High DC *output* MPPT

[xenodius]

Where shall I start?



Reasonable for a 100 pound curb weight vehicle


Sure you can buy them I did it. A 92 volt 90 AH battery weighs 220 pounds and is about what DIY EV cars use. In my case a golf cart which is over kill. Capacity and weight are directly related to each other. E-bikes use 36 volt 10 AH batteries that weigh about 6 to 7 pounds with a capacity of 360 watt hours. A 92 volt 90 AH battery weighs 220 pounds with a capacity 8280 watt hours. You cannot get around the energy density of 100 wh/Kg. Do the math.

A 8280 wh battery to be recharged in 1 single day with a 4 Sun Hour day takes 8280 wh / .9 efficiency / 4 Sun Hours = 2300 watts of solar panels. Very simple 5th grade math

An E-bike 36 volt 10 AH battery to recharge in 1 single day will require 360 wh / .9 eff / 4 Sun Hours = 100 watts

You can pretend all you want, the math and physics say you are nucking futs.

Do you want to talk about how much energy it requires to take a mass and accelerate it to X mph in x amount of seconds, and how much energy it takes it to keep going X mph? I bet Mr Newton would love to hear your method because James Watt is rolling with laughter in his grave right now.

Weight has almost nothing to do with it. It's all about aerodynamics. Sure, you lose energy as heat when you accelerate and then regen brake; about 25% of KE in my case, so about 10 watt-hours lost to ride up to 88mph and immediately brake back down. But aerodynamics are what really bite you. In my simulation with my controller, battery pack, motor, and tires, the difference between my fairing and a typical DF bike at 40mph is 2.6kw; 900 watts in the first case, and 3500 watts in the latter.

What chemistry are you figuring that off of? Here's a datasheet for the 18650 cells I will be using. Keep in mind, if charge rate and longevity wasn't a concern I could have switched to 3.4 ah Panasonics which would net me 11.3kwhr with about the same weight. With spacers and wiring they should still be under 40kg. Batteries are getting better all the time and NMC/NCA are no exception; you need to update your numbers. Packs around 200wh/kg are already available, and these use cheap, large, and heavy BMS's, unlike mine. There's a reason Tesla uses 18650 format cells.

Like I said, I don't expect that battery to be recharged in a day. I'd only like to get a few hundred watt hours of charge each day. If I need it charged fast, I'll plug in-- but that'll only be necessary on long trips. I never said anywhere I expected a full charge in one day. ELF rates their single semi-flex 100W panel at 568wh/7 hours, or 300wh/4 hours of sunlight. It's not unreasonable to expect the same performance from a system with four times the surface area.

Check your numbers, run them again, and if you still need me to explain to you why they're wrong, let me know. I'll be around.

[russ]

For the fairing on your bike - count maybe 10% of the panel rating - I still say you are smoking something. This is not real.

[xenodius]

For the fairing on your bike - count maybe 10% of the panel rating - I still say you are smoking something. This is not real.

Why so low? Like I said, I'd be happy with 18%. The ELF has been praised, despite the fact it has the aerodynamics of a wall and its low-end electric components have about twice as much waste heat as mine... A consequence of it's shape means that it is left with room for only a single 100W panel. That is enough for some people despite it's inefficiencies!

Tilt is important, but it's not 90% of the game... according to sources I've read, daily yield will be 30-40% higher by matching tilt. I see no other challenges in this case other than developing my own electronics.

@SunEagle; Thanks. Yeah I did wear a full-face helmet on that 70mph run, and it was on a 2% downgrade so it doesn't really count. I can get 63mph solid though, and stay over 60mph even up a 10% grade. Voltage limited speed is only 43mph with that bike, I have to use field weakening beyond that which requires a tremendous amount of power (up to 25kw total). It needs smooth roads, thankfully they're almost all brand new in my town. But the bike needs to be improved anyway =)

If I simply told you I was mounting fixed panels on a canvas or fiberglass, I seriously doubt I would be getting so much negative feedback. I suppose you think EV's are eclectic and shoddy, and that a homebuilt EV would be worse than a commercial one, yet it is the DIY EV's that are most impressive. Why would these panels be worse off in a fairing than on top of a boat?

[russ]

Why so low? Like I said, I'd be happy with 18%.

Tilt and orientation - you are going to get little to nothing. No big deal - the concept is foolish.

[CraziFuzzy]

The solar is unlikely to ever result in being worth the effort. If making it as a part of the fairing, you're looking at a situation where at all times, the different cells in the strings are going to be getting very different solar input, resulting in poor performance of the entire string. You'll likely get just as much benefit from simply eliminating the extra weight. If the shape of the vehicle was more solar friendly, it would be a different story.

