I need a high current DC to AC

M

manie

Senior Member
Working from 12V or 24V DC (cars and tucks) I now need to look at half wave rectified (to retain the ripple as speudo "pulses") stepped up AC. The idea is to step 12V/24V up to say 110V AC as these transformers will probably be available ex the US and also the energy requirement into the gas production cells requires lower Amperage at higher Voltage.

I know that transformers are linear devices for Voltage, stepping down one way and stepping up the other way, how does current capacity fit into this ?

How difficult (or impossible) is it to generate reasonably good Sin Wave AC to be half-wave rectified ?

Any sites or links I can look at for good circuits and ideas ? Your ideas especially welcome.
Manie
 
C

centrex

Senior Member
Hi Manie
It would be interesting to know what you are up to.
The current out of the low voltage supply is going to be related to the power you want at 110 volts plus losses ie 110 at 10amps is 1100 watts that would be over 90 amps at 12 volts.

Why not use a relatively cheap modified sine wave invertor they put out square waves to look like sine.

c
 
S

SilentScreamer

Senior Member
manie said:
I know that transformers are linear devices for Voltage, stepping down one way and stepping up the other way, how does current capacity fit into this?
Click to expand...
Transformers are very efficient. The power in is approximately equal to the power out. So if the voltage drops by half, the ampage doubles.

So:
12V 20A in = 240V 1A out
240V 1A in = 60V 4A out
 
A

allison1947

New Member
Mannie do you need AC or DC for the gas producer and at what voltage?

If AC what frequency? Why do you need the "ripple"?

Two approaches come to mind, either a cheap DC to AC inverter or a DC to DC supply. Both can be made using power transistors and transformers.

Allison
 
M

moxhamj

New Member
Sine wave inverters have come down a lot in price. They are now much cheaper than you could possibly build them yourself. You can get true sine or modified square (the latter is a bit cheaper but can damage some devices). Inverters used to be something you would only find in an electronics shop but they now are in camping stores and car accessory stores and all sorts of places. They come in a wide range of wattages. I have a little 50W one and a big 1500W one. Diferent sized power for different applications. There are even ones the size of a coke can that fit in your drinks holder in the car and plug into the cigarette lighter.

Once you have your mains voltage you can do all sorts of standard things with that. Eg, I wanted to charge a 24V battery from a 12V battery. The cheapest way to do this ended up being a 12V to 240V inverter, and then a 240V 24V battery charger.

If you wanted 15V half wave rectified AC from a 12V DC supply, it could well be easiest to step up to mains with an off the shelf inverter, then use a standard mains to 15V transformer and then a standard half wave rectifier using diodes.

If you do step up to mains, treat it the same way as mains, ie don't go building circuits that have mains on them that you don't understand. Personally, I keep mains in 'standard' plugs and sockets, ie I plug my 15V wall wart directly into my 100W mains inverter. Now I have 15V DC from 12VDC, and this didn't require any electronics knowledge at all.

What do you need to do?

Re current and volts, if you multiply the volts by 10 then you divide the current by 10. And then subtract 10% which is the energy you lose every time you do a conversion through an inverter etc.
 
A

Andrew Cowan

Senior Member
I would agree with DR Acula - buy yourseld a cheap invertoer, and be done with that. It will be cheaper than making one yourself.

A
 
D

Dippy

Moderator
" (to retain the ripple as speudo "pulses")"

- eh? Whats this for?

Just buy one Manie. They are cheap as chips and it'll save pages and pages of Forum webspace too ;)
 
M

manie

Senior Member
At this point it looks like I'll probably need around 50Hz freq. The idea is to place not more than 1.5V across a single (+)/(-) plate pair. Water is broken down into hydrogen and oxygen at 1.23V, excess voltage (power) is wasted as heat production. In addition, pulsing the current is beneficial (seems like) in terms of better efficiency (gas volume per watt) and thus less heat loss and more gas per watt energy.

I was asked by someone to look at feeding 100 plate pairs (cells) in series, with "pulsed" (ripple) DC of 150V, the power source is either 12V or 24V automobile alternator with capacity of 160 Amps (12V) which is 1920 watts. I have succesfully produced a LOT of gas with pure DC, at 3V/cell (4 cells in series/12V) running at 142 Amps max. This is with 6 containers each drawing +- 22-24 Amps.

