switching regulator + single cell reading about gruvin9x

[GoogleCodeExporter]

Thanks for the ideas. Please allow me to switch to "devil's advocate" mode and 
take a critical look ...

As it happens, the present regulators are both low drop out and work fine with 
2S Li-Po packs. The unit runs right down to below 6V, although my Fr-Sky 
transmitter module also requires minimum 6V. The 3.3V regulator was especially 
chosen too. It's one that doesn't like much more than 5V input, and give low 
drop-out, low loss in return.

The LCD back light, of course, already runs off a switching, step-up supply, 
and uses precious little power, thanks to the phosphorus at work.

The LDR idea is an interesting one, except there isn't really any spare place 
to connector such a sensor -- certainly not if we want to read analogue values 
by the CPU. 

Also, with my current 2S Li-Ion cells (TrustFire 2Sx3.7V at 2400mAH), I get 
days of run time, even with the backlight on a lot of the time. So, whilst it 
would be a nice feature, I don't feel it's needed.

Certainly, when running from an 8S dry-cell pack, things are not so efficient. 
I do not recommend this at all. Far too expensive in the mid to long run.

More comments always welcome.

Original comment by [email protected] on 17 Jan 2011 at 9:29

[GoogleCodeExporter]

Oh and, single cell reading -- perhaps even with a built in Li-Ion charger ... 
again, there are no spare analogue ports. I suppose I could multiplex other ADC 
ports "only in charge mode". But that's a poor solution.

In theory, though I'm yet to test it, the TrustFire "protected" cells (they 
have a little round circuit board at the negative end of each cell) should be 
safely chargeable with a simple, crude charger -- the likes of what could 
already be plugged in the existing charge socket. I need to find more 
information on this. But the sales literature does say, "protects from over 
charging."

See this article concerning the cells I use: 
http://www.rcmodelreviews.com/tendollarbattery.shtml 

Original comment by [email protected] on 17 Jan 2011 at 9:33

[GoogleCodeExporter]

hi, they are the same cells I use :)
last times I charged them one by one with crocodile plugs, but I noticed they 
keep well balanced. I also removed the diode on the charge plug so I can also 
use the jack port on the side.

I'll keep you informed with new ideas.

Original comment by [email protected] on 17 Jan 2011 at 9:36

[GoogleCodeExporter]

On the other side of the coin, I am thinking of going further with the next 
custom board design, by moving up to a more powerful chip, with at least 
another 20 odd inputs. But that's a complete redesign from the ground up, as 
well as major changes to the code base. So it won't be happening soon. For one 
thing, I'm completely out of money to fund such a project, so that too would 
have to be remedied, somehow.

P.S: The guy pictured at the ten-dollar-battery site is Bruce, not me. :p

Bryan.

Original comment by [email protected] on 17 Jan 2011 at 9:37

[GoogleCodeExporter]

what about a cortex m3? 32 bits and plenty of power.
I really like this radio and I think it can become the perfect one.

Original comment by [email protected] on 17 Jan 2011 at 9:42

[GoogleCodeExporter]

The V4 PCB design, being produced by a new contributor (Cam), is using the 
100-pin ATmega2560 chip. I believe there will be free ADC ports available. So 
the individual cell voltage measurement could possibly be added. I'll point Cam 
to this issue's comment thread.

Original comment by [email protected] on 21 Jun 2011 at 10:43

[GoogleCodeExporter]

Voltage dividers to the spare ADC channels have been added to the V4 PCB.

Cam.

Original comment by [email protected] on 21 Jun 2011 at 7:44

[GoogleCodeExporter]

Plans are being discussed to design a built-in LiPo charger -- as a separate, 
optional add-on board, probably replacing the transmitter module (all but 
blank) PCB in the rear of the cabinet. Might make sense to stick the SD card 
interface on there also. We shall see. Such a board will require the single 
cell reading feature, so the V4 (or V5) PCB(s) will include whatever is needed 
for that interface, later.

Original comment by [email protected] on 28 Jun 2011 at 6:01

• Changed state: Accepted

[GoogleCodeExporter]

This one has been lost in the noise and no longer thought to be of priority. 
We're letting it go.

Original comment by [email protected] on 30 Dec 2011 at 2:30

• Changed state: WontFix