www.recumbents.com - Practicable e-streamliner project bullet points
www.recumbents.com
www.recumbents.com
Home | Profile | Register | Active Topics | Members | Search | FAQ
Username:
Password:
Save Password
 All Forums
 www.recumbents.com
 Technical / Bike Building
 Practicable e-streamliner project bullet points
 New Topic  Reply to Topic
 Printer Friendly
Previous Page
Author Previous Topic Topic Next Topic
Page: of 2

Tom Schneider
recumbent enthusiast

146 Posts

Posted - 10/10/2020 :  14:05:43  Show Profile  Reply with Quote
quote:
Originally posted by estoudz

Also remember in a low power application like this efficiency is of primary importance. Even the 1.4 voltage drop for the pair of rectifier diodes in each section of the three phase bridge can reduce efficiency
drastically. You would probably be better off using synchronous rectification using a three phase mosfet bridge. I believe feeding your generator into a bldc ebike controller in regen mode will achieve synchronous rectification and act as an upconverter to raise the voltage up to the system battery voltage. Go to endless sphere for more information.

eric stoudemire



Sounds interesting. I will need to look into this. But first I must finish the Carbon DLR I am building.

Tom
Go to Top of Page

estoudz
Starting Member

USA
40 Posts

Posted - 10/10/2020 :  14:35:23  Show Profile  Reply with Quote
Tony in order to utilize the full wave rectifier he would have to have access to the center tap of the motor windings, most of the time only the three phase windings are accessible.

eric stoudemire
Go to Top of Page

estoudz
Starting Member

USA
40 Posts

Posted - 10/10/2020 :  20:22:39  Show Profile  Reply with Quote
After giving this some further thought I decided the best solution would be using a hub motor as a generator feeding into a brushless ebike controller as the rectifier and upconverter. The hub motor is already designed for low rpm's and has good current carrying capability for low I squared R losses. The controller is already designed for synchronous rectification and up conversion and due to both of these components being mass produced they can be bought cheaply.

eric stoudemire
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/11/2020 :  16:26:31  Show Profile  Reply with Quote
quote:
Originally posted by estoudz

After giving this some further thought I decided the best solution would be using a hub motor as a generator feeding into a brushless ebike controller as the rectifier and upconverter. The hub motor is already designed for low rpm's and has good current carrying capability for low I squared R losses. The controller is already designed for synchronous rectification and up conversion and due to both of these components being mass produced they can be bought cheaply.

eric stoudemire



Makes sense. Not most weight-efficient solution, but can be the most watt-effcient one (and most silent to boot), if you pick a very good motor like one from Grin Tech.

I wonder if you can fix the hub shell (lace it to the frame for instance :)), attach custom cranks to axles and use it in full direct drive?

Btw, talking of 'lacing a hub to the frame' - I've seen projects of bikes with a hub laced inside the frame. While this is certainly for aestetic value, I think it might actually be fine structurally, if hard to pull off I suspect.
Go to Top of Page

estoudz
Starting Member

USA
40 Posts

Posted - 10/11/2020 :  21:39:59  Show Profile  Reply with Quote
Balor the axle would need to be fixed due to the wires coming out of it and most hubs like to spin about 300 rpm so you can use a rear wheel bldc hub put a fixed cog on it and use a three times larger sprocket on your pedals.

eric stoudemire
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/12/2020 :  02:58:45  Show Profile  Reply with Quote
quote:
Originally posted by estoudz

Balor the axle would need to be fixed due to the wires coming out of it and most hubs like to spin about 300 rpm so you can use a rear wheel bldc hub put a fixed cog on it and use a three times larger sprocket on your pedals.

eric stoudemire



Some motors have a plate where wires come out of... but yea, unless you rewind the motor I guess only 100 rpm will not very efficient.

I've just went ahead and tested one geared motor (rated for 250w at 24v) on a Vesc from a buddy.
It has resistance of 0.250 ohms!
But no-load current of 1 amp at max rpm...

Funny enough, it is much more efficient (almost 86%) at 50v, 500w. But that is like 500 rpm. Will more or less nicely work at 36v on a 20" wheel I presume, but useless for me...
Go to Top of Page

estoudz
Starting Member

USA
40 Posts

Posted - 10/12/2020 :  10:07:24  Show Profile  Reply with Quote
That is the bad thing about hub motors, they require a lot of current when starting off but when they get up to speed the current falls off due to the counter emf. That is why I used a mid drive on my recumbent so that I could gear it down and have the motor run at a more efficient
rpm. I guess that is why people run hub motors at higher voltages to overcome the resistance of the windings and the counter emf. Like Tony said a low kv motor
will have more wire length in it thus raising the resistance and I squared R losses. I guess you could get a higher powered motor with more windings in parallel but at a higher cost and mass.

eric stoudemire

Edited by - estoudz on 10/12/2020 14:41:14
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/12/2020 :  16:51:38  Show Profile  Reply with Quote
On a good controller like Vesc you can do current control and have the motor stay within optimal currents all the time... at the cost of sluggish acceleration, of course.
Anyway, this is rather simple:
volts - rpm
amps - torque

volts times amps - power
More KV - makes for more rpm, but less torque.

