Sciencemadness Discussion Board

Power Source

Darkfire - 26-3-2003 at 15:35

If i use a cord cut from an old lamp as my power source will this work, or just get me killed?


vulture - 27-3-2003 at 13:20


Are you serious?? I hope not.

It's no use for electrolysis because you'll be using AC at a way too high voltage.

Apart from that, it's dangerous.

Darkfire - 27-3-2003 at 15:34

Yeah thats what i figured, im just checking, i have a battery charger at my old house which is what i planed on using that seems like what you all use. It will just be a while before i could go home and get it; so i had an idea that didnt seem good but i figured in a long shot it might be safe enoght but thats why we have these forums. And ive never done any electrochemistry before so im new.


Cappy - 2-4-2003 at 18:57

Electricity is caused by the movement of electrons. Electrons want to move from regions of high concetration to low concentration, because they repell each other, and are attracted to relatively positive regions. Direct Current is produced by causing a difference in electron concentrations. The electrons move in one direction from high to low concentrations as said before, making electricity.

If the electrons are moved back and forth short distances quickly, they will also cause electricity. This type of electricity is called Alternating Current. Obviously this isn't done with a constant difference in electron concentration.

Electrolysis uses a difference in electron concentration in order to cause ions to be pulled towards either electrode. One electrode has a higher electron concetration (negative), and the other has a lower electron concetration (positive). The positive ions are attracted to the negative electrode and the negative ions are attracted to the positive electrode. While AC will allow electricity to pass through the solution, the positive and negative electrodes will switch back and forth so quickly that no accumulation of ions can occur at the electrodes.

[Edited on 4/3/2003 by Cappy]

madscientist - 2-4-2003 at 20:34

Several years ago, I electrolyzed a sodium chloride solution with wall current (as in, what Darkfire was proposing). The electrolysis proceeded very rapidly - until its insatiable appetite for electrons blew the power out!

Darkfire - 2-4-2003 at 20:40

Thanks Cappy, that really explained it well, after making a kew post i leave enlightend.


Cappy - 3-4-2003 at 01:24

You're very welcome. I seem to be dishing out far more questions than answers around here, so I'm glad to help when I can.

You can make use of wall outlets if you have something to convert AC to DC (like electric train set transformers/controllers).

Batteries create DC of course, so you can just take some 1.5 volt batteries and tape them together in one long series. I don't know if the batteries will overheat if you run too many in series. I'll have to experiment.

I've used a model rocket launch controller (4 1.5 volt batteries for 6 volts) for an hour at a time with no overheating problems. The batteries weren't drained completely either.

[Edited on 4/3/2003 by Cappy]

Ramiel - 8-4-2003 at 05:58

I'm just making my own rectifier... it'll look like this…

<html><img src=""></html>

Now I can’t recommend you go out and make it just like this, it needs to be made just right or you’ll kill yourself right away or short the house… If you do it right, it’s just hellishly dangerous – which is acceptable to me. Now don’t laugh, but I know of a guy who killed himself with 8V and a few milliamps, it went straight through his bloodstream and defibrillated his heart… dead as a dingo [crikey, danger danger... mate ; ) ].

If you really like you could adjust the Current Limiting resistor to allow more or less amperage/voltage in (I suppose you could even put in a variable resistor right at the start). This sort of setup is perfect for me. How about you guys out there?

[Edited on 8-4-2003 by Ramiel]

[Edited on 14-4-2003 by Ramiel]

Organikum - 10-4-2003 at 19:16

For the linguistic depraved like me:
Whats MOT ?
I guess the T is for transformer.

The 25 bridge rectifiers will go up in smoke one after the other (fast, very fast). Because of the minimal differences in the diodes. The diode with the lowest breakthrough voltage will go first and on.....
Why don´t take a single rectifier from a old PSU (computer?) or from a batterie charger for cars? I cannot see the advantage of the concept and I cannot see the current limiting resistor and not why there is 1A at 240V?