Regarding charging a higher voltage pack, off-the-shelf isn't going to cut it. If you are already making your own FET based motor controller, it likely wouldn't be any more difficult to built a voltage boosting MPPT charge controller. But like I said, you are never going to get anything usable from those flex panels bend over a curved shape.

[thastinger]

Do you see any aero shaped cells on the PV cars the Universities race? No, they are pretty much flat and angled simi correctly. Every degree off from the Sun striking you panels perpendicular is lost power. If you are set on trying it, I would wire the panels vertically so that when you turn a corner one side or the other still makes some power.

[SunEagle]

Why so low? Like I said, I'd be happy with 18%. The ELF has been praised, despite the fact it has the aerodynamics of a wall and its low-end electric components have about twice as much waste heat as mine... A consequence of it's shape means that it is left with room for only a single 100W panel. That is enough for some people despite it's inefficiencies!

Tilt is important, but it's not 90% of the game... according to sources I've read, daily yield will be 30-40% higher by matching tilt. I see no other challenges in this case other than developing my own electronics.

@SunEagle; Thanks. Yeah I did wear a full-face helmet on that 70mph run, and it was on a 2% downgrade so it doesn't really count. I can get 63mph solid though, and stay over 60mph even up a 10% grade. Voltage limited speed is only 43mph with that bike, I have to use field weakening beyond that which requires a tremendous amount of power (up to 25kw total). It needs smooth roads, thankfully they're almost all brand new in my town. But the bike needs to be improved anyway =)

If I simply told you I was mounting fixed panels on a canvas or fiberglass, I seriously doubt I would be getting so much negative feedback. I suppose you think EV's are eclectic and shoddy, and that a homebuilt EV would be worse than a commercial one, yet it is the DIY EV's that are most impressive. Why would these panels be worse off in a fairing than on top of a boat?

Whatever you decide to use to mount those panels please take in consideration the additional drag and loss of aerodynamics. Having a sail in front and the wind in back works for sailboats but having a parachute in back for a bike is a drag.

With panels on top of a boat you have a higher percentage of them getting more sun due to lack of shadows being cast on them. When the panels are lower (like on a fairing) the chances of getting shadows (from the bike rider) is higher which hurts the amount of energy those panels can produce. Your 18% may be a high estimate if you bike into the sun. And there is no guarantee that even if the sun in behind you those panels won't get some shade.

[xenodius]

The solar is unlikely to ever result in being worth the effort. If making it as a part of the fairing, you're looking at a situation where at all times, the different cells in the strings are going to be getting very different solar input, resulting in poor performance of the entire string. You'll likely get just as much benefit from simply eliminating the extra weight. If the shape of the vehicle was more solar friendly, it would be a different story.

Regarding charging a higher voltage pack, off-the-shelf isn't going to cut it. If you are already making your own FET based motor controller, it likely wouldn't be any more difficult to built a voltage boosting MPPT charge controller. But like I said, you are never going to get anything usable from those flex panels bend over a curved shape.

I understand that different cell-to-cell insolation can really kill a module, and a shaded module in a simple string can bring down the other modules, but isn't that what diodes between modules are for? In fact, I actually planned to build separate MPPT's for each side. They are not very flexible so the cell-to-cell insolation in each module will not be very different in the absence of shadow; the ELF's panels are laid down along a 30-degree arc, yet multiple reviewers have stated even just 6 hours in ideal full sun is enough to net a full charge, though the official rating is 7.

I'm not familiar with these semi-flex panels outside of real-world applications I've seen like the ELF and their datasheets, and I know that's not always a perfect representation. But it tells me that this is certainly possible, and that's not what I'm hearing here, but I haven't seen any substantiation. So forgive me if I'm dismissive of your and others input, it's because I haven't seen any numbers or reports from anyone here and my own searches seem to indicate its quite possible.

People have suggested adding solar panels to EV's for a long time, but it's always been a pipe dream because electric cars use so much power that it simply doesn't matter. You might get a mile of range a day, parked in the sun... But I'd be using less than a tenth as much power and live in a small town, so it's not so inane. If I had simply told you guys this was for a trickle charger on a replacement hood for a VW Bug, would that have been as crazy?

Whatever you decide to use to mount those panels please take in consideration the additional drag and loss of aerodynamics. Having a sail in front and the wind in back works for sailboats but having a parachute in back for a bike is a drag.