So, if your hypothtics direct drive motor at 100 kv spins at 100 rpm the system while current limited to 10 amp your system voltage is 1 and your watts are 10.

If you add 10x mechanical reduction to the system, you'll get 10 volts and 100 watt, and you have to deduct 'copper losses' (IRsquared) and 'iron losses' (no-load current, which is a floating value that slowly grow with motor speed and is multiplied by resulting bems voltage) - it represents magnetic hysteresis and eddy currents.

I have a handy excel spreadsheet that can calculate optimal motor currents and volts from it's KV, resistance (Vesc can calculate it for you) and no-load current (the latter is an approximation because, like I said, it depends on speed and grows with it).

On a direct drive motor you will not be able to run faster than your voltage * Kv * wheel curcumference in meters per minute, btw. your torque willl drop off first, and once BEMF will grow higher than pack voltage you'll get forced regen. If your battery is overcharge and BMS shuts it off - your controller may blow.
Go to Top of Page

estoudz
Starting Member

USA
40 Posts

Posted - 10/12/2020 :  18:18:14  Show Profile  Reply with Quote
Good points! what do you think of high power axial flux motors? Do they use those in electric vehicles?

eric stoudemire
Go to Top of Page

Toecutter
New Member

USA
64 Posts

Posted - 10/15/2020 :  11:28:46  Show Profile  Reply with Quote
quote:
Originally posted by estoudz

Good points! what do you think of high power axial flux motors? Do they use those in electric vehicles?

eric stoudemire



The inventor of this type of motor does:

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

It would be nice to find them in a smaller size, around 6-8 lbs, but the ones available are too heavy for almost any pedalable ebike application.
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/15/2020 :  11:42:06  Show Profile  Reply with Quote
Frankly, after I've discovered drone BLPM motors and their insane power to weight rations and efficiencies... who needs more exotic tech when you can have 90+ efficiency and 1kw per pound of motor easily (if rather noisily, but there are way around that)?
Well, *I* do for electric transmission (where every 1% counts), but current tech is frankly just *good enough*.
The problem is with batteries. Always batteries.
Go to Top of Page

Toecutter
New Member

USA
64 Posts

Posted - 10/15/2020 :  12:18:39  Show Profile  Reply with Quote
If the goal is to be able to get car-like range at car-like speeds, the batteries are also *good enough* and have been so for about 10 years if the vehicle's aerodynamics are kind of crap(average new car), and 30 years if the vehicle is very streamlined(solar race cars, electric faired recumbents, velomobiles, streamlined cars like the Solectria Sunrise, ect). Recall that Cedric Lynch once rode his electric recumbent from London, UK to Birmingham, UK on a single 12V lead acid battery, a distance of over 100 miles, and at the time, chemistries with 2x or more the energy density like NiMH existed.

Today's batteries are awesome. Once my power supply battery arrives for my spot welder, I'll be setting up a 1.5 kWh pack that weighs about 12 lbs! That's going into my DIY coroplast velomobile. I don't have a fancy Axial Flux motor though, just plain old PMDC for me. It is said to peak at about 91%. My range is going to be in the hundreds of miles, for certain.

Bicycles and motorcycles generally don't even have aero as a design consideration. Making them work with batteries in a competitive manner to fossil fueled vehicles is much more difficult because of that.

The advantage of higher efficency for electric motors isn't going to impact range much, it is the potential for peak power that is impacted more than anything due to reduction of heat output. That is how AMZ can get 50 horsepower peak out of a 7 lb hub motor. Something like that, if released onto the market, would be awesome for ebikes...
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/15/2020 :  14:21:20  Show Profile  Reply with Quote
quote:
Originally posted by Toecutter

If the goal is to be able to get car-like range at car-like speeds, the batteries are also *good enough* and have been so for about 10 years if the vehicle's aerodynamics are kind of crap(average new car), and 30 years if the vehicle is very streamlined(solar race cars, electric faired recumbents, velomobiles, streamlined cars like the Solectria Sunrise, ect). Recall that Cedric Lynch once rode his electric recumbent from London, UK to Birmingham, UK on a single 12V lead acid battery, a distance of over 100 miles, and at the time, chemistries with 2x or more the energy density like NiMH existed.