But I have something for the brave the dead not fearing scientist:
from the book "Improvised Laboratory Devices" (on the FTP now), an arc furnace with exquiste load/power regulation! I don´t recommened this, realy not, but I admire the authors of the book.
(there is saltwater in the pot with the electrodes of the rheostat. my god!)

Attachment: Carbon_arc_rheostat1.tif (47kB)
This file has been downloaded 1265 times

PHILOU Zrealone - 11-4-2003 at 05:36

Metal Oxyde Transformer?



rikkitikkitavi - 11-4-2003 at 10:59


Typically 230V in, 2300 V out.
One of the secondaries is usually connected to the core.

check out for more info about what you can do with MOTs.

May I suggest old PC-AT powersupplies?
usually 5V @ 20 A, as long as you pull atleast 1-2 A from 5V it can supply 12V aswell. Can be series connected too since they are galvanically isolated from the 120/230 grid.


Organikum - 12-4-2003 at 09:52

The variable resistor is now present I see.
But I still don´t get the logic behind the 25x rectifier units with 25x variable resistors (which as now inserted indeed should save the rectifier circuits from burning out).

And the microwave oven transformer? Do you want to interchange inpit/output sides?


transformer calculation and more electrical information:
transformer calculator ad more

Ramiel - 12-4-2003 at 19:36

The rectifier circuitry in paralell is designed such that an equal current (25 rectifiers /25A = 1A per rectifier) flows through the diode system. Note:
1) the 1A diodes will not encounter breakdown if they are forward biased, and only at a reverse bias voltage of 10V or more will breakdown occur.
2) The actual circuit should be made such that the resistance from the Microwave Oven Transformer (which I might add has had the direction of transformance swapped) and the rectifier diodes is equal.

Okay, so I'm not an electrical engineer (yet!) and this is a work in progress, but i have been assured that this design will work. Shortly I'll get some exact values for the shunt capacitance and so forth based on a normal rectifier diodes sold at radioshack (it'll be just like in the good ole' Anarchist's Cookbook! ; P ).

In earnest

Organikum - 14-4-2003 at 15:11

A usual MOT transforms 240V to about 2200V what is a relation of 1 / 9,17. Reversed this would produce 26V AC output on the now secondary side of the transformer.
If I push 5 ampere through a 1 ampere rated diode it will burn out. Forward biased. Any load or also the diode which suddenly materialized in the drawing (V-diode) will hinder this. As I wrote it will burn out nothing was present to protect the rectifier circuits.
(this way of yours how to edit your post makes things not easier)

But what is the sense of using 25 rectifiers? Where is the advantage? Why? Rectifiers rated high enough are for free or cheapo as can be. Makes design and building far easier and effort could be made in stabilizing the power output so you get a more universal laboratory power supply. (filters, stabilization, power factor control and and and...possibilities without end.)

btw. the fact that MOTs are often grounded (N on core) to the ironcore on the original secondary side should be considered for circumventing surprises of the special kind.

My opinion:
AT PSU. Sense line manipulation used for voltage adjustion.

Ramiel, I am not here posting for to take you down in any way. If you want me not to comment on posts of yours in future PM me and I will do so. No problem.


Ramiel - 14-4-2003 at 20:49

<html><u><b> IGNORE THIS POST IF YOU GET BORED EASILY</b></u></html>
How could I not enjoy your response to my posts? Your responses have been instrumental in clearing up design faults that I would not have seen; by all means, keep it up!
I also apologize for changing the picture over, it does make things harder, sorry!

I do agree that a power supply unit could provide all the right voltages, but if I am not mistaken (which is quite possible!), it is still an alternating current. For electrolysis, you would still need a Direct current; which would involve some sort of rectification.

If one were to re-wire the Microwave Oven Transformer with high amperage wire (which is necessary because the smaller wires would easily burn out) one could at the same time change the coil ratio such that the conversion ratio is somewhere more in the order of 1/27 in the now secondary side of the transformer.

