos is definitely one of salvage and recycling where possible, however I'm not as committed to that aspect as you are. A motor of optimum efficiency would be very important to me, so I don't mind spending money on things like good magnets or the ideal gauge of magnetic wire. Home made is important too.

    Since first reading your blog, I did a bit of reading else-where and found that regen is possible with a brushless, achieved via controller complexity.

    I'm really hoping I can go the brushless route (more efficient, less wear on parts, more suited to outrunning motor, etc) so I guess I need to start figuring out these electronic controllers.

    Thanks for the speedy reply, by the way! You typed out quite a lot, so I'm still taking it all in. I might post back when I've developed some more thoughts on the matter.

    ReplyDelete
  4. > Sorry, I meant to type "I'm surprised to note that you say that brushless
    > motors CAN'T recapture braking power", which means that everything should
    > make slightly more sense now.

    :-) Everyone makes mistakes. Leaving stuff out is the most common one. Just don't leave out important parts of the stuff you build, cuz that's a lot harder to fix. :-P


    > I will indeed be blogging my build, when I start, however matters are
    > currently totally in the land of scientific theory at the moment as other
    > ongoing hobbies and projects take priority.

    Even theory can be good to publish, because as you write it out for others to read, you might ponder an idea you hadn't considered, or someone else might read it and suggest something you didn't think of. :-)



    > My plan was to construct a 3-wheeled
    > contraption, probably with a
    > custom-welded frame.

    Sounds like my eventual plan for a power-assisted recumbent. Got a few ideas for it as napkinsketches, but nothing up on the blog yet. I really ought to do a post for the idea, so I have a record of it in case I lose my sketches. :-(


    > The whole idea came from a set of old
    > motorcycle wheels I saw laying about.
    > They had really over-sized hubs to hold
    > chunky old drum brakes (rather
    > larger than most drum hubs I've seen)
    > and I decided that it looked an
    > absolutely ideal space to build hub motors.

    Now that is something I would like to find and build from. Only catch is the weight may require a lot more power than I could use legally on a bicycle. 1000Watts is, IIRC, the legal Arizona limit for power-assist on a bicycle, before it becomes classified as something higher that requires insurance, registration, and licensing, or at the very least is not allowed on bike trails and paths (which I depend on to get places, as I can't handle the traffic in certain areas; it's simply too dangerous due to some drivers that are inattentive and/or highly aggressive).


    > The design would call for what I think is is called an "outrunner" motor,
    > where the rotor is built to the wall of the hub (where the brakes used to
    > push against), and the stator is a fixed roughly in the place of the shoes
    > (attached to the non-spinning side cover).

    Yes, that's the type of motor usually used for a hub motor, and is what cieling fans are designed as (which is why I am trying to collect old ones for parts). If your hubs are thick steel, you can probably use them as the flux rings, and simply bolt your magnets directly to them, instead of adding more weight for flux rings that insert into the hubs.


    > This design appealed because it takes
    > care of the casings for me, as well as
    > suitable bearings (I'm sure motorcycle
    > wheel bearings are up to the job).
    > The wheels are also super-tough, and
    > having a motor in each wheel would give
    > slick handling and nice off-road potential.

    Yes, although you will significantly increase the weight with two motors, and thus twice the batteries to power them. A second controller is insigificant in weight, though not in cost if purchased instead of built from scratch (especially since many of the parts needed for a controller are available free or cheap in old electronics found scrapped out around the average city).


    > My ethos is definitely one of salvage
    > and recycling where possible, however
    > I'm not as committed to that aspect as
    > you are. A motor of optimum
    > efficiency would be very important to
    > me, so I don't mind spending money on
    > things like good magnets or the ideal
    > gauge of magnetic wire. Home made is
    > important too.

    Unfortunately using recycled parts *only* would be extremely limiting, which is why there are some things I simply won't be using them for. LEDs are very very hard to find in used stuff, at least for the superbright ones I need. Same goes for high-intensity magnets; the larger ones are very hard to find in recycled items, and I have yet to find more than one of a specific shape, except for harddisk magnets, which are all magnetized the wrong way. So purchasing some items is almost required. :-(


    > Since first reading your blog, I did a
    > bit of reading else-where and found
    > that regen is possible with a brushless,
    > achieved via controller complexity.

    Did the place you found the info discuss methods, or have even partial schematics? If so, a link there would be helpful to me. :-)


    > I'm really hoping I can go the brushless
    > route (more efficient, less wear on
    > parts, more suited to outrunning motor,
    > etc) so I guess I need to start
    > figuring out these electronic controllers.

    There is a project called OSMC Open Source Motor Controller,
    http://www.robotpower.com/osmc_info/
    which has a mailing list I'm subscribed to. I have not yet had time to read many of the posts or it's archives, but have been saving them for when I get to the real controller-design stages, in case I can use one of the variations they've come up with, or parts of it, at least. It's a microcontroller-based design, so I don't want to use it unless I have to, at least until I have learned enough programming to be able to understand and modify code where needed for my own purposes.

    ReplyDelete

Alternate suggestions or improvements to anything that's been posted is very welcome, and extreme detail is preferred to brevity.

Keep in mind that unless you leave an email address in your comment, I haven't any way to reply to you except to reply to your comment here. That means if you want a reply, you'll have to come back to *this* blog entry and it's comments to see my reply to you, unless you leave some method of contact within your comment.