Today's batteries are awesome. Once my power supply battery arrives for my spot welder, I'll be setting up a 1.5 kWh pack that weighs about 12 lbs! That's going into my DIY coroplast velomobile. I don't have a fancy Axial Flux motor though, just plain old PMDC for me. It is said to peak at about 91%. My range is going to be in the hundreds of miles, for certain.

Bicycles and motorcycles generally don't even have aero as a design consideration. Making them work with batteries in a competitive manner to fossil fueled vehicles is much more difficult because of that.

The advantage of higher efficency for electric motors isn't going to impact range much, it is the potential for peak power that is impacted more than anything due to reduction of heat output. That is how AMZ can get 50 horsepower peak out of a 7 lb hub motor. Something like that, if released onto the market, would be awesome for ebikes...



50 HP out of 7Lb hub motor? That's seriously cool... as in - there most be liquid cooling involved at the very least :) 0.05mm laminations, I'd like that! And it seems like a reluctance motor, too! I'd be happy for something 1/10 of the size on my bike :)

And yea, motorcycles, despite their much less size and weight are actually nearly about as 'aero' as cars and have about the same rolling resistance (if we are to compare racing bikes to racing cars, apples to apples).
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/16/2020 :  08:34:06  Show Profile  Reply with Quote
quote:
Originally posted by Toecutter

Today's batteries are awesome. Once my power supply battery arrives for my spot welder, I'll be setting up a 1.5 kWh pack that weighs about 12 lbs!



Compared to energy dencity of gasoline (~8kwh per liter, AFAIK) that is not exactly 'awesome'. And not exactly cheap, either.
Go to Top of Page

jjackstone
recumbent enthusiast

USA
281 Posts

Posted - 10/16/2020 :  12:15:38  Show Profile  Reply with Quote
I did a small survey of some of the guys with electric powered velomobiles a couple years ago. Most of them said they were averaging 5 to 6 whrs/mile energy usage without pedaling on the flats. IIRC the average speeds were 20+ mph. That translates to about 200 miles/kwhr. At 33.7 kwhr per gallon of gasoline there would be an equivalent of around 6700 mpg. That is of course with the energy already stored in the battery. Depending on the charger used the "mileage" would obviously vary based on energy extracted from the grid to charge the batteries, but charging lithium batteries is generally in the 90% efficiency range.

I still have the electric hub motored mountain bike I built back in 2006. At an average 18 to 20 mph with no pedaling I used to get about 1500 mpge. I ran that bike off a couple DeWalt LiFePo4 36 volt packs that I series up to make a 72 volt pack. I quit riding it after about 4000 miles but I still pull it out occasionally just to test it. It still works. The kicker is that I still have those original batteries after 14 years and they still work in both the e-bike and in the drill they were originally purposed for. Now I wouldn't expect the packs to have lasted the whole 14 years if I only used them on the bike but this does show that lithium based batteries can have a long and useful life. These packs were originally rated at either 2300 or 2400 mAhrs and they are now down to about 2000 mAhrs. I hated spending the $100+ for each pack at the time, but to me they have been well worth the cost. Anyway, I ramble. Hope you enjoyed the story.

JJ
Go to Top of Page

estoudz
Starting Member

USA
40 Posts

Posted - 10/17/2020 :  16:13:28  Show Profile  Reply with Quote
There is an article from the IHPVA journal 42-v12n3-1996 that talks about combining electric assist with streamlining. He achieves crazy ranges even with lead acid batteries.

eric stoudemire

Edited by - estoudz on 10/17/2020 21:33:36
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/19/2020 :  10:13:02  Show Profile  Reply with Quote
In fact, the best way to utilise e-assist on a higly streamlined 'hybrid HPV' in the most economical fashion is use it only for uphills and than ride regen brakes downhill (while maintain your most efficient human power output both uphilll and downhill, too).

This way you can disregard the low end spectrum of your gearing (it will be covered by the motor), and make the transmission to work as efficient as possible for cruizing speeds achievable on flats on human power (around 40-45 kmh depending on streamlning efficiency and your fitness), hence you will incur low drivetrain losses while cruizing, and will be able to get back at least 50% of your Wh back by regen braking downhill.