With panels on top of a boat you have a higher percentage of them getting more sun due to lack of shadows being cast on them. When the panels are lower (like on a fairing) the chances of getting shadows (from the bike rider) is higher which hurts the amount of energy those panels can produce. Your 18% may be a high estimate if you bike into the sun. And there is no guarantee that even if the sun in behind you those panels won't get some shade.

I planned on laying them up right over my carbon fiber, and bagging them in Sylgard-184, so they should be an integral part of the fairing... I'm not sure I explained my intentions well enough, but they would be running lengthwise along the fairing at a relatively shallow angle, likely less than 30 degrees. I can imagine the case you described depending on the angle of my rear panels.

I couldn't really shade the panels while riding, though. A picture will make that clear. My fairing will be somewhere between VeloTilt and the Rocket project, but with a more upright seating position than the Rocket and a wider fairing than perhaps either; I want the cargo space. These bikes are in the .1 square meter CdA range, Velotilt is certainly less. And I'd be happy with .15 instead of .7! As you can imagine, with a lower front and wider sides, there would be a lot of very flat area for mounting panels and it could be made even flatter. And depending on how often I ride, yes, I could park it at an angle for temporary peak production from one side =) According to reports from VeloTilt though, I believe their tilting geometry is quite poor; their center-of-gravity lowers faster while leaning than a linear track vehicle would, that is, it is unstable. I will use swingarm and rocker offsets to achieve the opposite effect.

Thanks for not dismissing me outright.

@ Russ; I addressed tilt right after that line... did you even read that far? If it's so foolish, how did a bike like the ELF raise over $100,000 on Kickstarter and garner so much pre- and post-production praise despite it's ridiculous appearance and poor performance?

I have been mocked plenty, but I haven't heard a real reason why this is crazy or it won't work, only misunderstanding and misinformation. If there's a real flaw though, I'd like to find it sooner than later, so anything beyond one-liners are welcome.

[J.P.M.]

If I was sure I was right about all this thing, I'd build one, and if it worked, document the testing. Then, come back and rub the proof of concept in everyone's nose. Seems to be one possible way to settle the issue. Before I did that however, I'd be sure I had a better grounding in the basics of solar energy availability and utilization than the OP seems to exhibit.

[Mike90250]

...
I have been mocked plenty, but I haven't heard a real reason why this is crazy or it won't work, only misunderstanding and misinformation. If there's a real flaw though, I'd like to find it sooner than later, so anything beyond one-liners are welcome.

For an ultra light bike, sure, you might notice a bit of a difference. But it's going to cost you plenty in effort, to save 5 min of pedaling per day.

I don't have the background to address the battery weight/power issues.

But I think the solar aspect is flawed, because as others have said, until you have optimum angles on the panels, they are going to produce negligible power. PV works as a current source, they will quickly develop open circuit voltage in dim light, but no current. Diodes help in a large array, by bypassing shaded sections, but your conditions are going to be just the opposite, most of the array at any time, is going to be shaded or non-productive. if only 10% of your array is producing, at a generous 20V, there is no off the shelf MPPT that can manage that power, and then 80% power later in the day parked outside a large mirror, and deliver a useable harvest to your batteries.

Maybe it's a gut feel those of us with hands on experience have, or maybe you can give us a sketch of the final design, because the photo of stripped down frame, didn't help me with panel placement visualization.

[russ]

If I had simply told you guys this was for a trickle charger on a replacement hood for a VW Bug, would that have been as crazy? Crazy I don't kmow but useless

I can imagine the case you described depending on the angle of my rear panels. 90% of which will never be aimed in a decent direction for PV production


@ Russ; I addressed tilt right after that line... did you even read that far? If it's so foolish, how did a bike like the ELF raise over $100,000 on Kickstarter and garner so much pre- and post-production praise despite it's ridiculous appearance and poor performance?

Kickstarter has more silly shyt on it than real technology - fools put up money for anything - including potato salad.

The potato salad came close to what ELF collected if I remember right.

Anything like this can collect praise as long as you stick to the green sites - there they will polish a turd if told it is a green thing to do.

[Sunking]

Lets start off on another foot. First I am going to direct you to a forum that specializes in DIY EV's where thousands have already done what you are trying to do including myself. You are going to get the same negative results there because they already know the limits and have pushed the limits and know what the limits are are.

You claim to be a PHD. I don't know what this so called degree is, but I can tell it has nothing to do with any engineering or science disciplines. Me, I am just an idiot Master Degree in Electrical Engineering with 35 years experience and have a passion for electric vehicles. I also have a lot of professional experience designing and building off grid solar systems. I am not an expert in EV's, but I know some and what I do know is you are blowing smoke.

