Solar panel for boost while coasting?

[matterrr]

Just wanted to begin by saying that Endless Sphere has been an awesome resource for me and I have avoided posting questions to this point because you guys have answered all of them many times over. I thought some people might find this interesting.

For some background:

I have built a working electric vehicle that incorporates a brushless motor, Infineon controller, and a 37V 10S LiPo pack. I am making a second vehicle with the same specifications but, as my goal is to make this new vehicle as efficient as possible, and I'm toying with the idea of a small solar panel. The size of the solar panel will be restricted to about 2 square feet, or about 15 - 20 W.

The two thoughts I have had for the panel (bear with me, I am new to solar):

- Use the panel power to trickle charge battery
- Use the panel to provide a very small boost to the motor while coasting

The first option would necessitate adding active cell-balancing on board the vehicle, which would add complexity and cost (I am using CellLogs with an alarm circuit and a Cycle Analyst). I would also obviously need to have some type of boost supply and diode arrangement to provide a more suitable charging source for the battery.

The second option is more interesting to me. I ride in a flat area and utilize the pulse-and-glide method for efficiency. I would use the main battery for high-torque situations (starting from a dead stop and accelerating) but then switch to the solar panel power for glide situations when very little torque is needed to increase the glide distance. To maintain a speed of about 15 mph, the motor currently draws about 2 A, so I figure a 20 W solar panel boosted to 36V could supply about 0.5 A.

My question is, is this feasible? What would I need between the solar panel and the motor controller? Would switching between the two power sources pose any problems (huge transient voltages)?

Thanks for the help.

 

[amberwolf]

Dont' forget the extra wind resistance and weight of the panel, and how that affects performance and power usage.

If the panel is "rated" at 20W, then remember that is probably at peak sunlight at noon on a clear day, so you're not likely to get that much (probably not nearly taht much) most of the rest of the day, unless you can keep the panel perpendicular to the sun.


Don't forget efficiency of conversions is also low. A fair factor to use is 70-80%, but it could be less. So let's say you get 20W out of the panel. Whatever you use to convert that to the charging voltage for the pack will only then output say 75% of that, so call it 15W (might be better, might be worse, probalby worse).

For a charger, assuming it's a 12-24V panel, you could just use an RC charger, whcih will already typically do the balancing for you as well (although some may not do it in the same mode as charging; you'd have to check before you buy). This could save you a fair bit of complication to the system of charging. However, if the panel gets shaded (signage, buildings, etc) the charger will stop (probably power off) and you must manually restart it. A panel-voltage battery (1 or 2 very very small SLA) between the panel and charger will fix that, but adds weight.

If you assume that the solar panel and power converter weigh a few pounds, you might be able to simply add that much more battery and have it actually be more efficient, because the conversion of power from the battery vs the weight/air resistance of them is probably a lot higher than the panel, plus you can probably get a lot more Wh out of the battery than the panel.


If the goal is to let it charge all the time, even when not in use or near a charging spot, then the panel may be helpful and worth it.

 

[Drunkskunk]

Welcome to the forum. You pose an interesting idea.
You will only get your rated power from the panel when it's facing directly at the sun, otherwise it will generate significantly less. you can make the panel tilt, but if it tilts forward or back you have to factor it as wind resistance, and since something close to 90% of that 74 watts the motor is consuming is going to overcome wind resistance of the bike, that poses a problem. So tilting may only work side to side and be effective. That really only helps if you ride north or south, otherwise, only expect an average of 10 watts or so from the panels.

Then you have to factor the efficiency losses to get the power to the motor. The most efficient way is to feed it through an inverter into the battery like a charger. You can ignore balancing the pack for this, but having a BMS to monitor the pack is a good idea.

You'll also have to consider how much energy will be lost even when the panel is flat on the bike. a 2 foot wide, 1 inch thick panel will still be adding 24 square inches of frontal area to the bike and will consume extra power to get it down the road, even if the panel isn't in use.

 

[CamLight]

Interesting idea!

A couple of thoughts...
- A 20W panel will only deliver that when aimed perfectly at the sun during a clear day. How will you be able to aim the panel? An off-axis panel can deliver as little as 10% or so of the rated power. Same for cloudy days or when the sun is blocked.
- Figure on about 70%, or so, efficiency for a boost power supply from a solar source. That wold give you about 1.38A in full sun (if aimed properly) and perhaps 200mA on other days and if not aimed or it's blocked.
- For best results, you would need a "MPPT" (max. power-point tracking) boost converter to prevent drawing too much current from the panel (which collapses the panel's voltage) when it's not aimed at the full sun, on cloudy days, or if the sun is blocked by your body, trees, buildings, etc.
- Is this amount of boost worth the time, cost, and trouble of carrying the panel and buying the converter?

