Rogers postbag

Tell us about the cool things you've recently got
stevestrong
Posts: 1829
Joined: Mon Oct 19, 2015 12:06 am
Location: Munich, Germany

Re: Rogers postbag

Post by stevestrong » Thu Nov 09, 2017 12:47 pm

What are the dimensions of the panels?
The power will rather be 0.6W, see here an efficiency diagram
Image
as feedback of one of the aliexpress buyers for this product: https://www.aliexpress.com/item/Portabl ... 54698.html

User avatar
RogerClark
Posts: 7482
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: Rogers postbag

Post by RogerClark » Thu Nov 09, 2017 9:34 pm

Steve

Panel outer dimensions are.

110 x 60mm.
I should take a photo of the back of the panel, as I now see it has 110*60 written on the back, which is the size;-)

Thickness is 3mm

Acrive area of the solar cells is approx 105 x 53 mm

The central line of these panels is black, not silver like on that AliExpress link

I will check again, but when I tested one of the panels using a 33 ohm resistor load, the output was 5V

So the power would be 5 x 5 / 33 = 0.75W and for a 27 ohm load i think the voltage was around 4.9V , so the power is 4.9 x 4.9 / 27 = 0.89W

I did not test with any other resistors, as the next value I have is 22 ohms and I think it would be too much load, and I did not have time to make resistances of 26, 25 etc

But for me the panel seems to give more power then 0.6 W and peehaps at 4.8V may be very close to 1W under ideal conditions.

I now need to work out the best way ti charge a 2000mAH LiPo with this cell.
At the moment, I am thinking of putting a regulator in parallel with the panel, so that above 4V it just dissapates the power through the regulator.

Or perhasps just feed the LiPo via the LDO regulator.

Or perhaps just use a 4 or 5 dioedes in series to drop the max voltage of 6V to 4V

(I think Ray has posted about charging LiPo batteries, or perhaps it was Andy... I will try to find their posts on this subject

User avatar
ahull
Posts: 1657
Joined: Mon Apr 27, 2015 11:04 pm
Location: Sunny Scotland
Contact:

Re: Rogers postbag

Post by ahull » Thu Nov 09, 2017 10:42 pm

Perhaps this is what you are thinking of.

Image

Quick weatherproof camera box.(CHDK Wiki)
- Andy Hull -

User avatar
Pito
Posts: 1628
Joined: Sat Mar 26, 2016 3:26 pm
Location: Rapa Nui

Re: Rogers postbag

Post by Pito » Fri Nov 10, 2017 12:04 am

I got a similar charger with almost the same solar panel, inside 2 accus, and a MC34063 switcher. You get 5V off it via an usb socket.. Cost 1Euro.
The simplest solution for you is a switcher running from 0.7-6V input and w/ 5V output, followed by a 3.7V Lion charger module..
Not sure you will get 150mA out of it most of the sunny day..
Pukao Hats Cleaning Services Ltd.

User avatar
RogerClark
Posts: 7482
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: Rogers postbag

Post by RogerClark » Fri Nov 10, 2017 12:21 am

Andy

Interesting...

Looking on eBay I can see similar things costing $20

Or what appears to be just the panel

http://www.ebay.com.au/itm/Outdoor-Camp ... 2938603575

for a couple of dollars

Without battery

or perhaps

https://www.ebay.com.au/itm/Portable-Wa ... 2358081041

with battery

But, I'm ultra cautious now about buying anything from eBay that claims to have a lipo in it, as either the lipo isnt there at all, or is no where near the advertised capacity

i.e in the cast of the last link, I'm sure it won't have 8000mAH for under $10 AUD

I found some interesting posts one web about using one of those $1 USB LiPo charger modules, and how its possible to change the charging current by changing the resistor on pin 2 of the controller IC, or someone has postulated that it would be possible to use a couple of transistors to control the charging current to optimise the voltage at 5V


e.g.

Image

Another way could potentially to use a LDR, instead of the charging resistor, but I don't know if the LDR changes resistance in the same was as needed, e.g. higher resistance may mean more current which is the opposite of what I want.

As every I was hoping for a simple and cheap solution...

PS. I know that Linear e.g make IC's to do just this job, but they cost $$$

Perhaps its a job for a STM8 ? Not sure how much current they take, perhaps they can be run really slowly a low current ?

User avatar
Pito
Posts: 1628
Joined: Sat Mar 26, 2016 3:26 pm
Location: Rapa Nui

Re: Rogers postbag

Post by Pito » Fri Nov 10, 2017 12:34 am

If I were you I would first hook up your panel on a Bpill and logging the currents and voltages on an sdcard. Let it run for a week. Then I would start with the actual design :)
PS: you will discover soon you need a 20x bigger solar panel for charging a 2Ah cell in a sunny day.. :P
Pukao Hats Cleaning Services Ltd.

