Selfrunning HHO system with 400 Watts additional output

[Paul-R]

NOBODY has claimed that THIS thread is about the Anton cell!

There is a clue in the title.
.

[mscoffman]

Hey Les,

I've actually learned something! I think.

First, your designs are not just Analog they are Pulse Width Proportional hybrid
Analog/Digital. If one can translate a control system to an entirely linear system
then one can model it entirely as control sequences and pulse-proportional
modules. (PID proportional control is actually pulse-proportional where the circuit
attempts to learn one important proportionary hidden variable of the system by
operational trial and error. Not required here.)

If one can translate control entirely to linear systems then one can ignore the
non-linear control laws which most often result in the more complex differential
equations intermediates.  Efficiency calculations can then be looked at as linear
equations.

---

Somewhat along the same lines with the system in question.  What I hear you saying is;
"Get the subsystem function from whatever the source you can, overunity comes with it.
Then carefully construct a demand control structure so that as the next subsystem raises
vs lowers it's energy demand, the current subsytem raises or lower it's demand in response."
Which make the chain efficiency more or less constant by PWP means. Try to get the
HydrOdxy to stay at a constant pressure so the proportioning injector can accurately
control how much hydrogen is injected into the engine manifold based in energy demand.

Avoid those subsystems that attempt to run at constant fixed power level then behave
very inefficiently at demand limits.

---

Ok..Thanks. You've made something very valuable available to us here.

[Les Banki]

Hey Les,

I've actually learned something! I think.

First, your designs are not just Analog they are Pulse Width Proportional hybrid
Analog/Digital. If one can translate a control system to an entirely linear system
then one can model it entirely as control sequences and pulse-proportional
modules. (PID proportional control is actually pulse-proportional where the circuit
attempts to learn one important proportionary hidden variable of the system by
operational trial and error. Not required here.)

If one can translate control entirely to linear systems then one can ignore the
non-linear control laws which most often result in the more complex differential
equations intermediates.  Efficiency calculations can then be looked at as linear
equations.

---

Somewhat along the same lines with the system in question.  What I hear you saying is;
"Get the subsystem function from whatever the source you can, overunity comes with it.
Then carefully construct a demand control structure so that as the next subsystem raises
vs lowers it's energy demand, the current subsytem raises or lower it's demand in response."
Which make the chain efficiency more or less constant by PWP means. Try to get the
HydrOdxy to stay at a constant pressure so the proportioning injector can accurately
control how much hydrogen is injected into the engine manifold based in energy demand.

Avoid those subsystems that attempt to run at constant fixed power level then behave
very inefficiently at demand limits.

---

Ok..Thanks. You've made something very valuable available to us here.

Mark,

Thank you very much for your kind words and even more thanks for your SUPERB analysis!

While highly "technical", I sincerely hope that your analysis does not fly above too many heads here!

As for me, I simply ACCEPT and thank SOURCE for the ideas and then try my best to implement them into usable form.

I also want you to know that ever since I started researching and experimenting with "water fuel" 20 years ago,

I have "copped" nothing but abuse!

So your posts are certainly "a breath of fresh air" and are much appreciated!

Thanks again!

Cheers,

Les Banki

[weirdal]

I'm hoping someone can help me size an appropriate transformer for Les' super simple 2kw electrolyzer power supply circuit (SS2KWEPS), I have a 21 plate dry cell electrolyzer (19*19*11 cm) that can 'reportedly' handle up to 35v/70amp (2.45KW). If I wanted to get as close to 2kw as possible what transformer should I use?

I already have a couple AC Motor Run 100uF Capacitors, I haven't calculated or measured how much current these will limit yet so I will likely need a few more. I'm in north america so mains here is 120v, the bridge rectifier I'm using is KBPC5010W, any suggestions would be helpful.

Capacitors:
http://www.amazon.com/Conditioner-CBB65A-1-50uF-Motor-Capacitor/dp/B008DEZIM0

-Al

[MarkE]

I'm hoping someone can help me size an appropriate transformer for Les' super simple 2kw electrolyzer power supply circuit (SS2KWEPS), I have a 21 plate dry cell electrolyzer (19*19*11 cm) that can 'reportedly' handle up to 35v/70amp (2.45KW). If I wanted to get as close to 2kw as possible what transformer should I use?

I already have a couple AC Motor Run 100uF Capacitors, I haven't calculated or measured how much current these will limit yet so I will likely need a few more. I'm in north america so mains here is 120v, the bridge rectifier I'm using is KBPC5010W, any suggestions would be helpful.

Capacitors:
http://www.amazon.com/Conditioner-CBB65A-1-50uF-Motor-Capacitor/dp/B008DEZIM0

-Al

2.5V * (N-1) is enough voltage to eletrolyze with reasonable electrolyte.  So you want at least 50V @ 40A.  35Vrms will peak at 50V, and your capacitors will droop after that. All is not lost, but you are running kind of ragged.  The simplest solution is to reduce the number of plates from 21 down to 14 or 15.    Your capacitors aren't really what you want.  At line frequency, your rail will droop about 1V for every ampere at 8,333 uF.    It would be a good idea to limit the droop to no more than 30% peak or 40% of the rms value.  So:

C = 8,333uF / A / V * 40A / ( 35V * 40% ) ~ 24,000uF.  You will need a ton of 100uF motor start capacitors to get there.  The caps should be rated at least 80V.

Another option is to remain light with the capacitors.  In that case, the electorlysis will start and stop 120 times per second as the voltage rises above 1.23V + margin per plate pair.  If you frop to 14 or 15 plates then you will get electrolysis about 50% of the time at a current that rises and falls 120 times per second.  So, you will maybe get about 25% the output as you will if you use 24,000uF of caps.  You will be more limited by the ripple current.  If you put ten of these in parallel, your ripple will be under control for a 14-15 plate design:

http://www.digikey.com/product-detail/en/SLPX682M080E9P3/338-1653-ND/1882130