 

[matterrr]

Thanks for the thoughts, everyone. These are the things I was hoping to learn. A few more details, if it helps:

- the vehicle will already have a body shell and I would most likely be adding individual cells that conform to the body and wiring them together, rather than using a large flat panel. So the added wind resistance will be negligible.

I'm beginning to understand that the most efficient use of this energy is to charge the batteries. So from what I'm hearing, I might not need to balance the batteries on board if I am planning on a small increase in voltage while driving?

So a strategy might be to go from the solar cells to an MPPT to regulate, through a power diode,to the battery terminals?

 

[CamLight]

So a strategy might be to go from the solar cells to an MPPT to regulate, through a power diode,to the battery terminals?

That would work just fine. The electrons won't care where you connect though. The solar cells' electrons will go to whatever is drawing current, either the load when it's on or the battery when the load is off (or perhaps the battery even if the load is on, if the load is small enough).

Check the MPPT boost converter (charger) features when shopping around, many already come with the diode equivalent (a MOSFET acting as a diode) already installed.

I'm beginning to understand that the most efficient use of this energy is to charge the batteries.

As long as the solar cells can provide more current than is drawn from the batteries due to the additional weight of the solar cells, cabling, converter, etc. They would be a terrific way to recharge whenever you're stopped though.

 

[Jeremy Harris]

I can add some practical experience from an electric boat with 200W of solar panels mounted on a canopy. I have a really efficient buck MPPT to try and get the most out of the panels (one with a very low quiescent current and synchronous rectification), and this does reasonably well, it gives me around another 15 to 20% from the panels over a straight diode connection to the battery pack. On a good day I can get maybe 140 to 150 W, but the average is probably more like 60 to 80W. The array is quite big, it almost completely fills a canopy that's about 8ft x 3ft. The panels I used are semi-flexible Sunflex ones, that are lighter than glass panels and will conform to a slight curve easily.

As others have said in effect, the panel rating is for 1000W/m² solar power hitting the panel, which is roughly what you get at midday on the equator for a horizontal panel, not the typical conditions you might get anywhere north or south of the equator unless the panels are aligned and track the sun. As soon as the panel is off axis to the sun the power drops off, due to power being spread over a greater area. Also, as the panels heat up the output drops, as their efficiency is somewhat temperature dependent. Clouds have a pretty big effect, too.

I'd work on getting maybe 30% of the panel rated power as a working assumption for panel sizing (with an MPPT, less if you don't use one) and see if fitting panels is worthwhile. In my case I take advantage of the usage duty cycle. The panels charge the boat battery pack all the time the sun is out, I can cruise for around 8 hours on a full charge (with no sun at all) and as long as I keep the cruise time to total daylight time ratio at less than 1:2 I am pretty much OK (the boat uses around 120 W or so to cruise).

 

[dogman dan]

Do it for the fun of it, rather than the gain. Bear in mind, mild pedaling will get you 50w, at the rubber. With no additional wind drag. Spectacular gains in efficiency are there for you to harvest, simply by riding slower, and the pedaling you already do.

The math is a bit discouraging. Your pack has about 300 wh in it, so in one hour, your panel is going to only be capable of adding 20wh, and actually less. But vigourous pedaling for an hour, adding 100w can give you 100wh more. That 100 wh will make a noticeable difference in your range. 15 wh won't.

My own experiments, with both geared and dd motors, found no spectatular gains from pulse and glide vs just slow and steady. For a really long ride, I found slow and steady far more comfortable.

Any wattmeter, preferably a cyclanalyst, will give you the moment by moment real world feedback on the watts you are using, and wh/mi on the whole ride. Using a tool, you can train yourself to ride the most efficient possible ride.

What I settled in on, at any speed, was to set a speed with the throttle, then select a gear that allows you to pedal up to 100w, aiming to increase your speed by about 2 mph over the throttle setting. You get the gear and your pedaling right, and see at least 100w drop from your wattmeter. The slower you go, the bigger that 100w is proportional to the total. 15-18 mph turned out to be my best compromise between efficiency and getting there in a reasonable time.

 

[arkmundi]

The future of the planet requires a full address to the fossil-fuel, heavy, carbon intensive vehicle sector. The marriage of modern-day solar and light electric vehicles needs to be fully consumated. So go for it! Even if you don't get but marginal power from it, the advertizing value would be great. We need to get people thinking how to go 100% solar for their home/travel needs. My own solution, which I'm still in the process of implementing, is three fold:

• roof-top grid-tied solar greater than my actual usage (I get paid for the excess)
• my ebike used for transporation, enhanced by bus & rail for long-distance
• batteries, batteries, batteries (the reason I'm on this forum)

The idea is always have excess solar charged capacity on hand.

I'm also working on a solar pedicab project with earn-a-bike, our local youth oriented bicycle recycling non-profit enterprise. Still design phase. But it'll get people in the shop, where we can educate them on a sustainable transportation future. Best of luck!

