[RE-wrenches] Diversion load for AC coupled system

[Hugh]

At 12:16 -0700 30/11/11, Larry Crutcher,
>Here's a scenario: Sun is up and voltage rises slowly to the 
>diversion controller set point. PWM starts and maintains the 
>voltage. Current continues to increase but the load is keeping the 
>voltage constant until the controller is overloaded.

I think the point that most people don't get is that you can't 
overload the controller if you choose the correct dump load.  You can 
get to the point where the dump load is fully turned on (100% duty 
cycle) and then it is 'overloaded' in the sense that it has not 
further ability to limit the rise in voltage.  However it will not be 
damaged.   Then the voltage will rise further and hopefully another 
controller will start dumping or something good will happen to 
protect the batteries from damage.

If you have 2 controllers with dump loads that can handle 150% of the 
incoming PV current between them then you can be sure of never 
'overloading' them.  But it will not matter if you 'overload' one 
before the other comes on because it will never draw more current 
than the dump load takes at the prevailing voltage.

Here is a calculation:

3.8 kW at float voltage 54 volts gives 3800/54 = 70.4 amps.
According to NEC you multiply this by 1.5 to get 106 amps.
Now you need to choose dump loads that will each dump at least 53 
amps at 54 volts. (1.02 ohms, maybe one ohm plus resistance of leads.)
These will be the same loads as will do 60 amps at 61 volts and they 
are the lowest resistance you can safely use with the TS60.  If you 
plan to EQ at voltages higher than 61 then you need to use higher 
resistance and you cannot meet the NEC target of 150% of incoming 
current at float voltages.

That's how I see it anyway.  It's possible that the NEC thing does 
not required you to look at float voltage.

all the best,
-- 
Hugh Piggott

Scoraig
http://www.scoraigwind.co.uk