One boundary you will soon learn there is a direct relationship mass and speed. If you were really had a PHD in any field of science you would know that from Sir Issac Newton Laws of Physics, and his formulas does not take wind, rolling, and internal resistance into consideration, that takes even more power to over come. From the Laws of Physics I know it is not possible to do what you have claimed to have already done. First thing I can check is your efficiency claims any one can perform and understand because the math is so simple. It is a starting point for any EV build Efficiency In Wh/mile is: Gross Vehicle Weight / 9 = roughly wh/mile. Having said that rolling, internal, and wind resistance are not factored into thatr estimate. So if you have a 500 pound bike with you fat butt parked in it the absolute best you can do is 500 / 9 = 55 wh / mile. And that mile will be very slow. If you really had a PHD then you would also know to take a vehicle from say 10 mph to 20 mph takes a minimum of 4 times more power.

Another barrier which you have not discovered yet is what we call Barrier #10. It cost a ton of money to build a vehicle that can go 10 miles in one day on solar power alone. To do it takes a 4-wheeled vehicle that is very flat and wide. They look like a aerodynamic cigar box with a bubble canopy known as solar racers. It takes a team of engineers/scientist with real deep sponsorship pockets to build one. To run those solar racers has to be in areas with excellent solar insolation around the hours of noon. They cannot afford the weigh of batteries, and all the panels are oriented basically facing straight up because that is the only orientation that can work. You cannot do any of that with a 2-wheeled vehicle.

Anyway I am done with you. I suggest you go over to DIY EV Forum and try your luck there. But they will not believe you either because they have already been there and done that. Many of the members are from the V industry including the engineers and scientist. They helped me build my 3 EV's. I will bet you any amount of money mine can go faster and longer than anything you can build.

[xenodius]

=

For an ultra light bike, sure, you might notice a bit of a difference. But it's going to cost you plenty in effort, to save 5 min of pedaling per day.

I don't have the background to address the battery weight/power issues.

But I think the solar aspect is flawed, because as others have said, until you have optimum angles on the panels, they are going to produce negligible power. PV works as a current source, they will quickly develop open circuit voltage in dim light, but no current. Diodes help in a large array, by bypassing shaded sections, but your conditions are going to be just the opposite, most of the array at any time, is going to be shaded or non-productive. if only 10% of your array is producing, at a generous 20V, there is no off the shelf MPPT that can manage that power, and then 80% power later in the day parked outside a large mirror, and deliver a useable harvest to your batteries.

Maybe it's a gut feel those of us with hands on experience have, or maybe you can give us a sketch of the final design, because the photo of stripped down frame, didn't help me with panel placement visualization.

I'd actually have to pedal for 1.5 to 3 hours to produce 300wh of energy! Most riders fall into the 100-200W range. I can float near 200w for at least an hour, but it's definitely work. Pedaling for your energy really puts the consumption of climate control or a car into perspective and gives you a healthy respect for power. My battery pack will hold the equivalent of 7,000 Calories of energy, but the body is only 20-25% efficient...

I don't think I'm misinformed about PV characteristics.

There are some papers from '91 into the 2000's detailing the losses at various angles of incidence. They found that the worst result was a 30% yield of what was geometrically expected at a 60-degree angle of incidence, that is, ~15% of rated output at a 60 degree angle. In this example, angle of incidence is degrees from perpendicular; that is, θ = 0 degrees = 100% production. So this example is a pretty sharp angle. The amount of geometrically expected production is cosine(θ). This site has a good interactive animation of this effect at the bottom of the page that accounts for the reduced efficiency at these angles. Please note that the default latitude is zero degrees, and as I live at 45 degrees I'm actually in a great situation to mount them at 30-45 degrees on my fairing. Output is pretty low after 45 degrees but not too bad before that point, and I could not possibly see an incident angle greater than that in more than half my panels at any time.

I may work on the model some over these holidays, but I'm away from my desktop and don't like to model off a trackpad. I've been changing parts of my sketches too quickly over the last year to keep up in Solidworks but I'm beginning to settle on all those details.

Maybe in a year or three I'll be able to come back here and show off a finished product... I enjoy fabrication so I'm not in a hurry, and there are many minutia such as this charge controller that can easily consume months of time. But it's a long way off.

PS: Thank you for a helpful response.

[xenodius]

Lets start off on another foot. First I am going to direct you to a forum that specializes in DIY EV's where thousands have already done what you are trying to do including myself. You are going to get the same negative results there because they already know the limits and have pushed the limits and know what the limits are are.

