Tuesday, 15 March 2011

DIY Voltage Cell #2

Cell-stack #2 was constructed on Feb 17th 2011, using a similar construction to Cell-stack #1, with just a few differences:-

- there was no paper separator layer glued 
  with Starch to the Zinc sheet;
- the cotton layer was sprinkled with undiluted honey
  (ie. no added water);
- the cells were not initially heated at higher than ambient;

the circuit also was mostly the same, the main differences were:

- a multi-toroid transformer was used in place of the inductor;
- the buffer capacitor was doubled in capacity to 4700uF;
- the leakage current feed to the 'relaxation oscillator' 
  sub-circuit was repositioned at the Collector of the 2N2222
  to allow any residual coil-collapse current in the primary
  to flow back into the cell-stack 

(the 'relaxation oscillator' sub-circuit is shown as a generic oscillator here - it's the same component configuration as for the Cell #1 schematic)

the initial o/c terminal voltage was approx 1.75V, ie. higher than cell-stack #1 o/c voltage, but it decreased after connecting the load circuit (instead of increasing steadily, like cell-stack #1)

the voltage/temperature readings gave a slight indication that this system had inverse temperature characteristics compared to the Cell #1 system

after a few days the in-circuit terminal voltage started to level-out around 1.6V and, after about a week, the LCD clock was moved from cell-stack #1 to cell-stack #2 as an additional load with the LED flasher

within a few days, the terminal voltage of cell-stack #2 decreased to approx 1.2V and the LCD was removed

after approx 400 hours of continuous operation from construction, the in-circuit terminal voltage was centred around 1.2V and the voltage/temperature characteristic of the system became very clear - the terminal voltage was a remarkably close inverse-correlation with the ambient temperature (the opposite of the Cell #1 system, which showed a positive correlation)

since passing the 400 hours on-load point, the system appears now to be in a self-sustaining stage of operation (which has persisted for an additional 400 hours, so far):

  - the trend line for the cell voltage is the inverse of the trend line for the
    temperature readings

  - the 'crossover' points of the two sets of graphical data have kept level
    (see the 21 degC ruled line in the DIY Cell #2 graph,
    immediately below this text)

(blue = Vnimh;  red = Tnimh)                   

compare the graph data for the DIY Cell type #2 with the graph data for a system with a commercial rechargeable NiMH** cell type
(the graph immediately above this text) powering a similar LED flasher circuit

the NiMH powered system shows a positive correlation with temperature (similar to DIY Cell #1) but the trend of the in-circuit terminal voltage is decreasing steadily whilst the trend for the temperature is rising

therefore the graph data shows that the NiMH powered system is steadily discharging

(** the 1000mAh NiMH cell was used with a low state of charge to bring it closer to the very low energy capacity of the DIY cell and make it easier to see how the NiMH terminal voltage varied with continuous operation of a similar load circuit)

No comments:

Post a Comment