Portable solar charging questions

[Pseudoabdul]

Hi everyone,

For a long time I've been contemplating building an E-bike with a solar trailer that charges the battery as you ride for high speed touring. I've read most of the information here and around the web but I still have a few more questions. There is a big race in europe each year based on this concept, but the website is light on technical details. I'm pretty electrically savvy but I still have a few questions.

My biggest issue is deciding on what kind of panel to buy. I definitely need a flexible style one to save on weight, the drawback here is that most of them are rated at 12V (i assume thats closed circuit). To charge my 36V lipo in need about 42V. For my prototyping I want about 100W. One option would be to get four 20W panels in series and then drop it down to the right voltage, but that is expensive and not space efficient. I"m looking at getting a 100W panel at 12V and stepping up the volatge. I'll probably use a boost converter but a transformer system is also on the cards. One thing that concerns me is the relatively large power going through these circuits.

If anyone has a good type of panel for testing and prototyping or any thoughts on this it would be much appreciated.

 

[amberwolf]

How much power will your bike use while in motion?

Do you need the panel to keep it charged all the time, or will you plug in to charge when it runs out?

(if you need to keep charged without plugging in, you have to have a panel setup that can provide on average *at least* as much as you use up. meaning that at it's peak it has to provide a lot more than you are using up, to make up for the off-peak-sun times early and late in the day when it can't output as much because of sun-angle on the panels, and more air between the sun and you)



Do you have pure bright sunny days all day long with nothing to shade the panel for the entire day and ride?

(Less sun = less power)



What is the efficiency of your charger, and of your voltage converters?

(that efficiency multiplies in series with your panel wattage to determine how much *maximum* wattage you can get out of the complete system. so if your charger is 60% efficient, and your converter is 70% efficient, and your panels put out a total of 80W peak, and 20W average, then you get 0.60 x 0.70 x 80 = 33.6W peak, and 8.4W average.

so if you are using up 100W constantly, you'd need about 20x (more, really) the panel area to make up for the losses in conversions...if your efficiency is better than that, you'd need less panel, but all of these things are important considerations)

 

[dogman dan]

This thread should have some good info for you.

https://endless-sphere.com/forums/viewtopic.php?f=3&t=62700&hilit=solar+trike

I like the way he eventually set it up, with the panels putting out pretty close to the voltage he needed, then using a voltage converter to pop it up just a little bit higher for the controller to use.

My crude understanding was that the closer his panel was to the desired voltage, the less the converter wasted it into heat. So I'd say go for multiple 12v panels, like 3 or 4 of them.

No charger losses, so he gets the most from his panels. Almost, but not quite just running straight from the panels.

 

[Pseudoabdul]

Hi everyone,

This is a project I've been working on for some time. As a bit of background I got a job in an electric bike shop. I live in Australia where there is plenty of sun and I want to do long distance touring with my bike. So for this project I need to have the solar panels charging while the motor is running.

The electric motor runs at 36V which gets its juice from a controller, which in turn gets its juice from a lithium ion battery. This battery is charged at about 42V. My biggest concern is what circuits to use between the panels and the battery.

One issue I forsee is that charging the battery and discharging it at the same time might cause problems. Lithium Ion batteries require a specific charge cycle and if you deplete it while it is charging it may damage the battery. One way around this would be to get a second battery and charge/use them separately. Another option would be to bypass the battery when the motor is running.

The other issue is the best way to power from the panels to the battery. For this project I need to use light weight flexible panels which I can only find in 12V. The battery needs 42V but the battery charger runs on mains voltage. What I'm wondering is how I can get the voltage up high enough while retaining efficiency. Three options come to mind.

1. Use a step up circuit to boost the voltage to the 42V and feed it directly into the battery.
2. Use a different curcuit to raise the voltage to mains level and use the mains charger.
3. Run several panels in series to increase the voltage.

I'm very keen to see wha tother people would do to get this working.

 

[Pseudoabdul]

Thanks for replying. I felt the other thread wasn't as specific as I would like it to be. I did read all the replies and I have been digging for information.

 

[dogman dan]

I found this guys setup pretty impressive. Lean, mean, and practical.

https://endless-sphere.com/forums/viewtopic.php?f=3&t=62700&hilit=solar+trike

 

[Pseudoabdul]

THanks, I've read through this before but I think there is plenty of detail I missed.

