Well, after a lot of thinking and looking I think I’ve come up with a little better design for my plates.  I’ve chosen to stay with the plate design for now because it is much easier to change things up.  Once the tube design is built its a little more difficult to add more tubs or change things around for different tests such as plate seperation and so on.

 Like the previous design we still have the 11 plates that make up the unit.  The approximate spacing between plates is from 1-2mm.  The new design has the two top and bottom holes drilled straight through at 1/2″.  Then I have added two more alternating sets of holes on the left and right side of the plates.  These holes alternate between a threaded 5/16″ rod and a 1/2″ hole allowing for one plate to conduct and the other not.  The holes alternate from side to side to allow for cathode - anode - cathode -anode….. - cathode layout.  This design has the two outside plates being cathodes and therefore not conducting on the outside surfaces.

I’ve placed hard plastic supports on the 4 outside holes in order to support the assembly better, and have rubber spacers between each plate.  The assembly is put together before the threaded rods are inserted into place to allow each plate to stay more evenly spaced.

The device currently averages out to around 2.07amps with 12.9V for around 27watts.  During operation my MOSFET is getting extremely hot and I can already see that I’m going to have to put a heat sink next to it.  I’m a little worried about the 1/4watt resistors I have on the board and am waiting from input from a few of you on this to see if I need bigger ones.  So far the board is still working and I haven’t fried anything but there are a few things getting warm because I can smell them.

The video below shows all of this in more detail and also shows the device working.  I’m still only using 1/2 of the circuit and plan to start building the rest of the circuit later today.  Hopefully by tonight we can see what difference the entire circuit makes on the overall production.  There are still a few more things to consider on the circuit but we should be getting closer to something that will work.

Last note before video:  I noticed that on this design I immediately get massive bubble formation on all plates before anything starts to happen.  This is the first time I’ve seen anything like this and it may be what some people were talking about requiring the plates to be conditioned before they perform correctly.  I can definately see a problem with this now.  However, I didn’t see any drop in current and we actually got an increase compared to tests the other day.  So I’m not sure what to attribute to it at the moment.

Overall changes:  Plate spacing more consistent at between 1 and 2mm.  Larger 5/16 threaded rod for electrodes between plates compared to earlier 1/4″.  Plate spacing held firm by other non conduncting stainless steel 1/4″ threaded rod.  We are pulling over 1/2amp more juice and I really don’t yet know why.  I wouldn’t think the 5/16″ rod over the 1/4″ would make this much difference by itself…

Here is the video.  Please comment and advise if you see anything I need to address…

Test-18: This test shows new design assembly using 2 assembly (non conducting) bolts to hold plates together, then tapping 5/16Rod for anode and cathode electrodes. Also shows assembly in action.