 

[solarshift]

10-20 watts wont do all that mutch but why not! even if it takes a week to get a full charge thats one charge less a week that comes from the mains power supply. coudent agree more wyth you arkmundi your solar pedicab sonds really intresting. to me it makes lots of sence to directly solar charge ebike battries,

 

[dogman dan]

For sure go ahead and do it, but not for the range for the fun, for the PR value, etc.

I often discourage carrying panels around because a large heavy one might cost more watts in drag than it makes. If you were discussing a trailer in use where the wind drag is there already, covering it's top with a panel makes plenty of sense.

Nothing wrong with solar paneling your roof to power ev's an all the rest. I keep wanting to, but low income and poor health keeps getting in the way of financing it. I homebuilt solar heat panels years ago, and slashed my heat bill by about 80%. Once the house payment is gone in 6 years, solar pv is high on the list of house improvements.

 

[Jeremy Harris]

I've always thought that the idea of an ebike shelter, either at work or home, with solar panels for the roof made a lot of sense. If you wanted an overnight charge, then add a set of cheap, second hand, lead acid storage batteries to the shelter and then use them to charge the bike overnight with power they've stored during the day from the panels. Even a 200W to 250W rated panel would probably be enough to charge an average ebike like this and keep it completely off-grid.

 

[solarshift]

if you can ceep pannels horazontall and find ones that arnt hevy glass based ones it is posable to carry sevrel hundred watts of solar pannel wythoult a massive increse in power use. Ive had up to 400 watts of pannels as a roof and only increased the power consumption by about 4wh per kilometer from 20wh per kilometer at about 45kph. but it was already on a pritty big trike. 200 watts of solar would be compleatly adequate for most use, I curently have about 200 watts of pannels on my (trike) roof and I have never needed to wall charge, on an avrage day I can pull in arond 15h at 48 volts. it would be intresting to compare the increased power use caused by carrying pannels (and charging eh ebike battery directly) wyth the iinificencys involved wyth charging a lead acid battery, runnign an inverter, then running a battery charger.

 

[Jeremy Harris]

if you can ceep pannels horazontall and find ones that arnt hevy glass based ones it is posable to carry sevrel hundred watts of solar pannel wythoult a massive increse in power use. Ive had up to 400 watts of pannels as a roof and only increased the power consumption by about 4wh per kilometer from 20wh per kilometer at about 45kph. but it was already on a pritty big trike. 200 watts of solar would be compleatly adequate for most use, I curently have about 200 watts of pannels on my (trike) roof and I have never needed to wall charge, on an avrage day I can pull in arond 15h at 48 volts. it would be intresting to compare the increased power use caused by carrying pannels (and charging eh ebike battery directly) wyth the iinificencys involved wyth charging a lead acid battery, runnign an inverter, then running a battery charger.

Carrying panels on a trike is certainly a heck of a lot easier than on a bike. My suggestion of a charger/shelter was really aimed at ebikes, rather than trikes, as carrying panels around would be pretty impractical on a bike. It's easy to charge from 12 V batteries, as pretty much all RC-type chargers run directly from 12V, so there would be no need for inverters or much additional loss, as the chargers could run directly from the batteries.

 

[dak664]

Maximum efficiency would have the solar shed directly charge a second pack through a buck converter, and swap packs as needed. The charge efficiency of lithium ion is much better than lead-acid, and no need to waste PV output during the daily absorption period.

 

[solarshift]

youre right tit would not be posablr/ practical to carry anywhere that amont of pannels on a bike, idident realise that yiu could get chargers that ran of 12v,thats worth knowing.

Maximum efficiency would have the solar shed directly charge a second pack through a buck converter, and swap packs as needed. The charge efficiency of lithium ion is much better than lead-acid, and no need to waste PV output during the daily absorption period.

that would be most effecnt, it may be a bit more flexabld on couldy days/cheep having a cheep lead acid storage? and mabie slightly bigger pannel to make up for the power lost threw. but just having a mppt and 2 ebike battries is mutch simpler and more efficent.

 

[John in CR]

25-35W per meter of panel is sobering. I hope most of that is due to your latitude up in the UK. The greater power of the sun here is readily felt even compared to the southern US.

Charge control has been my biggest hold back. Where did you get your buck MPPT? I need cheap and a reliable and accurate cutoff of 82-83V. Any suggestions?

John

I can add some practical experience from an electric boat with 200W of solar panels mounted on a canopy. I have a really efficient buck MPPT to try and get the most out of the panels (one with a very low quiescent current and synchronous rectification), and this does reasonably well, it gives me around another 15 to 20% from the panels over a straight diode connection to the battery pack. On a good day I can get maybe 140 to 150 W, but the average is probably more like 60 to 80W. The array is quite big, it almost completely fills a canopy that's about 8ft x 3ft. The panels I used are semi-flexible Sunflex ones, that are lighter than glass panels and will conform to a slight curve easily.

