From: nospam@needed.com   
      
   JF Mezei wrote:   
   > question:   
   >   
   > Say battery is at 11vdc and Solar panel is at 15vdc (difference of 4vdc).   
   >   
   >   
   > If the difference between the 2 segments is low, do I have to worry   
   > about the diode letting some of the 15vdc flow towards the 11vdc ?   
   > (which would bypass the solar charger and potentially damage battery)   
   >   
   > Or are the diodes good at blocking current flow in the blocked direction   
   > for anything betwene 0 volt difference and the max voltage difference   
   > rating of the diode ?   
      
   Draw your diagram, and work out the forward or reverse bias conditions.   
      
   First off, you may need to modify your diagram a bit. Two things are   
   bothering me right now.   
      
   1) The manual for the charge controller, shows it has six terminals.   
       In your "diode diagram", you're assuming one side of the thing is a   
       "common voltage rail" (which we'll call ground for lack of a better   
       word). Yet, the diagram in the manual is drawn a bit upside-down.   
       As a result, you'll have to be a bit careful about where you   
       stick the diodes. You're assuming a negative ground, and the diode   
       magic is on the positive side of things. The diode scheme might need   
       to be placed in the negative side of the circuit. That manual... sucks.   
       A typical Chinese manual.   
      
       Of the six terminals, it might turn out three of the terminals are   
       at a common potential. You would be seeking to verify they haven't   
       done anything goofy to endanger your simple diode idea. If all   
       six terminals were independently floating, the situation would be...   
       interesting. This needs to be resolved before deciding what to do.   
       You may have to acquire the unit, and use an ohmmeter to do some   
       checks on the six terminals. (Ideally, you'd want a schematic, but   
       what are the odds they'll give you one ?) It is possible to make   
       a fully floating output, in a similar way to how ATX computer power   
       supplies work. Or, they can do a boost converter, with common ground   
       (which would be more to your liking). Since the unit is so cheap,   
       I'm guessing boost converter. And I chose the word "boost", because   
       the output stays at a steady 12.0V output, as the battery goes from   
       11.1V to 8.25V. It could even be a buck-boost though, for all I know.   
       That would cost more to build. The buck-boost would only be needed, if   
       there was a danger the LiPo battery could go above 12V.   
      
   2) For your diode circuit, you make no direct connections to the battery.   
       The battery connects to the charge controller, and that's it. By doing   
       it that way, you tie into the charge controller's "low battery" cutoff.   
       That's so the GPS doesn't drain the LiPo battery below 8V or so.   
       If you make diode connections to this diagram, it'll be between   
       the left-most pair and the right-most pair, rather than involve the   
       battery terminals at all.   
      
        Solar_Panel --- Charge     ----- 12V regulated   
                    --- Controller ----- output (switching supply, from LiPo   
   battery)   
                           |   |   
                          Battery   
                          (Make no   
                           other   
                           connections)   
      
       If you made a connection directly to the battery, then the GPS (load)   
       would need to have a policy of stopping current draw at 8.25V minus   
       one diode drop. It's just better to let the charge controller   
       manage that, by not making any diode connection to the battery.   
      
   *******   
      
   If there is nothing goofy going on, you do your electrical analysis by   
   looking at whether the diodes are forward or reverse biased. Now, the   
   numbers I put on here are no longer correct (since the charge controller   
   puts out 12V on the right, and not 11.1V while it's running). So this   
   diagram at the moment, is just an example of how you work it out,   
   rather than being the exact circuit.   
      
   [solar panel]----+----[solar charger]---+-[lithium polymer battery]   
   15V              +                      +------[load]   
                     +----[>]---+----[<]----+ 11V   
                           ^    +   
                           |   [GPS]   
                           |      ^   
                   1.2V drop at    \__ Node voltage is 13.8V and probably a   
                   3 amps, less        bit higher. The other (right hand) diode   
                   drop at lower       is reverse biased by 2.8V, and is   
                   current             not conducting.   
      
   When the solar panel drops to 11V, the current will be equally split between   
   the two power sources.   
      
   When the solar panel drops to 9.8V, the right hand diode will be   
   fully conducting and able to deliver 3 amps if needed. The left hand   
   diode will be zero (close to reverse) biased.   
      
   So the characteristic is "mushy", and doesn't cleanly move from   
   one condition to the other.   
      
   If the solar panel is in the dark, the GPS continues to draw power from   
   the 12V output side of the charge controller. When the LiPo battery   
   hits 8.25V, the 12V output drops to zero volts (charge controller cuts   
   off the load side). That is to protect the LiPo battery. By allowing the   
   charge controller to manage things, you get protection against "flattening"   
   the LiPo battery. If you connected one of your diodes right to the LiPo   
   battery,   
   you could drain it down to around a volt or so (so low, the charge controller   
   may refuse to charge it when the sun comes out later).   
      
   The batteryuniversity articles address some of the reasons for things like   
   a charge controller not being allowed to charge a completely flat battery.   
      
   *******   
      
   Jesus, that manual is annoying... Should the word "Float" even be in that   
   manual ? I thought "Float" was terminology for lead-acid battery technology.   
   As far as I know, there's no "Float" for a system like this.   
      
   http://batteryspace.com/prod-specs/SC-LI11V.pdf   
      
       Paul   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)