SWIFT Solar Wide Functional Headlamp, developing and discussing.

Nice project! It is interesting to hear about how to make a light as efficient as possible, starting with the solar panel all the way to the amount of light that is produced.
I have litlle input for this because it is very much an electronical thing and that is not my subject.
Keep us informed! :slight_smile:

I do appreciate your interest! I agree, it impresses, when light - produces light. Searching for the li-ion solar charger was a challenge. I digged everything around, it’s very rare IC, even TI Instruments had nothing suitable, I wish, the charger could consume more power but 10W is maximum at the moment. All other components have strong characteristics, for example step-up converter can easily provide 3.5 A (90-97% efficiency depending on li-ion state of charge SOC) load up to 8A maximum, of course it will heat but even an average power bank isn’t always able to do the same.

Tomorrow will be the judgement day, the time to solder a new pcb has come, if something burn out, you might be the fist who hear my voice :slight_smile: Debug path is painful.

By the way, I’ve read your “Springs” test recently, find out it extremely useful. Thank you man. I decided to choose Fasttech ‘carobronze’ because of not high load currents of the single li-ion, current path of the external li-ion will come through the XT30 plug.

Efficiency is the high-ground.

Look forward to your results and further steps you take.

Interesting topic, I am also looking around for solutions to charge li-ions from solar.

Do you know about the CN3722 chip? Here is the datasheet 上海如韵电子有限公司-欢迎访问上海如韵电子有限公司主页!

Here is a ready made charge controller using this chip http://www.ebay.de/itm/181908611611

And here a review of this charge controller Cheap(est?) Lithium MPPT Solar Charge Controller CN3722 - 12v Solar Shed - YouTube

I plan to order this and use a 75W solar panel I already have lying around to charge a 3S li-ion battery pack, should work I hope.

edit: Where do you got the li-ion charger from?

If you can make your led engine fit some headlamp that already exist, it can be popular, you will be able to sell them as transformation kit.
Otherwise making plastic cases on 3D printer by 1pcs will cost more than zebralight for 95% of users.

Yes, it’s well known from me. But as for the efficiency, it doesn’t have a real mppt, so if weather conditions changing, device won’t track solar panel’s IV curve, basically this device is functioning with a static mpp instead of tracking. It will work and not really bad, anyway there’s an issue whether it worth to use advanced system, that in a few times more expensive. Check here, some good fully integrated solar chargers.
http://genasun.com/products-store/mppt-solar-charge-controllers/mppt-for-lithium-batteries/

I’m just worrying about rather high system power, you know… pcb will heat. I’m not sure that it’s possible to load this charger more than 25W without radiator, you should take it into account. As a rule power is growing together with lossless therefore a total efficiency should be in a priority here or not really if you have enough solar power. You know, if it weren’t for so expensive solar W/$ and mppt controllers cost, everything would be ok, that’s why we have to find balance.

I know one man, who’s actively exploiting CN3722 at the moment. As for me, I always read his blog with passion, especially the electronics chapter, but there’s no english language. Here’s is the link. Just try to translate, hope you’ll find it useful.
http://www.shpilev.net/archives/663

Well, I was lucky and bought a few pcs in my local store, when it costed around 3$, but Zmdi company is rebranding to Idt, that’s why this IC haven’t been selling so far. The total cost of the charger’s pcb is about 8$ including all recommending components. You may find test boards on mouser or farnell but they cost quite a lot.

I wish, I could… Look, there’s no suitable headlamp having enough place inside for all this electronics. I might find some examples but they’re not wide spread, their quality and price deserve better.

About the price, I don’t know, what’s a real self-cost of an average zebralight. There’re much more functions in my device, that’s why it’s hard to discuss price difference. It has already cost me 45$. I can’t mark any benefits of zebralight headlamp compared to my fully completed device, except one - more lumens but happiness lies not in lumens.

