Computer Control of a PWM design. work in progress.

[z.monkey]

Howdy Lefferdink,

Digikey, Mouser, Allied, and Jameco are a few that I can think of.  All have internet catalogs and delivery service.  Even in a one horse town you can find something.  The hardware store might have some electric stuff.  You might find a guy that has a bunch of electronic junk.  You may not find the exact parts that you are looking for but may be able to cross reference them and come up with something usable.  I have a lot of electronic junk, and I am always digging out a part here or there.  Old TVs have lots of high voltage caps and diodes.  Audio equipment is a good source of OP Amps and Power Amps.  Just about everything that plugs into a wall socket has a transformer in it.  Even an old Radio Shack should have the basic LM741, IRF511, and CMOS gates laying around.  I had a need to build a bridge rectifier last night, so I went digging through the junk and found 4 brand new 1N5404s, nice and free!

My wife hates it, but electronic junk is gold!

Blessed Be Brother...

[Lanmasterd]

Hey lads was Up late again last nite designing.. so far Usb section is finished.. and the Dedicated 5v supply for the Atmel Atmega88 28pin finished.. 

Pcb will be double sided and components on single side only ..
Keeping with An Automotive design im going to be making all components through hole.. on the model.
we can worry about making things smalled later.. the over all unit size so far will be 5 inch tall 6.5 inch wide mounted in a aluminium Box with side finned design to act as a sort of cooling fins.. similar to a specker system amp.. in boy racer cars..

the cell design will be posted upon soon but im trying to do the electrics first.. keep the cell design till ive built all my needs into the pcb circuit..

With out drawing majour ampage does any one have ideas for stimulating production ...
ie ideas other then add extra power or electrolits . Water Has to stay Pure.. ull see y eventually..

yeah Digi key are great and if u are international custom to them if u ring them u get better delivery deals. i got nabed the first time didnt realise it was ups delivery and got charger 200 dollers for tax and delivery lol live ya learn..

Mouser are good and straight forward..

monkey i checked out that PIC18F4520 looks ok and might be suited for wat we would do.. but i see it dont have a built in frequancy generator..
that will add more to the board and make code writing a little longer..

I write in hex and c++ as they are the usable commands in All Atmel chips..
Not a mbr of IPC

I have a degree in Computer Programming and my skill in pcb design are from the work i do .. i work in a small prototyping company in ireland.. based in cork .. i do alot of my work from home.. in waterford..

the mower being B&S is handy there very rugged machines..

Ill message back soon

[z.monkey]

Howdy Y'all,

As for the Electrolzer it's self I have seen several designs.  One has the spiral of wires on an insulating core in a glass jar of water, don't like that one.  Another as concentric sets of stainless steel tubes, like that one better.  A third has one set of stainless tubes and a second, smaller set of stainless tubes, the two set are oriented so that the smaller set sit inside the larger set, this is about the same as the concentric stainless tube set.  All of these examples are producing HHO, but not efficiently for HHO on demand usage in a high volume application.

What I am visualizing is something akin to the plates inside of a lead acid battery.  We have to maximize the surface area of the electrodes in the water.  We need a plate like system, but not solid plates.  There need to be perforations, or a grid like structure to the plate.  Then we alternate the high voltage plates with the ground plates in rectangular vessel.  I would like to think that this thing would look something like a car battery in its final form, including the electrodes on the top.  Except that it would have a port for the HHO to come out, and a water in port.  I think that high voltage is the way to go.  The pulse train is something that I have not worked out yet, but it seems that other pulsers are not maintaining a continuous frequency.  There is a packet of several pulses, and then a quiet period.  Perhaps you could call it a frequency burst.  The housing for the electrolyzer should be something that is automotive quality, and can take a lot of torsional stress so that it will be durable.  I imagine that this technology, like most automotive technologies, will be proven in racing.  So if we design it to be as rugged and durable as possible it would reduce the amount of redesign we would have to do later when someone wants to put it in a race car.

OK, Mo Later...

[lefferdink]

hi guys:  I have been studying different types of Generators and  would like to elaborate on the styles, if that is ok.  The Meyers type is stronger than the box type.  Some box types need a flat area of 20" x10" and liquid can slosh around.  One person is using this method and the plates are 6"x6" @101 pieces of 316L @.062 thick and it needs 1/8" separation of the plates; plus he has to use koh additives.  We are not going to use additives are we? To me it is a crutch.  Meyers didn't use additives. His plates ( loose terms )are 1/6" spaced.  The closer the plates the conductivity gets better.

[Feynman]

Hi!  I don't know anything about generators or HHO, so I can't help with that yet. 

But I am working on a precision computer-controlled frequency generator, originally for TPU research and pulse charging.  Now it can't do PWM (yet), but perhaps we can use it as the clock source for other circuits.   

I am still working on the latest revision, and am waiting for parts.  My hopes is that maybe by posting I can save others some time, so we don't always have to re-invent the wheel.  My idea was to make the system to "scan" across a large frequency range, while the operator looked for interesting effects or there was automated data collection. To get the PWM you'd need to send the clock into high-speed shift-registers and adders/divers.  So that is not built. But what I do have is programmable frequency generator (1Khz - 67Mhz) controlled by Arduino (Atmel) microcontroller, which is ultimately controlled by Java computer program using Serial over USB.  So I hope this helps get more ideas out, feel free to check my blog or msg me, I am always interested in collaborating.



This was the original design. The main design change now is an OPA633 high-speed buffer before the MOSFET.



The 1khz - 67Mhz LTC6904 oscillator chips were MSOPs, so I soldered them to DIP converters.



This is the Arduino Microcontroller (Atmel), easily controlled over USB.  The Arduino talks to the LTC6904 oscillator using the industry-standard two-wire I2C bus.  Perhaps the only better bus protocol would be CAN (the automotive bus).


This shows the results of the first test, which was a steady 1.039khz square wave.  We were very close.


This is the code for the Arduino, showing how ridiculously easy it is to program.  What you see sets the frequency to 8.21 khz. The entire program is pictured, there is no other code.


Lanmasterd, are you using a computer (serial over usb) to talk to your microcontroller? 

If so, Free Energy Group is trying to make open-source standard for computer <-> microcontroller communications, so we can have modular Java control software which works on all our devices.  Please let me know if you are interested in designing a common control protocol.  The draft is based on get/set.

GK/Feynman/Free Energy Group Microcontroller Control Protocol v0.1draft


----==== Set Commands ====----

syntax: S,variable,value

S,E,0 #Set Enable to False (stop oscillators)
S,F1,21400 #Set Frequency1 to 21400 hz
S,F2,315200 #Set Frequency2 to 315200 hz
S,F3,40323 #Set Frequency3 to 40323 hz
S,E,1 #Set Enable to true (start oscillators)

----==== Get Commands ====----

syntax: G,variable
response: value

G,E # Get enable setting
G,F1 # Get Frequency1
G,F2 # Get Frequency2
G,F3 # Get Frequency3