So in the end, as per Dr.A, 30A@ 12V(360W) in --> 150V @ 2A(300W minus losses[60W]) allowing for 20% inefficiency.

A 28x1 or 40x1 will monitor and control the various voltages etc. Does this help more ?
Manie
 
Last edited:
B

boriz

Senior Member
Ah yes, the old HHO ploy. *In an Inspector Clouseau accent*

Did you tell your friend that all the extra energy gained from the burning of HHO in the engine is used up generating the gas. In fact, due to various losses, the sum is actually lower fuel efficiency?

The only way to improve fuel efficiency is to generate the gas elsewhere. But then of course, you will need to take into account the cost of electricity etc. And the fact that HHO is very dangerous (probably illegal) to store. And the fact that the increased cylinder compression will significantly increase engine wear and heating. And the fact that you could blow yourself to bits Etc.

Apart from all that. Sonoluminescence is the way to go…
 
J

jglenn

Senior Member
If the source really is a car alternator, those already have 3 phase ac. There is an internal bridge that turns it into dc. You can tap into the windings before the bridge and get ac, which could be transformed. Break into the world of 3 phase transformers! :eek:
 
M

manie

Senior Member
There's that word again ! THREE phase, I understand and can work with single phase 220V 50Hz......... 3 phase though ????????
Manie
 
P

pilko

Senior Member
Check "allaboutcircuits.com" forum there are several guys working on WFC's Search "WFC" also search "Farlander"
 
J

jglenn

Senior Member
Manie, 3 phase is not scary, just more windings and nodes. Tesla built the first large scale devices, 3 seems to be the minimum for efficient operation of power transmission. To turn 3 phase ac into dc, you need 2 diodes per phase, so that is 6 total. If you are handling a lot of power, put 3 of them on one heat sink, the other 3 on another (one set will be "R" version stud package, anode/cathode is reversed). These heat sinks will be the + and - out busses.
 
I

inglewoodpete

Senior Member
Oh dear: Brown Gas.

I prefer Alchemy - so much more scientific and practical. I have so much gold I could destroy the gold market. I have to hide it.

Manie, I'm sorry to say you're on a hiding for nothing. Please don't injure yourself in the process. The components of water are quite dangerous in the presence of electricity.
 
M

moxhamj

New Member
If you want 1.5V per cell, then you want to step the voltage down, not up. Putting things in series won't work - water is conductive and will short out your high voltage. Unless you have each cell insulated from the next. But then if you are doing that, there is no point in having high voltage. Plus high voltage and water is dangerous++

So, you want 1.5V DC with a bit of ripple? Ok, use an inverter to get mains, then use a standard off the shelf transformer to get down to low voltages. Ideally, one in a box.

I'm not sure what you want this gas for - others here have made some guesses though. But you might like to do some experiments first with a low current setup. Sure, you can make lots of gas, but what you will find is that your metal plates will corrode very quickly, unless you use platinum or gold or graphite. Also you will need an electrolyte to increase the conductivity. Sulphuric acid is one choice. You could use salt, but then you get chlorine instead of oxygen off one electrode. Hmm - all the fun and dangerous chemicals. Personally, I'd be keeping this to test-tube sized experiments for the moment.

Just as a quick experiment, take two copper wires and put them in a test tube and hook them up to a power supply. Anything really, 3V to 12V. Run it for three days and then see if any copper is left. Then try with steel and aluminium. They all disappear into solution.
 
B

BeanieBots

Moderator
As Doc says, you will get a lot more than just H & O.
PURE water does not conduct at all. It's a very good insulator. Adding salt will make it conduct but you will get lots of chlorine. Tap water will also produce lots of chlorine and a few other nasties which will all speed up the electrode corrosion process.

As for you electrical issue. Square wave or psudo sine into a transformer is not a good thing to do. Applying half wave rectification afterwards will be a disaster. It will cause a net DC current in the transformer which eventually cause permanent magnetisation of the core and reduce the efficiency dramatically. So, the 50% efficiency of the half-wave coupled with a magnetised transformer trying to transfer square modualtion will give about 20% efficiency if you're lucky. If you're not lucky, the magnetised transformer will just pop a fuse.
 
