Weather Station Hardware > Remote Weather Monitoring
Davis Battery backup
Flag:
Ok so how can the best use be made of the Davis battery backup especially with regard WLIP and in remote systems?
The Envoy 3*AA and Console 3*C are pretty much useless when running a WLIP providing only a few hours.
So how can we make use of the battery backup curcuit to improve backup times? Thoughts?
One thought I have had is to extend the battery backup +/- wires from the circuit board (very easy with the envoy) and run external backup but what voltage will this curcuit handle and how does this curcuit interconnect with the primary external power input?
Typical power spec is rated at 5 volts and we know the primary cuircuit can handle 6 volts (even up to 7 volts). Battery backup is typically 4.7-4.8 volts with new batteries so the scenario might be
Primary 5 - 7 volts, backup 4.7 volts (typical)
What if it was
Primary 6 volts, backup 6 volts?
Primary 5 volts, backup 6 volts?
Thinking more of trying to eliminate DC/DC convertors (like 6 to 4.5v or 12 to 4.5v) from the system as this only introduces something else in the system that could fail and run direct from battery. I have seen 4.5v SLA batteries but what about a charger?
johnd:
--- Quote from: Flag on March 26, 2011, 07:08:40 PM ---The Envoy 3*AA and Console 3*C are pretty much useless when running a WLIP providing only a few hours.
--- End quote ---
I'm not convinced that the WLIP will run reliably on battery power at all. Whenever I've had the power go down on a WLIP system, it's always required a power recycle to reset the logger even though the Envoy/console had batteries in. Maybe I just wasn't lucky enough to catch it in the first few hours of outage or maybe the consoles batteries were already partially depeted - must admit that I haven't run any rigorous tests of WLIP on battery power alone. But I do also wonder whether the WLIP actually does benefit from being powered from 5v and not the 4.0-4.5v of a partially depleted battery.
But maybe you have done some tests?
--- Quote ---So how can we make use of the battery backup curcuit to improve backup times? Thoughts?
--- End quote ---
I'd suggest that the only simple solution is to wire the battery contacts to an external battery box using primary cells. If you had a 6-cell box of D cells wired as two parallel banks of 3 cells then that would give you a respectable current reserve. (But this is obviously on the assumption that 4.5v is indeed sufficient to power the WLIP reliably.) BTW I'm not sure how you can have a 4.5v SLA battery: two LA cells gives you a voltage of around 4.0v (maybe 4.2v @ literally 100% charge and no/minimal load) and three LA cells would be 6v nominal. Are you sure it wasn't say a 4-cell nicad pack that might give you eg 4.8v?
Also the max recommended voltage to apply to a VP2 console is 6v (personal communication from Davis).
--- Quote ---Thinking more of trying to eliminate DC/DC convertors (like 6 to 4.5v or 12 to 4.5v) from the system as this only introduces something else in the system that could fail and run direct from battery.
--- End quote ---
Hmm, I'm not sure that as soon as you move from a simple external battery box (which could I guess be an offsite-rechargeable pack in principle) then any of the solutions is swapping one type of added complexity for another.
For example, it is possible to imagine the console being powered from an efficient 5v or 6v UPS that is fed 12v DC from a solar PSU. I think the circuitry for this sort of device is fairly straightforward and it could use any size of 6v SLA that you cared to specify. But I haven't come across anything like this that is readily commercially available (I dare say that they do exist but where do you get one from and at a reasonable price?)
Obviously, another option is to use a separate 6v solar panel/controller/SLA just to power the console. Although 6v components are not as easy to track down as 12/24v, you can with a little effort find 5W 6v panels and regulators, which are potentially powerful enough to be useful if they're just powering the console/WLIP. (At some point in the design process you have to bite the bullet and decide just how much current/power reserve you want to build in. So a 17Ah 6v battery, to take a size at random, would give you around 170 hours - roughly a week - WLIP run-time. )
But I'm not convinced that putting in eg a separate 6v PSU for the console/WLIP or using a 6v PSU (if you can find a suitable part) is any simpler than the solution I use which is to reserve the last 20% or so of the capacity of a 12v/100Ah battery purely for use by the console/WLIP. Yes there's an electronic voltage switch involved and also a 5v DCDC converter, but both parts seem pretty reliable and are relatively inexpensive to buy.