Here are a comple of links, estimating my motor efficiency and streamlining level:

Uphill:
https://www.gribble.org/cycling/power_v_speed.html?units=metric&rp_wr=110&rp_wb=50&rp_a=0.509&rp_cd=0.2&rp_dtl=12&ep_crr=0.005&ep_rho=1.22601&ep_g=5&ep_headwind=0&p2v=1000&v2p=40

Downhill:

https://www.gribble.org/cycling/power_v_speed.html?units=metric&rp_wr=110&rp_wb=50&rp_a=0.509&rp_cd=0.2&rp_dtl=12&ep_crr=0.005&ep_rho=1.22601&ep_g=-5&ep_headwind=0&p2v=1000&v2p=40


AS you see, you can easily get back 50% of power used for climbing moderate grades at decent speed, and if you keep pedalling - it gets even better, provided your motor is efficient enough.
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/19/2020 :  10:41:37  Show Profile  Reply with Quote
Btw, given my weight and distances I want to ride, a ride from Tver to Sanct-Petersburg has 550 km and has 5 kilometers of ascent. I'll need 2kwh just to climb that much with 100% efficiency.
Go to Top of Page

alevand
human power expert

USA
3753 Posts

Posted - 10/19/2020 :  10:44:45  Show Profile  Visit alevand's Homepage  Reply with Quote

How big of a battery would be needed for a 100 kg bike on a 100 m hill with a e-braking terminal velocity of 50 kph? The slower one descends the lower the aero losses, but at some point momentum is lost.

C:
Tony Levand
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/19/2020 :  12:01:20  Show Profile  Reply with Quote
Well, play with the sliders!

Funny enough, the slower your speed (and, obviously, less weight), the less the requirement for motor power and, hence, battery capacity to take the charging watts, because gravity is a constant force (actually, it is not a 'force' per se but anyway) when it comes to 'watts' it gets greater as your speed gets greater as well - untill you get hit with an aero 'wall', but your overall store of kinetic energy remains unchanged and by going slower you'll be able to extract it better provided your motors will be able to provide enough torque and maintain efficiency given amps generated.

All in all, in your case for moderate hill of 5%, you'll need about 500w of braking power (and you'll want to make sure that your motor operates at peak capacity in this case) and 1kwh battery - that is about 50v, 20ah I presume, if we are not to exceecd 0.5c of recommended charge rate.


Lifepo chemistry can take 1C charge rather easiliy, but heavier (but more reliable!). Exceed that and you'll be wasting power heating the battery and degrading it in the process.

If you wave very steep hills around, double the numbers.
Go to Top of Page

alevand
human power expert

USA
3753 Posts

Posted - 10/19/2020 :  12:30:09  Show Profile  Visit alevand's Homepage  Reply with Quote
https://www.maxwell.com/products/ultracapacitors/

C:
Tony Levand
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/19/2020 :  13:28:23  Show Profile  Reply with Quote
As far as I know, they are very expencive, have extremely low capacity (depsite being 'ultra') and most importantly their voltage curve makes them unsuitable for typical EV electronics.

Lithium titanate is a better fit, but ~1kwh lifepo battery can be had in ~5kg nowadays - so you'll get decent useful capacity for long rides, will not add to vehicle mass much and will be able to use 'off the shelf' chargers and controllers.
Go to Top of Page

estoudz
Starting Member

USA
40 Posts

Posted - 10/19/2020 :  18:26:36  Show Profile  Reply with Quote
There is a guy on endless sphere that built a serial hybrid trike. His pedal generator was buffered by a 16v 990f ultracapacitor which in turn fed a 220v inverter that powered a vfd for his ac induction motor.

eric stoudemire
Go to Top of Page

Balor
human power supergeek

Russia
1029 Posts

Posted - 10/20/2020 :  05:59:39  Show Profile  Reply with Quote
quote:
Originally posted by estoudz

There is a guy on endless sphere that built a serial hybrid trike. His pedal generator was buffered by a 16v 990f ultracapacitor which in turn fed a 220v inverter that powered a vfd for his ac induction motor.

eric stoudemire



That's a lot of steps... does not sound very efficient to be fair!

Anyway, graphene supercapacitors sound extremely interesting in theory but as far as I understand the tech is not quite there yet.

Unless you want to conquer 1 second to 60 mph barrier, or operate a cargo vehicle in mountaineous terrain, I really don't think that you need to use any supercaps in your system - the weight and complexity increase compared to simply increasing the size of your battery is simply not worth it...
Go to Top of Page

estoudz
Starting Member

USA
40 Posts

Posted - 10/20/2020 :  22:16:38  Show Profile  Reply with Quote
Not only inefficient but heavy as hell to boot!

eric stoudemire
Go to Top of Page
Page: of 2 Previous Topic Topic Next Topic  
Previous Page
 New Topic  Reply to Topic
 Printer Friendly
Jump To:
www.recumbents.com © 2019 www.recumbents.com Go To Top Of Page
Snitz Forums 2000