M

moxhamj

New Member
Brown's Gas, Inglewoodpete? Thanks for the link - I looked it up. Seems to have pretty much all the standard scam stuff. Plus anyone with more than Year 8 Chemistry will be able to spot the flaws. I am kind of concerned though about the statement below. I might be out of a job!

"We're learning that water can be 'programed' to pass on health. This is easy to do with the water formed by the 'exhaust' of the Browns Gas flame. The production of Browns Gas 'erases' previous programing and then we can impress new programing on the water as it is newly reformed. We can make water that gives a feeling of relaxation and well-being just by drinking it. "
 
B

BeanieBots

Moderator
Absolutely Doc,
I'd read about "programmed" water before. It's a miracle!

What I find hard to believe is that nobody has taken it further and used it to replace EEPROM. Maybe silicon is more abundant and that's why it's used in preference to H2O but surely it takes far less energy to program the much lighter water molecule than it does a silicon one? Aparently, the memory retention of water is greater than silcon.

Also, aren't we 5 days late to be discussing the merits of Brown's Gas?
 
M

manie

Senior Member
Thanks for the concern guys ! Fact is, Dr.A, after one year of running quite vicious currents through cells with 3Volt across them, there is zero evidence of metal corrosion. We use 302 or 304 Stainless steel plates. They do however develop a "bronze" sheen on the reactive side and with higher voltages the cells form a brown sludge after quite a long period. Incidentally, we use distilled water and NaOH (drain cleaner) as the elctrolyte, and before you get concerned, I've been busy in chemistry for a very long time, comming from the mining industry as Extractive Metallurgist. So I (we) do take appropriate safety measures. Pulsing the current through the cells does have a beneficial effect and produces more gas than straight DC. Whether the technology has developed far enough to really work in cars/trucks is still debatable, I agree. However, it does provide a chance to learn more about electronics etc. without having to pay for it out of my own pocket. Here is the funny part:

I (we) have seen GOOD improvement in fuel consumption, HOWEVER, I (we) have also seen ZERO benefit. Figure out what caused both situations and then you have a winner.

From the metallurgical side, Dr.A, NEVER get Cl2 anywhere NEAR Stainless Steel. It will dissolve it as if it was not there. H2SO4 (sulphuric acid) however, does not touch stainless due to the Chrome oxidation layer acting as protection. I suspect this is also the case with the NaOH, producing an oxidised layer on the stainless, hence the "bronze" colour it develops.

BB: I take note of the transformer/magnetism issue, which is why I asked about Sin Wave generation in the beginning of the thread.

Dippy and others(all): Buying an inverter with enough "oomphh" (2kW and up) is totally OUT ! Here in RSA, anything decent starts off at US $900.00, that is entry price of ZAR 9000.00 !!! Totally too much to bring any monetary benefit to this issue.

As to the "healing waters", ya well !?? ! There are a lot of crackpots and very GULLIBLE people out there. If they're dumb enough to be caught out, so be it !

Your turn now....
Manie

PS: Since there are ENOUGH of these watertofuel and "free" energy sites I'd rather call this thread completed. The issue for was whether a homebuilt Sin Wave inverter is feasible. Being a "semi" believer in the Browns Gas issue, I'd rather close this now. If you concur, good.
 
Last edited:
B

BeanieBots

Moderator
Unless you have a supply of free components and time doesn't cost, you won't be able to make a sine inverter cheaper than you can buy one ready made (they are not simple things to design or build but have been around for years and are well established hence 'cheap') , so yes, that probably ends this thread.
 
N

nbw

Senior Member
I love the 'modified' sine wave tag some inverters have. It's like saying to a traffic officer when pulled over: 'I was travelling within a modified speed limit' LOL

Modified sine wave = marketing speak for very liberal, ugly, abstraction of a sine wave.
 
J

jglenn

Senior Member
Yep, Beaniebots has it right, transformers do not like non sine waves, or some motors too. Square, and maybe to a lesser extent the modified sine waves will cause heating in the core. Overall efficiency sinks like the Titantic.