You claim to be a PHD. I don't know what this so called degree is, but I can tell it has nothing to do with any engineering or science disciplines. Me, I am just an idiot Master Degree in Electrical Engineering with 35 years experience and have a passion for electric vehicles. I also have a lot of professional experience designing and building off grid solar systems. I am not an expert in EV's, but I know some and what I do know is you are blowing smoke.

One boundary you will soon learn there is a direct relationship mass and speed. If you were really had a PHD in any field of science you would know that from Sir Issac Newton Laws of Physics, and his formulas does not take wind, rolling, and internal resistance into consideration, that takes even more power to over come. From the Laws of Physics I know it is not possible to do what you have claimed to have already done. First thing I can check is your efficiency claims any one can perform and understand because the math is so simple. It is a starting point for any EV build Efficiency In Wh/mile is: Gross Vehicle Weight / 9 = roughly wh/mile. Having said that rolling, internal, and wind resistance are not factored into thatr estimate. So if you have a 500 pound bike with you fat butt parked in it the absolute best you can do is 500 / 9 = 55 wh / mile. And that mile will be very slow. If you really had a PHD then you would also know to take a vehicle from say 10 mph to 20 mph takes a minimum of 4 times more power.

Another barrier which you have not discovered yet is what we call Barrier #10. It cost a ton of money to build a vehicle that can go 10 miles in one day on solar power alone. To do it takes a 4-wheeled vehicle that is very flat and wide. They look like a aerodynamic cigar box with a bubble canopy known as solar racers. It takes a team of engineers/scientist with real deep sponsorship pockets to build one. To run those solar racers has to be in areas with excellent solar insolation around the hours of noon. They cannot afford the weigh of batteries, and all the panels are oriented basically facing straight up because that is the only orientation that can work. You cannot do any of that with a 2-wheeled vehicle.

Anyway I am done with you. I suggest you go over to DIY EV Forum and try your luck there. But they will not believe you either because they have already been there and done that. Many of the members are from the V industry including the engineers and scientist. They helped me build my 3 EV's. I will bet you any amount of money mine can go faster and longer than anything you can build.

Neuroscience; certainly not engineering, but I enjoy fabrication. Education has nothing to do with it, obviously I understand that kinetic energy increases exponentially with speed. That envelope calculation of EV efficiency you're using is bull****, to model the efficiency of a vehicle you need to know it's weight, rolling resistance, effective cross-dimensional area, and everything about it's drivetrain. I did that for my numbers. A stock velomobile with a 500W mid-drive can do 40mph on the flat, even with a total weight of 300 pounds. That's 1/3rd of your envelope math. Weight is quite unimportant in this scale of vehicle, especially with efficient regenerative braking in excess of 600 newton-meters peak; it is aerodynamics that win the day.

I've been to that forum but spend most of my time on endless-sphere.com. In fact, the motor I'm using was recommended by Luke, aka liveforphysics, who worked on the motor and battery design of the Zero motorcycles. You are still either not reading what I wrote, trolling, or you're actually too stupid to understand it. That's not what I want to do. Those university vehicles are bound by all sorts of ****ty restrictions. I just want to trickle charge my big battery a few percent per day average, not run a vehicle on nothing but solar. Running right off of solar is a truly awful idea.

You may have experience with electric cars, but those have nowhere near the potential that smaller EV's do, in my opinion. I'm not the only guy who took a Cromotor and a decent controller and pack and made a 60mph bike. Lots of guys have done it... just google it.

Kickstarter has more silly shyt on it than real technology - fools put up money for anything - including potato salad.

The potato salad came close to what ELF collected if I remember right.

Anything like this can collect praise as long as you stick to the green sites - there they will polish a turd if told it is a green thing to do.

My point about the Bug was that I don't have unreasonable power requirements or expectations, and the reason anyone is throwing flak is because it sounds too good to be true. But really, it's all been done before-- just not all together.

It collected over $100,000, they're pretty damn overpriced... but certainly got a lot of attention and remains well-reviewed.

[CraziFuzzy]

Without showroom the shape of your fairing and how the dollar panels will be arranged, there's no way to asses the effectiveness of the array. If I'm picturing a typical aerodynamic fairing, I'm thinking in perfect conditions. Mauve a quarter of the cells would be within 15° of perpendicular. This means your MPPT will have to buck so much that the power available will be near zero. Of you have drawing or mock ups, the dismissal and mockery would be far less. Also, making this with individual cells is likely to allow you more control on shape and performance than using the cheap flex panels.