User avatar
RogerClark
Posts: 7482
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: Rogers postbag

Post by RogerClark » Fri Nov 10, 2017 1:11 am

Pito wrote:
Fri Nov 10, 2017 12:34 am
If I were you I would first hook up your panel on a Bpill and logging the currents and voltages on an sdcard. Let it run for a week. Then I would start with the actual design :)
PS: you will discover soon you need a 20x bigger solar panel for charging a 2Ah cell in a sunny day.. :P
Logging is a good idea

Re: Size of panel

Depends on the power needed by the load device, I have some sensors that use the nRF51822 BLE device and a BMP280 which have been running for well over a year on a 800mAH battery.

So I don't see any problem powering them via a 0.75W panel

User avatar
RogerClark
Posts: 7482
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: Rogers postbag

Post by RogerClark » Tue Nov 14, 2017 3:22 am

I did some experiments with using the solar panel to charge a 2000mAH Lipo cell, via a TP4056 charge controller and there was some interesting results.

Basically there is no point controlling the current so that the voltage on the panel remains at 5V, because the TP4056 is not a "buck converter"

Hence the input current is virtually the same as the output current, even if Vin = 5V and Vout = 3V.

In terms of wattage, at Vin= 5V the max current is around 150mA (0.15A) = 0.75W but the battery is receiving 3V @ 0.15A = 0.45W

So, 0.3 of a watt is being lost somewhere.

Looking at the spec on the TP4056 it says its a "Linear" Li-lon Battery Charger, so I presume the 0.3W is being lost to heat in the charger when the battery is at 3V

This is nearly half the power of the panel, when the battery is completely flat, but more efficient as the battery voltage increases e.g.

Mid charge at 3.6V the loss would be 0.225W

The obvious way to recover this lost power is to use a Buck converter on the input of the TP4056 as I measured that the TP4056 only seems to need its Vin about 0.1V higher than Vout for it to operate.

However the problem with using a buck converter is economics. They cost $1 minimum and would also need modification, so that the output voltage could be regulated so that the input voltage on the panel could be maintained at 5V.

The cheap buck converters on eBay mainly use the a LM2596 (clone), and its output voltage is controlled via a feedback pin (4).

I think in theory it would be possible to replace the variable resistor that sets the feedback as a proportion of the output voltage, with a transistor which is driven by the input voltage from the solar panel.

Because the voltage on the panel is proportional to the load, and the load is proportional to the voltage difference between the charge controller and the battery.



So I think it could cost $2 to build a system to extract perhaps 0.25W maximum.

This however is not economic, because another "1W" panel (actually 0.6 to 0.75W is only $1.50)
Hence if space is not a premium it makes sense just to get a larger panel.

User avatar
RogerClark
Posts: 7482
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: Rogers postbag

Post by RogerClark » Tue Nov 14, 2017 4:00 am

Update.

I thought I'd try to manually replicate having a buck converter between the solar panel and the charge controller and I fitted a external 1k pot to the buck converter so I could easily change the output voltage.

I attached a volt meter to the panel and carefully adjusted the voltage on the panel until it dropped to approximately 5v (around 5.05V), and looked at the output voltage which was 3.7V

The battery voltage at this time was 3.55v, but the result was not what I expected, and rather than in excess of 150mA from the panel, the battery was only receiving around 125mA of current.

This was not what I expected, so I looked at the datasheet for the LM2596 and it looks like the efficiency is not that good when the input voltage is only 5V

On page 7 of the datasheet, under 3A load the efficiency with input of 5V and output of 3V is less than 70%

http://www.ti.com/lit/ds/symlink/lm2596.pdf

My load was only 125mA, but I suspect that the efficiency may be even worse.

So the idea of using a cheap LM2596 buck converter seems to be a non-starter

User avatar
Pito
Posts: 1628
Joined: Sat Mar 26, 2016 3:26 pm
Location: Rapa Nui

Re: Rogers postbag

Post by Pito » Tue Nov 14, 2017 8:50 am

Roger, as I wrote above you need a switcher (step up converter) working from 0.7V to 6V input, giving you say 5V stable at the output. How much current? - it depends, but at least the voltage will be stable. There are such chips, long time back I used to use a 5pin sot MAX17xx something
https://datasheets.maximintegrated.com/ ... AX1724.pdf
(chip+2 capacitors+10uH inductor), but there are better one chips today, sure..
PS: several worked here from 0.5V input, giving 5V stable output (or 3.3V stable based on the part number)..
Pukao Hats Cleaning Services Ltd.

Post Reply