Flag:
--- Quote ---Are you sure it wasn't say a 4-cell nicad pack that might give you eg 4.8v?
--- End quote ---
You're right the 4.5 figure is A/h of a 2*2volt battery bank giving 4.8v
--- Quote ---Also the max recommended voltage to apply to a VP2 console is 6v (personal communication from Davis).
--- End quote ---
This then differs from what the Davis solar panels can put out and also the float charge of most 6 volt staged/smart chargers connected to a 6 volt battery, which is around 6.8 volts.
--- Quote ---But I'm not convinced that putting in eg a separate 6v PSU for the console/WLIP or using a 6v PSU (if you can find a suitable part) is any simpler than the solution I use which is to reserve the last 20% or so of the capacity of a 12v/100Ah battery purely for use by the console/WLIP. Yes there's an electronic voltage switch involved and also a 5v DCDC converter, but both parts seem pretty reliable and are relatively inexpensive to buy.
--- End quote ---
So I assume then you consider the battery backup circuit as redundant?
And maybe this is where I see an issue? Here we have a reasonably low power 5-6 volt system yet we have to use quite large Amp/h 12 volt systems and associated add-on curcuits to keep it running?
Flag:
Ok so this raises another query.
Variable input DC/DC convertor capable of outputing 5 volts with either an input of 6 or 12 volts.
Would the same A/h 6 volt or 12 volt battery provide 5 volt output for the same amount of time?
johnd:
--- Quote from: Flag on March 27, 2011, 08:15:09 AM ---This then differs from what the Davis solar panels can put out and also the float charge of most 6 volt staged/smart chargers connected to a 6 volt battery, which is around 6.8 volts.
--- End quote ---
Two comments: First I've never really taken a look at how sophisticated the regulator on eg a 7707 Davis solar PSU is. But if it was more than just a trivially simple controller (which possibly/probably isn't the case) then the load output might be a regulated 5-6v value.
Second, I'm sure that 6v isn't a hard limit in the sense that 6.1v will cause the unit to fail. It's more that the console PCBA components will be rated for a max continuous voltage of around 6v. There will obviously be some margin on this and so maybe 6.8v for perhaps 10-20% of the time might well be perfectly acceptable, though it might be on the margin of causing slightly premature ageing. But plugging in say a 7.5v or 9v mains supply for permanent use certainly wouldn't be advisable.
--- Quote ---I assume then you consider the battery backup circuit as redundant?
--- End quote ---
Possibly, but I'm really not sure. All I can say is that IME the battery backup doesn't seem to work too well with the WLIP logger installed, but I haven't done any rigorous tests.
--- Quote ---And maybe this is where I see an issue? Here we have a reasonably low power 5-6 volt system yet we have to use quite large Amp/h 12 volt systems and associated add-on curcuits to keep it running?
--- End quote ---
I'm not sure why this would be an issue. The power density of batteries is simply not very good at delivering the levels of current that we take for granted from a small mains adapter over an extended period. I have toyed with the idea of using a methanol fuel cell like this one:
http://www.efoy.com/en/fuel-cells-technology.html
But these are still very expensive to buy. Maybe in a couple of years time? (Though Toshiba - IIRC the relevant make - have been rumoured to be about to bring one out to power laptops for several years now that might well be substantially cheaper. It's obviously close to production but seemingly not quite close enough as yet for whatever reason. Or maybe early production has started but availability is only in Japan? Certainly AFAIK they're not readily available outside of Japan.)
Part of the problem seems to be that TCP/IP connections (ie as used by the WLIP) seem to need to use a substantial chunk of power. I don't understand enough about network design & engineering to know why this should be the case. We wouldn't be having the same discussion if all we wanted to do was to power a VP2 console with a serial logger say - this is why I say (as per my post in the previous thread) that the VC sort of architecture is probably the best if you want to consume the absolute minimum of power at a remote site.
Navigation
[0] Message Index
[#] Next page
Go to full version