> >@=q` bjbjqPqP ..::%
8,&$HVV"xxxSSS$hf
SSSSS
xxS
x
xS
xJ@,0H
SSSSSSSdSSSHSSSS
Running series cell electrolyzers on 50/60 Hz AC power
Many years ago, when I was determining the plate sizes for the 120 cell electrolyzer,
I made the assumption that the RMS voltage value of the rectified (but un-filtered) supply would be approx. the same as the mains power input.
In other words, I assumed that the rectified 240V AC (which is RMS), when LOADED, would be about the same value (minus rectifier losses) as the input.
Thereof the assumption that for a 240V AC input, divided by 120, would result in approx. 2V (RMS) across each and every cell.
Also, keeping current density to no more than 40 mA (0.04A) per cm.
However, when this cell was constructed, CURRENT draw turned out to be WAY in excess of what I assumed!!
This could only be due to MUCH higher cell voltage than I anticipated!
This required a fresh look at the power supply, since my previous experience has shown an EXPONENTIAL increase of current for increasing cell voltage!!
I found that measuring voltages with multimeters (including True RMS meters) are virtually meaningless!
Looking at the wave forms with oscilloscopes revealed the problem!
I used the following set-up for the tests:
A 240V/50VA transformer with secondary winding of 15V - 3.2A rating, connected to a 25A bridge rectifier.
The output of the bridge rectifier was loaded with a 3ohm/60W power resistor.
AC input to the bridge: 15V
DC + AC output from bridge: 13.2V (RMS)
(Current: 5A.)
However, the oscilloscope revealed a full wave rectified pulse waveform with a peak-to peak amplitude of 19.2V!!
(See attached oscilloscope screen image: rectacwave2.bmp)
Using the ratio 15:13.2 and 15:19.2 the values for 240V were calculated to be 211.2 and 307.2, respectively.
Dividing 211.2 by 120 (the number of cells in the electrolyzer) is 1.76V.
With only 1.76V per cell we would have, expressing it in a good old Aussie term, BUGGER ALL of gas!
Obviously that is NOT what is happening.
It is clear that the cells DO NOT respond to just RMS voltage but something else!
Dividing 307.2 by 120 is 2.56!
When the cells get 2.56V across them, the CURRENT sky rockets!!
I actually drew a chart (on a graph paper in those days, about 15 years ago!), showing the relationship between voltage across a single cell versus current.
I have now scanned it and found it good enough to show what is happening.
(Attached: Cell V-I graph.jpg)
Note (refer to the graph) that when the voltage is 2V, the current is about 0.15A.
As the voltage increases to 2.15V, the current is 0.5A.
When the voltage reach 2.5V, the current is around 3.43A!
Increasing the voltage to 2.6V (only 0.1V increase!) results in a current draw of 4.7A!
And so on.
To sum it up:
A 25% increase in cell voltage (from 2 to 2.5V) increases the current from 0.15A to 3.43A!
Expressing this in a more practical way:
From 2V, a 0.5V increase in cell voltage gives 22.86 times more current!
This explains why the 120 cell unit INSTANTLY blew all 15A mains fuses!
Further, it is also clear that the cells react to the PEAK applied voltage, NOT the RMS.
It needs to be pointed out that in order to make QUALITY gas (HHO, Hydroxy, Browns Gas, etc.), PULSING is necessary.
George Wiseman has also pointed this out in his Browns Gas Book Two which he published many years ago.
Quote (from page 18):
Power supply considerations
If we apply straight DC current to the electrolyzer, we find the oxygen and hydrogen devolving to their di-atomic state. We get NO Browns Gas.
The electricity MUST be pulsed to an electrolyzer to produce Browns Gas; 120 cps is sufficient to produce Browns Gas, even 100 cps will work; so regular wall cycles will work.
End quote.
So, the bottom line is that if we want to have only 2V per cell, (for optimum efficiency within practical limits) the electrolyzer running straight on 240VAC needs 153 cells!
That is 33 cells MORE than the present 120 cell design!
(In his book, George Wiseman suggests 138 cells for a 240V electrolyzer.)
Naturally, reducing plate sizes (surface area) would lower the current but this would drastically reduce efficiency due to the high current density.
(The plates would also erode quicker.)
Instead, we can limit the voltage input to the 120 cell unit by using phase control (already designed) which has negligible losses.
This will ensure that each cell will get only about 2V.
It also limits/regulates the current AND temperature as well.
Les Banki
(Electronic Design Engineer)
Water Fuel & LBE Technologies
PAGE
PAGE 1
!'*68<GZ B K x
y
z
,9.Lcq"
A
Z
s
v
̾ԇwwssh=hOhO>*hOhwhfh=h;>* htC>*h=h^">*h;h^"hhZ>*hZhZ5>*hZh86h86h,w5>*h,wh;htCh;5CJ aJ hH5CJ aJ h;h;5CJ aJ h9.h5CJ aJ *78 4
5
z
{
+,-.qr
Z
w
#` #J_`az"1ijtu}~.3ADL[COdƾƺưƬƨʣhgh`h`>* h=>*h`h (I>* hg>*hzhLhtCh>l5>*htChwh>l>*h>lhQkhwhxhwhx>*h=h,h^"hrhhOhOhO5>*4`a~[Cef+&opdef$)k}
%&0EUZenop.NPUry
%ƿxtttlhwhkf5hkfhq h5>*hwh5>*hwh5hgh`5hgh55h`5>*hOhO5>*CJaJhOhO5>*hO5>*h;hOh=hwhk>*h`hwhPC>*hPChPCh (I5>*hf~hrh (I* &'mxy(`aAhi&depgdggdxgdm%x&'(AESXfgipr%&JbdǿӳhH0JmHnHuhg
hg0Jjhg0JUhO)jhO)Uh;hB h6hh6hhR#hxhghBh86hmh`hwhf~hf~5>*hf~hwhkf5h]hgd;&`#$gdh;hBhO)hg
hg0J,1h/ =!"#$%@@@NormalCJ_HaJmH sH tH DA@DDefault Paragraph FontRiRTable Normal4
l4a(k(No List4 @4;Footer
!.)@.;Page Number.7845z{+,-.qrZw#`a~[C e f +
&op
&
'
mxy(`aAhi&dep00000000000000000000000000000000000000000000000000000000000000000000000000000000000000@0I00@0I00@0I00@0I00@0@0@0@0@0@0I00 $$$'d%
` '!!8@0(
B
S ?!'*8<[cAA635y
3333!'*8<[cAAJ^C e AESXfgc%*('ZO^"=x;R#O)PC (Ig9.hQkkum,wf~tCLrx86q HwO5kf>lf;`zgg,mB@**wV**@@UnknownGz Times New Roman5Symbol3&z Arial"1hdzF
!4d2QHX ?;2&Running series cells on 50 Hz AC power Les Banki Les BankiOh+'0 $0
P\
ht|(Running series cells on 50 Hz AC powerLes BankiNormalLes Banki6Microsoft Office Word@I@h@ :
՜.+,0hp|
*** 'Running series cells on 50 Hz AC powerTitle
!"#$%&'()*+,./012346789:;<?Root Entry F00AData
1Table WordDocument..SummaryInformation(-DocumentSummaryInformation85CompObjq
FMicrosoft Office Word Document
MSWordDocWord.Document.89q