As others have said in effect, the panel rating is for 1000W/m² solar power hitting the panel, which is roughly what you get at midday on the equator for a horizontal panel, not the typical conditions you might get anywhere north or south of the equator unless the panels are aligned and track the sun. As soon as the panel is off axis to the sun the power drops off, due to power being spread over a greater area. Also, as the panels heat up the output drops, as their efficiency is somewhat temperature dependent. Clouds have a pretty big effect, too.

I'd work on getting maybe 30% of the panel rated power as a working assumption for panel sizing (with an MPPT, less if you don't use one) and see if fitting panels is worthwhile. In my case I take advantage of the usage duty cycle. The panels charge the boat battery pack all the time the sun is out, I can cruise for around 8 hours on a full charge (with no sun at all) and as long as I keep the cruise time to total daylight time ratio at less than 1:2 I am pretty much OK (the boat uses around 120 W or so to cruise).

 

[dogman dan]

Yeah,( refering back a page) but a second pack just adds to the cost, if it's an expensive lithium. But more panel to make up the losses is costly too. flip a coin which way to spend the dough.

One thing to bear in mind, at least for an ebike, is how little fossil fuel is used to charge a 1000 wh battery. Is it really worth fretting about 10-20 cents of fossil fuel a day? Not knocking a slolar ebike, but carrying more battery to extend range works well. You already have outstanding efficiency just for riding an ebike, compared to an e motorcycle or e car, let alone any kind of ICE bigger than a moped.

A very different situation of course, if the nearest plug is many miles away.

 

[Jeremy Harris]

25-35W per meter of panel is sobering. I hope most of that is due to your latitude up in the UK. The greater power of the sun here is readily felt even compared to the southern US.

Charge control has been my biggest hold back. Where did you get your buck MPPT? I need cheap and a reliable and accurate cutoff of 82-83V. Any suggestions?

John

Latitude plays a big part. We're about the same latitude here in southern England as the north end of Vancouver Island, so a fair way north of pretty much all you folk on the American continent.

My MPPT isn't available anymore, unfortunately. It's a Fatmax (http://www.solarfreaks.com/fatmax-high-power-maximum-power-point-tracking-unit-t142.html) that I bought from a chap in Australia and is super efficient but only OK up to 30 V input and about 200W. It doesn't do any sort of output regulation, so a BMS is required to limit charge current. Mine charges a 4S 8P Headway pack, with a homebrew BMS.

 

[dak664]

85 volts is high for most integrated circuits, but see http://www.digikey.com/product-detail/en/LT3958EUHE%23PBF/LT3958EUHE%23PBF-ND/2333098

Many more would be available if you could split the pack and charge in parallel at half the voltage. 60 volts is common for simple pre-engineered buck or boost converters.

Note MPPT is not needed for battery charging since the only parameter that needs to be maximized during bulk charge is the current. Buck or boost converters can do that with minimal brains. But a voltage limit may not provide an appropriate taper, so some smarts to periodically stop the charging to measure the battery voltage might give a faster charge. For the higher voltages, buck converters that use embedded microprocessors and discrete components are surprisingly robust and easy to make, but boost converters not so much (disconnect the load and the circuit can self-destuct within microseconds). So if you need custom 85 volt chargeing I'd look into a buck converter on a 100 or so volt panel string.

And I'd question that temporary storage in lead-acid is cheaper than the direct charging of a second lithium battery pack. The cost is comparable when you give lead-acid the 2x derating, and the charge efficiency for lithium is much higher.

 

[dogman dan]

Yes, either way would cost a bit more, which is cheaper would depend on the exact choices. For a little bike charging station, you might get away with scrounging some cheap lead.

Which is why I pointed out that it would all be done to conserve a few pennies of fossil fuel. In other words, you already picked the low hanging fruit by having an ebike.

 

[Jeremy Harris]

Scrounging cheap LA was what I had in mind. My workshop lighting has been running off grid for a couple of years now, on a few scrounged car batteries, that wouldn't start a car in cold weather, a cracked solar panel (kindly donated by a friend) and a bunch of cheap LED lights bought from ebay. Not a big deal in the scheme of things, but the cost was small and I've not had to turn on the three 5ft fluorescent tubes in there for a couple of years now, which has probably saved a fair bit of power.

The nice thing about using 12V LA batteries as a charging solution is that they are cheap and you can charge pretty much any ebike battery from them, with a bit of ingenuity. RC type chargers are pretty much all designed to run from 12V, and by having a few 12V batteries you can easily arrange to have isolated supplies, allowing a big battery pack to be charged with multiple RC chargers without needing to separate out the series connections.