Archived Forum

This is what i built few weeks ago - emontx is basically same as jeenode. http://openenergymonitor.org/emon/node/2109 . You need external current sensing device.

I'm fairly certain that some users have at least thought about monitoring the d.c. side - though not low voltage d.c. as you are intending. You don't mention the currents you are envisaging, so I don't know whether Hall effect sensors or plain shunts would be in order, but if the latter then you'd need to keep their volt drops low to keep power losses to a minimum so some amplification would be required to get a usable voltage to feed the ADC.

The emonTx is designed for a.c. mains use so it's probably not the best starting point, you might be better starting with an Arduino board with a custom interface to your hardware. You also need to think about your control algorithm. Calypso_rae's Mk 2 router algorithm is intrinsically tied to mains frequency and the characteristics of the energy meter, it seems to me that - depending on the characteristics of your pump, you will be essentially free of these limitations and you could drive your pump at variable speed or on/off with whatever period and duty cycle you desire. However, the basic control algorithm should stand you in good stead.

Oh - did not fully read that in addition to dc monitoring also a way to control output is needed. Attached a link to a schematic and code (with google translator) to a mppt controlled dc hot water heater. Its not exactly same what you want but another view to the same challenge: http://translate.google.com/translate?sl=fi&tl=en&js=n&prev=_t&hl=en&ie=UTF-8&eotf=1&u=http%3A%2F%2Fwww.poikkis.net%2Findex.php%3Ftopic%3D5155.0

Basically this circuit has a constant load resitor and then with varying output it switches excess energy to additional hot water tank heater resistors based on how much energy is available. I.e you could have motor on constantly and then switch the heater elements on using this kind of setup.

Erm...

I think the fish would die if the water were to be heated.

Thanks, that looks great and will be very helpful.  a good place to start.  I didn't know much about the jeenode before, I will check it out.

The currents are low, I'm talking about a 50 Watt panel at 18 volts,  a fairly simple 12 DC charge controller with multiple charge modes.   Yes, probably use a shunt and OpAmp into the ADC.   The 12 VDC Pump is somewhat flexible, it draws less than 2 amps at most voltage levels, from 11 - 13 volts out of the load side of the charge controller.

The Battery is deep cycle 12VDC 40 Amp hours, so with good control routine, should keep it in the 50% - 100% range.  In full sun and the pump running, should just lightly or trickle charge the battery, so no need for a dump load, plus, just running the pump evening and/or early morning will almost ensure the battery is never fully charged, come peak solar time.

However, on cloudy or rainy days, don't want to run the fountain pump too much to drain the batteries too low.  And one of the main reasons to have the monitoring, is to know if maybe need a 60 or 80 watt PV panel, for example.  So part of this is to help tweak the setup under real site conditions, not just estimates, of average solar days, etc.  and of course, see how much power is made from sun and used on pond.  It's not for any significant saving on AC, more not paying to run AC power out to the pond / fountain waterscape.  

lol.... Yes, don't want to heat the water, but I may add an underwater LED light to burn any extra watts I might have in summer months.