 

[Ykick]

Hardly worth the trouble managing large array on a moving vehicle. Have you seen the “Martian” movie yet?

A practical/workable solution IMO is to carry enough panels for significant charge rates and deploy them when parked/camped or whatever the frock you do for long distance touring.

Mounting enough surface area to reel in 100W is gonna compromise chassis design and aerodynamics to the point of diminished return(s) and/or creating a death trap.

The tech is simply not there yet for anything other than a school science project IMO.

 

[Pseudoabdul]

Sorry, I am new here after all. Before I did post I'd been digging a lot through the forums but couldn't find any information specific to my problem. The other thread was old so I figured that any replies were long gone, so I decided to make a new one that better addressed the issue. I think you are being a bit harsh on me here though. I've read through all the information sent to me but can't find anything specific to my needs. What do you think my next course of action should be?

 

[dogman dan]

Welllll,,, I thought the thread I linked to, did show you a very elegant way to do it, and it did answer your specific questions.

But since you hate really reading a long thread and I don't mind typing some. Here it is.

SOLAR POWERED BIKE FOR DUMMIES. copyright 2015.

1. Connect panels in series to get reasonably close to the desired voltage. For example, if using a 12v panel, 4 in series for 48v.

2. Connect panel output to a voltage converter of sufficient wattage for your array. The panel voltage will vary constantly, with the angle to the sun you ride, with clouds going by, with time of day, etc. The converter will turn that variable voltage output into the voltage you want. For example, if your 48v battery charges to 56.4v, you could make your converter output 56v, thus insuring you could never overcharge you battery. The idea is not to sit there getting fully charged, but to insure a decent flow of usable voltage as you travel. If you want a full charge, then set the thing to 56.4v.

3 . Connect the voltage converter output to the charger port of your battery. Most bike battery bms can handle 250w of input, and no way you are riding along with more than that deployed into the sun easily. I'd aim for about 100w of power, to keep the solar panels small enough to handle. Two 50w 24v panels, or four, 12v 25w panels.

You are going to have to go dig for the link to the place that sells 24v flexible panels in that thread. Or,,,, wait for it,,,, google search "24v flexible solar panels".

There you go, specific answers to your question, but not without sarcasam.

 

[Pseudoabdul]

Thanks Dogman Dan.

I understand your frustration but I think its a misunderstanding. But you make some good points. While I think your proposed system will work I don't think it is the best possible system which is what i'm trying to obtain. I think its important to make the system as efficient as possible and I don't think your does that. One thing I am trying to incorporate is maximum power point tracking. I'm pretty sure this is part of the solar controller but I'm still trying to get information on how best to set it up.

In terms of multiple panels in series vs one larger panel with a voltage boost: At this early stage I'm considering all my options and trying to form an opinion of which is best, not just which will work. I don't want to have to invest a lot of capital without fully considering the alternatives only to have to switch over later. I'm sure you can appreciate measure twice, cut once. The advantage of the single panel is that it will be much cheaper, and I can tune the voltage to be exactly what I want, rather than a multiple of 12V (Of course, if a charge controller has an option of being able to manually set the voltage this wouldn't matter, but that isn't the case on the ones I have for tasting at the moment).

My number one concern however is trying to follow the charge curve of the lithium ion battery.

My understanding is that just imputing a constant voltage to the battery will damage it. I'm keen to get everyone's opinion on this and possible solutions. This is why I've been leaning towards using an inverter to mains voltage and using the mains charger. That way I will know its safe, but it probably won't be the best efficiency.

I have been looking at some suggestions I got from an electronics forum. They suggested something like this. http://genasun.com/all-products/sol...ium/gvb-8a-li-lithium-solar-boost-controller/

This one does seemingly everything I want. It charges to Lithium Ion, it follows the charge cycle, and it bumps up the voltage to work with my 36V battery. I'll do more digging today and check it out but I think this looks like a good solution.

I'll try to post another update this evening.

 

[John in CR]

Did you see Solarshift's thread? He circumnavigated Australia on his solar powered trike. see https://endless-sphere.com/forums/viewtopic.php?t=46587

 

[dogman dan]

Sorry, but you sounded like you wanted it handed to you on a platter, so I did. ES is not instructables, here we do all kinds of kooky experimenting on shit, and then write it up. Lots of it is talking about what didn't really work. It's a lot to read through.

