# Help needed with triangle wave circuit / Op amp mosfet driver.

#### Inspironator

Joined May 16, 2011
7
FOR YOUR CONSIDERATION: If your cell has Iron components, they will rust...basic chemistry. The cells will disintegrate with time and the water will have the resulting residue. You could use 316 stainless steel for all metal components in the water, but then your water would contaminate with Hexavalent chromium resulting from the stainless steel. Hexavalent chromium is toxic and you need be concerned about the waste stream in the environment. It would be best to scrap that paper and take an approach used by chemists: use carbon rods or carbon plates for the anode and cathode. Research hydrogen generators in the field of chemistry. You may even find some YouTube videos. Good luck.

#### RIKRIK

Joined Oct 11, 2019
60
Everything in the cell is stainless steel apart from the crimp connectors which are sandwiched between a stainless spacer and a stainless washer, to try to mitigate corrosion as its the best i could find at the time, both the papers use dry cells, one only uses mhz, cavitation happens at between 20/100khz. The second paper does use 25khz but doesnt focus on that frequency. It uses different frequencies, different voltages, different duty cycles. , iv got to see if it doesnt or does work and document what i find.so i wont be ditching it until i measure with pwm and with out. One side of the graph Gas output per min , the other current.

#### ci139

Joined Jul 11, 2016
1,194
the electrolyte temperature and resistance if not pH or if not chemical/molecular composition

#### Danko

Joined Nov 22, 2017
870
cavitation happens at between 20/100khz
Unfortunately it is impossible...
Molecules in electrolyte do not move under electrical current,
only positive charges are instantly jump from H2O to OH- molecules,
converting H2O to OH- and OH- to H2O.
So, electrolyte is not stressed mechanically at all.
Therefore cavitation absolutely will not happens.
Sorry...
BTW, micro bubbles, you saw in experiments are
not cavitation bubbles, they are produced gases.

#### RIKRIK

Joined Oct 11, 2019
60
impossible , i would say highly improbable, im not claiming anything unscientific, well i dont think i am, i have a cell , running off a 12v battery , 15a , tap water no added electrolyte,obviously theres going to be some trace elements in my water. all i got to do is test it with and without the circuit. If i get a better result, then all that is being claimed is that the cell is running more effective under those conditions. If not then you told me so.

#### RIKRIK

Joined Oct 11, 2019
60
Later on today im going to my local shop to see if i can get any distilled water, to see if the cell design works well with low electrolite tap water (most probable) or there is some other effect that works with distilled water. This sould give us all a better idea on what is going on. I did add a small amount bakingsoda to the cell after trying it with tap water, but the amps shot up enouth to blow a 30 amp fuse, so i stopped as my wires going to my cell, are not rated for that current.

Last edited:

#### RIKRIK

Joined Oct 11, 2019
60
Having issues with my circuit, will buy one of those cheap handheld oscilloscopes, couldnt find distilled water, so used some deionised water, heres a quick test

Last edited:

#### cmartinez

Joined Jan 17, 2007
6,786
Having issues with my circuit, will buy on of those cheap handheld oscilloscopes, couldnt find distilled water, so used some deionised water, heres tla quick test
Take a look at this tiny scope, it's proven to be a rather useful tool, at least for me. And it sells for less than $30 bucks! Thread Starter #### RIKRIK Joined Oct 11, 2019 60 Will have a look, theres some on ebay for about 20quid. All i need it for is seeing if im getting pwm. Also :0 my mosfet wasnt turning off, didnt add pull down. Do you reckon something like this will work. #### Danko Joined Nov 22, 2017 870 Also :0 my mosfet wasnt turning off, didnt add pull down. Do you reckon something like this will work. In experiment post #34 I used this$3.36 1Hz~150kHz PWM driver:
ZK-PP2K PWM DC 3.3~30V Regulator 8A 150W Pulse Frequency Duty Ratio

EDIT:
It will work:
(Circuit corrected after kind comments @AnalogKid and @ci139, Posts
#51 and post #52)
R3, R4 increased from 100 to 200 Ohm, models of real BJT selected, R5 decreased from 1 to 0.5 Ohm.

Of course Q1 and Q2 turn on during the brief interval when voltage V1 is travelling from high to low or low to high.
This means Q1 and Q2 will put a "short" across the power rail.
The addition of the R2 and R1 will allow a high current (240mA, 240ns) to flow but Q1 and Q2 will not be damaged.

Last edited:

#### AnalogKid

Joined Aug 1, 2013
8,308
In circuit #50, whatever is driving the circuit (V1) must be able to supply over 40 mA at a headroom of under 0.4 V. That is, an output swing of from 0.4 V or less to 4.6 V or more to guarantee that the opposing transistor is completely off. Even with that, there is going to be significant cross-conduction (shoot-through) in the two transistors during each transition.

ak

#### ci139

Joined Jul 11, 2016
1,194
In circuit #50
are used a (near) ideal transistor models also (no other danger than such may hang simulation or dc analysis at startup)
... it seems the Danko won't split the match unless there is a real need . . .

#### Danko

Joined Nov 22, 2017
870
Bit crude but do you think this would work.

IPP045N10N3GXKSA1 - £2.21

BTW high and low levels in this simplest circuit will reversed.
Real duty will 100% - DUTY_ON_SCREEN.
=====================
One question:
Design high efficiency electrolyzer
or high power boiler?
-------------------------------------
Electrolyzer has maximum efficiency = 70% at current density 400 A / sq.meter .
More density - lower efficiency.
So, for your 50 mm dia washers optimal current should be 0.785 A,
does not matter, is it DC or pulse current.

Last edited:

#### RIKRIK

Joined Oct 11, 2019
60
Well its just to give me abit pf headroom, it will work fine on my small cell 5a, 20% safety margin 6a , if i want to do tests with electrolite the current will be way above 8a. , The rated current of the module.

#### ci139

Joined Jul 11, 2016
1,194
as the latter can do over 100a
if you scroll down the sales site it says 50A . . . but with the amount of heat sinking surface provided - you should double--double-check it . . . especially if it's driven by (what i suspect is a slow & low power) opto . . . resulting increased dissipation at (linear region at) switch-over points