Manie: if you can get surplus step up xfmrs, you could build a sinewave driver, sort of like an audio amplifier, but simpler. Use a center tapped winding, it goes to B+. Fets to pull down the other two sides. Maybe pushpull. From 12 or 24V, you can step it up to normal 120VAC levels. Also you need a sinewave reference, there are many ways to do that. If you build it yourself it will be cheaper than commercial. The challenge is to minimize heating, normal amplifiers are not as good as switchmode, but that gets about 3-5 times hairier.:(
 
B

BeanieBots

Moderator
I guess the approach depends largely on the overall objective which I thought (I may be wrong) was an efficiency thing.
A linear amplifier to produce a sine wave can't be better than 50% efficient and in practice is unlikely to be better than 30% and home-made probably closer to 20% efficient at best.

The only way to handle such power levels is to use a switcher. Making a switcher produce a sine wave output requires VERY good understanding of power supply bandwidth and control methods. If you don't know what a "bode plot" is, you ain't going to build one.

The other method is to produce an isolated switched follower which is the most popular method used by grid connected solar inverters. Essentially, this is an isolated switch mode power supply which "sits on top" of the mains sine wave (which is low impedance) and attempts to "pull" the sine wave along. If it were not for the low impedance, the effect would be an increase in fequency but instead the net effect is a slight increase in voltage. Not a simple project.

However, I don't think the real issue here is how to produce a low impedance mains voltage sine wave. It's how to produce a high voltage which can be pulsed. That's actaully quite easy. All you need is a "basic" switcher for which there are countless published designs. Forward push-pull being the most popular at higher power levels. The rectified and smoothed DC output can then be pulsed (via a FET and suitable driver) into whatever the load is.
 
M

manie

Senior Member
BB: I just did not want to go into the merrits or deficiencies of gas production here. This is my site for electronic and Picaxe related stuff so thought I'd cut it short. There are, as said A LOT of those site out there and I do not have time for absurd gas production claims.

Efficiency yes but only for the "gas" thing... it seems that a good Sin Wave inverter is not practical as yet. That makes it one hell of a challenge though, don't you think ?

Switching with Mosfets ? Nah, we've had LOOOONNNNGGGGG discussions already and my sponsor has opted out of that system. Anyway, I want to remain friends with Dippy...........
 
B

BeanieBots

Moderator
Well if you don't want to use a single FET to do some basic switching then doing the complex stuff to make an inverter is out of the question and Dippy can go back to bed without having nightmares;)
 
M

manie

Senior Member
NO WAYS ! Wake up the DIP !!! Nightmares are good, especially for Dippy....

Correction there BB: The sponsor for this particular project is out of Mosfets, I'm NOT out, so Dippy wake up !
 
J

jglenn

Senior Member
Beanie: I have built a 500W audio amp with about 8 parts, a sine wave osc, power opamp, complementary driver/booster, fet stage, and a transformer. Does not have the dozens of predriver transistors, resistors, caps, diodes, etc that normal discrete amps have. It was not designed to cover a wide freq range, was used as a bass guitar experimental amp, and I think may be suited for UPS purposes, although I have not tried it for that.

The efficiency driving PC type switching power supplies will be far above 50%.
Note the supplies have diode bridge inputs charging a cap to about 150V.

SO, they only draw current when the line voltage is above 150V. If you ever saw the current draw waveform of a switching supply, it is a current spike when the line voltage sine wave is around max. This is what aggravates power factor on the mains, one thing anyway, and why Power Factor Correction is such a big deal these days. Anyway, I only based it on a sine wave to get the power thru the transformer efficiently. Plus, sine is nice to have. True, with a resistive load, like a heater, it will not be efficient.
 
M

moxhamj

New Member
Magic sine waves are brilliant - the maths is so clever. In practice, what you need is a way of outputting the data quickly. Beyond a picaxe for sure, though you might be able to program the data slowly into a ram chip and then clock it out much faster and repeatedly once it is all programmed. The picaxe could read it out of an external eeprom etc. As it says in the instructions, you would need a clock of at least 2.5Mhz. This is well within the capabilities of ordinary PCBs and xtals etc - I'm playing with Z80 chips at present that are clocked between 4 and 8Mhz and that clock signal goes all over the board.

The problem is going to be driving a mosfet on and off at that speed. You can't afford to have any errors in the width of the pulses - they really have to be precise to less than 1 microsecond. That would involve some sort of drive circuit on the gate. And a fast mosfet.