That charge controller does look ideal. Expensive, but ideal. Bear in mind, if the goal is only to add some watt hours while riding, your battery will never get full, until you stop for a long time. So damaging the battery is not very likely. To make that charge controller work the most efficiently, get the panel voltage closer to the output you need. So a 24v array better than 12v. 36v even better.

The only thing to worry about then, is whether your bikes battery has a bms that can handle the output of the charge controller, or whatever you use. Most bike bms can handle 250w easily, some up to 400w. So if the battery spec is 5 amps charge, then use no more than 250w of panels. Pretty hard to carry much more than 200w, but it can be done if you must. Often the solution for more than 200w of panel is a trailer.

There is a solar bike race in Europe, where you can see some examples of how they carry a shitload of solar panel. 200w worth is about the max that is reasonable. Because it's damn close to impossible to track the sun on a road that is not a straight line for miles, most just mount the panel horizontal, or close to it, while moving. So in fact, you may only see 100w.

But take 100w of pedaling, and add 100w of electric, and you can run 15 mph. That means a very low draw on a battery that is already charged to go 20 mph. But see what I mean, the net effect is still the battery discharging. That's why a lot of sophisticated charge controller is not needed till you stop for a long time, and sit charging up full. That's why you could do it cowboy style, with a cheap voltage converter.

https://www.youtube.com/watch?v=5U1QV3VNjcs

http://www.electricracenews.com/2013/03/sun-trip-7300-kilometer-solar-powered.html

These guys that ride the solar bikes from France to Morrocco do both stationary charging and gather power while riding. Check out this guys setup for stationary, and how it converts to riding mode.
https://www.youtube.com/watch?v=JYzF8uLbTNw

 

[inverse]

I did some research and simulations on this subject, and here's what I found:

based on the current size, weight and efficiency of commercial solar panels it quickly becomes impractical to carry more than about 60W worth of solar power.
Even if one uses flexibly panels, which weigh about one third of conventional panels, the surface area is going to subject you and the bike to strong aerodynamic forces. Any kind of wind is going to make the matter even worse.

Plus, vibrations are going to quickly damage the panels if they are too large to maintain rigidity. Also, keeping them oriented for maximum efficiency is not feasible, so expect to get about 40W worth of continuous power under the right conditions.

At this kind of power levels, rather than charging the batteries the energy you make is best thrown directly into the controller to supplement the battery power, effectively making it last longer.

When you stop, the panels may be used to charge the battery. If so, first you should stabilize the voltage via a buck-boost converter and then I'd just use a microcontroller to drive a buck-like downconverter in constant current (and finally constant voltage) mode with current and voltage feedback in order to correctly charge the battery.

 

[bobc]

I've made a couple of MPPT chargers for SLAs. Yes you want MPPT, you then get everything the panel can give rather than some or none of it. For my money what WORKS is a perturb & observe algorithm (or "hill climber") operating on array voltage to make the MPPT work. The charger section is a controller which dumps the array power into the batteries in a way that maintains the controlled array voltage. SO when you ride into shade, the charger reduces the charging current so the array voltage doesn't move. The charger has to do more if you're working with lipos for safety reasons. You have to get your head around separating the array voltage control from the stepping/optimisation.
If you have several arrays pointing in different directions then each needs a separate MPPT. I don't know how you'd make such a system safe with lipos (TBH it shouldn't be too hard)
Here's a picture of my 50W 24V SLA charger

and schematic

You can see these things can be made small cheap & compact & make quite a nice little project. I've not seen anything quite like this available commercially.
How it works? the 3845 is a standard switchmode controller chip doing the charger control. The 358 is a standard dual op amp, 1/2 is taking over from the 3845's internal error amp (2 modes, MPPT mode controlling array volts and charge control mode controlling battery volts). The other half amplifies a current shunt signal so that array power is known The tiny5/tiny10 is a little microcontroller from AVR which does the hill climbing MPPT function.
If you were DIYing this, I'd combine an arduino for stepping with some-ones switchmode demo board; should be reasonably straightforward
Hope some of that is of interest and or useful.