That’s really nice, when you’re not tied to a certain flashlight. 3D printing has began available as never before, even a kid is able to create a simple model. Prices are mainly depend on the filament and weight. I’m going to spent about 10$ for the first example, it’s not really expensive as seems. No matter, you order 1 pcs or 20 pcs, technology is the same. Therefore I suppose that’s a right direction. After making a 3D model, we could back to this point.

If I were you, I would prefer this charger (link below) to CN3722. Just because it’s TI, pcb probably has better thermal perfomance. I know very well this chip, that was my first acquaintance with this topic.

I would reccomend you to check YZXstudio DIY on Taobao. I have no idea what the voltage of your solar panel, it’s all up to you.
https://shop35123736.world.taobao.com/?spm=a312a.7728559.2015080705.3.JFsoJ7

Try this one, it’s must be the same and still available on sale. There are also chargers for Ni-Mh and Supercap.
http://eu.mouser.com/new/semtech/semtech-ts52001/

Thank you for your suggestion with the BQ24650. This looks like a good match to my 75W panel.

Edit: just ordered a board with this chip

Thread have been updated. “Update 1.”

A food for thought. What type of plugs would you choose? If you had a mini-power bank, that can provide high current 5V output. Especially, when there’s a possibility to connect external li-ion with higher capacity.

I’m using XT30 plugs at the moment, their disadvantage is low compatibility with different interfaces to accelerate the charge process. Usb type-c plugs might be widespread in the nearest future but they’re not serious in front of XT30 due to higher resistance. During the hiking, a tiny plug can be dropped in the mud and become unusable. So compatibility vs efficiency and reliability. On my mind there’s no sense to complicate current concept, you have 2 output wires, all the rest depends on your wishes. Am I on the right way?

About reverse polarity protection. Plugs will not allow to mix up the polarity. There’s only one path to burn the board while inserting the li-ion. Hope everybody has eyes. Yes, sometimes shit happens, it’s sad but also teaches us to be more responsible. I am for consequences :person_facepalming:

Let me know if you wanted to see more detailed DIY (Do It Yourself) guides dedicated to this project.

cool idea. :slight_smile: building Solar-recharged lights always catches my attention. :smiley:

All the best for your project Reyden. DBSAR is the resident camping light expert. Do I see a collaboration here in the future? :slight_smile:

Thank you! Well… who knows, I told you why I’m here and always opened for discussion.

For now, I have a few tasks: designing 3D model, soldering another pcb (hope sucsessful), layout changes are not expected. Once all the parts are adjusted, I will be able to improve each of them. I don’t ask for help but if someone wants to support me, that would be great. Failed debug slows down the process, I’m going to order new TPS61235 chip.

Solar charger, power bank, and led driver all in one, definitely an interesting and ambitious project. Best of luck with this.

No reverse battery protection? A lot of investment at risk for a sub $1 part? http://www.ti.com/lit/an/slva139/slva139.pdf

Thank you man! Frankly speaking, I didn’t find suitable place on the pcb without additional metallization holes. As for me I dont need it. But you may have been right. For example, any kid can spoil the device easily. I’ve bought a few SI7655ADN-T1, exactly costed 1$ each but decided not to use. Alright I’ll take it in attention.

By the way, I have one idea how to improve thermal resitance. Thermal vias on the factory pcb must be filled with solder. As a reult the surface become rough. To smooth roughness I would like to solder copper termopad 1mm on the back side or use silicone thermopad. I’ve not decided yet.

As I said, open source project is in priority. Manufacturers spend years developing upgraded versions. A great advantage when you’re able to make changes by yourself and not a small fix but a brand new case, that can be created from plastic filament on 3D printer. The same thing with a scheme, it’s quite easy to provide improvements, especially with a fresh ideas and great community.

So there are two ways: half and full DIY. I could order a factory pcb, that would be available on sale as a module with other parts and attached stl 3D model. For the rest people, it might be a guide, how to do it by yourself. Particularly, delivery and mass production issues are a part of most projects, as a result it goes slowly and require a lot of money. I would like to avoid this path.

