Energy requirement for shutdown

formatting link

formatting link

Just how easy do you want it to be? :-)
I use this with a Pi sat underneath a shortwave receive antenna. No noise picked up, no interference. Allows me to run my Pi in the field off the same battery that powers the short wave radio transceiver.
Reply to
mm0fmf
Loading thread data ...
But why would you want to? By driving up the voltage you get out the realm of the supercaps, you're now into high voltage, high capacitance types. Assuming that same 5V 300mA as earlier we need 90J to power the Pi for even 1 minute. At 40V we need 0.11F to store that much energy no matter how you get it out. Looking at Farnell's listings the cheapest suitable cap will set you back £44.56, part no 471460 if you're interested. Compare that with part no 2475275, a lead acid cell which costs £7.00, is 1/3rd the size and will power the Pi for four hours on the same basis.
It's not a question of what is possible, it is what is practical and proportionate.
--
Andrew Smallshaw 
andrews@sdf.org
Reply to
Andrew Smallshaw
I take your point, even if that is a buck rather than boost regulator. I was thinking of board level solutions but yes those modules are getting pretty cheap these days. On the other hand, a LF50CV costs me 75p and needs another 10p of caps, takes half a square inch even on stripboard, and the complete project will as such fit in a lot smaller case than one with a module like that. Switching regulators are always advancing but to suggest linear types inevitably belong to the 70s is churlish at best.
--
Andrew Smallshaw 
andrews@sdf.org
Reply to
Andrew Smallshaw
Actually, for the lower current parts it is not a lot different from linear regulators. Every data sheet I have seen from major manufacturers list vendors and multiple part numbers for the important parts and give good guidance in selecting those parts. Not much worse than looking to see which of the suggested parts have the best supply.
The only real design work is in the switch to change between battery to line when the AC power drops. That needs some thought. The simplest seems to be running the battery at a higher voltage as someone suggested which requires two switchers and not as common and simple parts as the ones made for 10 to 5 volts. Not that they are complex, there just aren't as many to choose from, at least when I was digging into these a decade ago.
But few of use are doing this for work. Either way, that is a decision they can make.
Manufacturers don't vary a lot, but yes, it can be hard to get good info on charging them. So why would you expect an off the shelf board would be better than what you would design? An engineer has the same problems I would in designing one.
You mean if it has not been used? Why not just deal with the voltage on the cell? When it drops below a threshold charge it. Both NiCd and NiMH self discharge significantly unless they are one of a few of the long shelf NiMH types out there. I think a maintenance charge on NiCd, NiMH or Lion is not recommended, but don't quote me, I'm no expert on charging. I just know what I've read... well, if I can remember it.
If you can trickle charge a battery, that is the best way to keep it up. Are you saying you can connect it to a voltage regulator indefinitely? I'm not familiar with gel cells. Are they lead acid?
--

Rick
Reply to
rickman
Supercaps can be strung in series to get the voltage rating up.
--

Rick
Reply to
rickman
That was a big advantage of NiCd - hang a 1% rate charger on it and you were done. I used to do that with d/t timers on model gliders: the timer didn't have a switch fitted and drew 80 uA (4, 50 mAh cell battery) or 300 uA (5 cell 50 mAh battery). I used constant current 900 uA chargers to keep them happy: that was enough to recharge the batter after a flying session while allowing for the current being used by the charger.
Battery maintenance was easy: leave it on continuous charge in the workshop and, on a competition or trimming day, disconnect the batteries and go fly. Check battery capacity every year and replace the batteries every 3 years or so when the capacity was down 20%.
I never tried NiMH cells, but they were known to have a higher self- discharge rate than NiCd.
--
martin@   | Martin Gregorie 
gregorie. | Essex, UK 
 Click to see the full signature
Reply to
Martin Gregorie
NiMH has a higher internal resistance, so not as good for high current apps. No NiMH in power tools.
--

Rick
Reply to
rickman
The batteries in my DeWalt are NiMH.
--
Steve O'Hara-Smith                          |   Directable Mirror Arrays 
C:>WIN                                      | A better way to focus the sun 
 Click to see the full signature
Reply to
Ahem A Rivet's Shot
Are you sure? They are usually NiCd or Lion. Your DeWalt what?
--

Rick
Reply to
rickman
That is not "also", that is just the phenomenon that there may be a lot of to-be-written blocks in RAM which I already mentioned.
Reply to
Rob
You should not build these from the chip level, as a hobbyist! Our Chinese friends make small modules with SMPS chip and associated inductor and caps, and sell them for 70 cents a piece including shipping. You buy them at AliExpress, Ebay, etc.
These drop in where you used a 78xx in the past, but they don't have the cooling issues as they operate more efficiently. (which results in longer cap or battery use in this case)
Reply to
Rob
DeWalt DC925 XRP drill/driver. The DE9130 charger handles NiCd or NiMH. The original batteries it came with were 2.6AH 18V NiMH (DC9503XJ) - there were also 2.0AH 18V NiCd available, now there are 2.4AH NiCd. Currently it has third party 3.0AH NiMH batteries. It delivers more torque than most mains drills apart from big SDS drills.
--
Steve O'Hara-Smith                          |   Directable Mirror Arrays 
C:>WIN                                      | A better way to focus the sun 
 Click to see the full signature
Reply to
Ahem A Rivet's Shot
Well spotted. The only real requirement is that the system can be turned off without being corrupted.
I often design power supplies in my day job, so naturally I think of a power supply solution. But it doesn't have to be that way. For example, one crude (but good enough) solution would be to clone the FS onto a whole lot of microSD cards and swap a new one in when the current one gets corrupted.
I'm rather missing the embedded systems I usually encounter, which might have something like UBIFS on SLC nand Flash. Turn it off, turn it on, repeat - it just keeps working. (We have a QA test here that does almost exactly that.)
Regards, Allan
Reply to
Allan Herriman
Interesting: I wonder whether those timers would have worked with NiMH. When the time expired they pulsed a solenoid to release a lever - 500mA for 15mS. This caused a 0.5v, bathtub-shaped drop at a new battery. Vertical voltage drop, flat 15mS base and 45 degree slope recovery over 5mS or so. As the battery aged, the drop increased and the flat-bottomed part of the pulse became shorter. When the drop was 2v and the pulse had become triangular it was time to bin the battery.
--
martin@   | Martin Gregorie 
gregorie. | Essex, UK 
 Click to see the full signature
Reply to
Martin Gregorie
I use these as drop-in 7805 replacements:
formatting link

Only 1A just like the 7805, but I've powered a few Pi projects with them.
Sadly 5x the price of a 7805, however...
Gordon
Reply to
Gordon Henderson
but from my high-school electronics that reduces the capacitance
1/c = 1/c1 +1/c2 +.... 1/cn
--
If you think nobody cares if you're alive, try missing a couple of car 
payments. 
 Click to see the full signature
Reply to
alister
And _when idle_ it mostly won't do any harm to just cut power without a graceful shutdown. That's not the case you care about.
Reply to
Alan Braggins
Ok, I stand corrected.
--

Rick
Reply to
rickman
With a high current draw, no matter how brief, higher internal resistance makes it worse. Alkalines are even worse than NiMH. I had an RC boat that would put around on alkalines and I put NiCds in to save a few bucks. The thing took off like a rocket! It only ran a short while though because of the faster drain.
--

Rick
Reply to
rickman
The trouble with those modules is they aren't well specified and good luck getting a data sheet on the IC on them. Battery charging needs to nail down specs so I don't recommend these modules as a rule. If you can make it work, great.
--

Rick
Reply to
rickman

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.