[jackbauer]

Ok major progress today. I've been dogged by the voltage readings going very high at anything over 10 amps. Charging 4 batts i had to set it to 200v to charge at any real power despite the voltage never exceeding 58v in reality. Its a pure noise issue related to the layout of the components etc and its a generally noisy environment for the sparkfun board in any event.

Last night I had a brainwave. The voltage measurement is fine with the charger off so i coded the voltage measuring subroutine to put the pwm at 0 , take a few readings and average and then return the pwm to its previous value! What a difference! Today I ran the batts down a bit and opened the charger up to 30amps! Voltage read perfectly and the heatsink topped out at 33C! Ran a full 3 stage charge with voltage based transisitions not a problem. At high currents the inductor makes a tick-tick-tick sound. Reminds me of faulty tv line output stages!

Another unintended consequence of this setup is that the charger is measuring the pack voltage at zero charger current. This makes the voltage reading much more vaild. I subscribe to Jack Rikard's views on the bottom balancing / bms issues and i have seen that the voltage of a battery on charge is entirely dependant on charge current. Even today on absorbsion charge i had set the charger to an equivalent of 14.5v per battery. 14.8 being the max. Voltages were 14.5 , 14.3 , 14.2 and 14.4. I had the opertunity to speak to odyssey yesterday and they confirmed that they no longer recommend charging to 14.7v for anything more than 2 series connected batteries. they suggest setting the charger to 14.4v to allow for the slight tolerance in capacity.