Mind you, if you can use the magic sine formula to make sine waves, you can use a number of them running in parallel and adding to produce any sort of waveform you want. Fun fourier addition - try making a square wave out of 3 sine waves. I did it once in excel and it was kind of strange to watch the waveform appear as you added in harmonics.

I'm still not 100% sure what you need this for - is it to experiment with efficiencies of electrolysis with DC vs AC vs any sort of arbitrary waveform? Because if so, you can set up a simple little experiment using test tubes and an audio amp and set up a signal generator (eg from a PC sound card) to 'play' any arbitrary waveform (you can check it on a cro) and measure the output under different conditions. Who knows, you might discover a more efficient way of doing things?

But better to do the experiment with low cost equipment on a small scale before investing in large scale sine wave inverters etc? Also if you want higher volts, audio amps certainly can go to higher volts. The power supply of the audio amp will give a clue - your clock radio might have a 9V supply, but your proper stereo might have +/-30V or +/-50V. You can 'play' a low voltage sine wave through that at 3V and you can add in high voltage pulses on top of that. Maybe don't play it through real speakers though, it would sound horrible!
 
Last edited:
J

jglenn

Senior Member
How do you post schematics from ExpressSCH? It will only create
.sch and .bmp, neither of which the forum accepts.
 
lbenson

lbenson

Senior Member
I copy .bmp files into Paint Shop Pro and convert to jpg. Microsoft Paint can also do this. For .sch files, I do a print screen, paste it into PSP or Paint, clip it, and save it as a jpg. If you wanted to put up the actual sch file for people to use, you can change the name to a .txt and post that, with the note that people have to change it back to use it. Print-screen, paste, clip, and save as jpg can also work for the .pcb files.
 
J

jglenn

Senior Member
Seems to open. Let me explain the circuit. It was a result of a study to build a high power audio amp with the minimum of parts. I'm not saying it is a good general purpose audio amp, as it has only been tested at low freq, works fine at 60HZ. Ran a bass amplifier speaker pretty well, with guitar.

Usual discrete amps have very complex predriver circuits, I tried implementing an old idea used in tube circuits, transformer coupling of stages. Any crummy little power transformer can be used to drive the base circuit of the NPN/PNP pair, you need the right turns ratio, the base voltage must approach the +\- 50V bus. This circuit looks simple, but works, and cranks the power out.
Must have big heat sinks on the output pair, and one on the EL2008.

Note it is open loop, no fancy feedback, yet. The EL2008 is a very unique part, a power booster amp that has an amp or more output, at up to 50 MHZ if you need it. Yes, that is MegaHertz. Also add 5000uF caps for the 50V supplies at the output. A ground plane pcb is suggested.

This is not a 12V to 120VAC UPS, but can be made into one. Try using two 12V batts for + and - supplies, instead of the 50V. Then instead of a speaker, drive a step up transformer. You will get a really BIG sine wave.

:cool:

pdf conv:
http://www.pdf995.com/download.html

http://cboards.co.za/pcb_tips.html
 
Last edited:
M

manie

Senior Member
Dr A. & Jglenn: A lot of info there to digest. Thanks, I'll look it over maybe during the week-end.

I built a small two tranny oscillator today. Here are the Schematic and pics. From the pics you can see a very nice Sin Wave produced. It shows 9.02V on the dig.meter. I'm including a shot of thr CRO controls, help me interpret this beautiful instrument please....

Shematic and pics (2 at a time)
 

Attachments

M

manie

Senior Member
From the above pic... YES, I did it on Veroboards first, a first in itself for me ! Lo and behold, it actually worked straight out first time too !!

The other two pics...
 

Attachments

E

eclectic

Moderator
@Manie.
A genuine question.

Why are you soldering components for a temporary test circuit?

Breadboard / AXE091 .......................

e
 
M

manie

Senior Member
EC:
I have a breadboard available BUT........ it does not work very well for me. Too many intermittent/loose/dead/dry/sometimes/etc. connections, so I developed the habit of soldering, as of this afternoon on Veroboard, before as you know, straight to PC board artwork etc. This way I have 100% (almost) success rate, breadboard ?? I use to slice bread on...., don't like it...
 
You must log in or register to reply here.
Top