What kind of realization do you expect from this project?

Reyden,

I want to congratulate you on your initiative and vision.

I use Solar panels to charge my Li-on cells on almost a daily basis. And I use a head lamp (NICHIA emitter modified) to read every evening.

With that said I have read your mission statement but to me I don’t see why. (but I am not a hard core climber either)(I will never sleep in a mummy bag spiked to the side of a mountain face)

Issue: Solar Panels are often sub optimized (don’t meet test max specifications in real life) I use a 50% fudge factor meaning I only expect 50% output from spec.
Solution: used overrated panels for task.

Issue: Power Banks Capacity, Maximum current, Weight
Solution: use Larger Power Banks and realize that the weight penalty is simply a current technological requirement, or bring extra cells to be charged.

Issue: Headlamps are often wanted to have high CRI floody beams and are light weight. Floody beams take lots of power to cover large areas.
Solution: High CRI is easy, Floody is easy and one needs to accept limits of how much area is/can be covered in flood with the power availible.

For head lamps close work is a good use and the obviates the issue of needing “large” flood lights. Use small light replaceable batteries with out carrying a large power bank around all the time.

My solution is to have a separate solar panel, a separate charger/power bank, and a separate head light. There is no doubt that my solution costs more and weighs more but it allows so much flexibility, for me it is simply a better solution.

For situations where more then one device needs power the ability to move batteries is so much more effective and efficient then trying to have every device have its own, (Charge controller/power bank) and possible solar panel.

Again, I am glad that you are out there solving the all in one solution.

Thank you for the response Ronin42. I’m glad to hear.
I will try to comment current issues through the prism of this project.

My Poweradd 14W is about 10W in a real life. But there are so many factors that could influence on the output parameters, in most cases users do not realize how to provide correct measurement. So, it’s a usual situation, when max specifications doesn’t not match. I do not exclude manufacturer’s fault. Yeah…, more powerful and quality solar panel, mppt controller have to solve this issue.

I appreciate high CRI light but it has much lower efficiency compared to normal CRI led bins. Sometimes it’s really hard to notice the color difference at the same temperature. I suppose that HICRI is not so important while reading, because paper color is mainly black and white. High CRI XHP Cree series has a bright future, they’re more powerful than Nichia leds and allows to get more lm/W. In addition EasyWhite technology helps to achieve more stable color without artefacts. Here’re the test reports of Nichia218B/XHP50.
https://drive.google.com/open?id=0B5UZoGmkPR9URTQzMUlnQm1vN1E
https://drive.google.com/open?id=0B5UZoGmkPR9UbnNkN0prMy1tdFk

As for more flexibility of a separate concept rather than everything in one. There’s no doubts, everything depends on the needs. Devices are usually fit on each other, if one of them does not meet the requirements the other one can be chosen. A separate device could have more functions as a result good flexibility. It’s all up to you, what exactly you want take. But I don’t want to pay for unused functions.

This project called “wide functional” because it combines a sufficient amount of quality functions for the outdoor/hiking/city usage. First of all I look at characteristics, some of the them are unique and rather hard to find analogs, therefore I appreciate customs. Of course it can’t be perfect for everyone. I’m on your side and understand why you decided to choose this way.

The question is next, what can I improve? There’s no other li-ion mppt charger on sale but 1.7A must be enough for a single li-ion. Headlamp power (at the moment 1.2W and 2.5W turbo) depends on the heatsink and power supply. I chose plastic case for a several reasons. 3D printing is cheap and available for everyone, easier to manufacture, can be made aluminium case from 3D model in future. No sense to develop another one headlamp like zebralight/skilhunt, their customs is also on sale. But I’m ready for the experiments.

Hey guys! I’ve gathered the most actual information in one video. Especially must be interesting the flashlight control and boost converter test, that I had never mentioned before.

SWIFT project begins here. Hope you enjoy!