Yes, I am aware of the grounding configuration of the Microwave Transformer! That is why I strongly recommend you not to attempt making this from the diagram yet!

The reason I recommend repeating the rectifier is to split up the large current so that the poor little diodes can handle it. Note that the tiny resistance in the wire will ensure that the current through each rectifier circuit in parallel is roughly equal. So if you had 500 amperes, and 5 A diodes, you might want 100 or more rectifier circuits (as an example). For my circuit, I use the example of 1A diodes.

The negative V<sub>DIODE</sub> is the voltage drop associated with silicon diodes, I just included that for completeness.

Now someone made a comment on filtering. Just for fun, let’s consider a shunt capacitor at the end of the rectifier circuit that could filter our rectified voltage…
Considering that;
<.>The electrolytic cell can be approximated by a resistance (call it the Load Resistance R<sub>LOAD</sub>;)
<.>Time from the end of capacitor charge to the end of capacitor discharge = t<sub>1</sub> =~ Rectified current Period, which equals in my case (on Australian mains), the frequency = 50Hz, then rectified effective frequency it is 50Hz x 2 = 100Hz, hence the <html><b>period is 1/100 s</b></html>
<.>V<sub>PEAK</sub> is about 8V for this example.
<.> C is the capacitance value we need.
SO, after all that, we use the equation:
V<sub>RIPPLE</sub> = V<sub>PEAK</sub> - V<sub>PEAK</sub> . (e^(-T/(R<sub>LOAD</sub> . C)))

Now, we want the ripple voltage to be somewhere within say %25 percent of the peak voltage (which is rather a large ripple voltage!). Then V<sub>RIPPLE</sub> = 2V
R<sub>LOAD</sub> can be approximated to 120mOhms

Solving gives a value of 1/3 Farrads or more. This is an awesome Capacitor, I’d love 25 of these, but I’m afraid that would set me back $10,000 plus the cost to hire the crane, so I’ve just given up on filtering and other fancy stuff for this model.

Organikum - 15-4-2003 at 04:05

Sorry , had some bad days lately - this got through to my posting style I fear.....

Ok, the rewiring of the MOT changes the situation - understood. This adds to the effort which is in my humble opinion in no relation to the outcome. As told with equal costs - perhaps even less and the same effort on time and soldering you can build something superior.
my 2 cents.

For the staging of rectifiers has no advantage at all and diodes or readybuilt bridge rectifiers capable to handle the current are asscheap to free (electronics scrap) I would skip this part. Many possible points of failure added for nothing - this is against all logic and safety basics.

If you like to have something special for this I would suggest to build a motor - dynamo device where a AC motor drives an DC motor which works as dynamo and produces the wanted current. Works fine, no joke, was in ancient times a usual constellation. A heavy flywheel makes it perfect for such an application.
Old motors are everywhere....

not bad, isn´t it? ;)

I am a fish - 30-4-2003 at 14:02

Another potential power supply is a lighting transformer (i.e. one used to power those minature halogen lamps that are sometimes used in shop displays). They are cheap and readily available. They generally give a 12VAC output, which is ideal for feeding into a homemade power-supply circuit.

Power source

Theoretic - 17-6-2003 at 07:15

You can use a phone recharger. 6V of DC
with a few hundred milliamps - good enough for any electrolysis.

not really

Organikum - 27-6-2003 at 07:39

or theoretic only. Telephone chargers (not chargers for lead accumulators) are selfrestricting for protection of the mobile and the LI/Io cells. If you don´t have a perfect electrolytic cell with extreme low inner resistance the charger will deliver almost no current as high inner resistance tells him "battery full".

True for all chargers of GSM phones I have seen and should be general true for all chargers of "not lead" cells.

Interesting are chargers which deliver DC with overlayed AC for some applications. For the suggested milliamp/6V I would advise to use a pack of rechargeable batteries directly - works better and you won´t ruin a phonecharger.


Theoretic - 30-6-2003 at 03:53

I don't mean the small ones that you put the mobile right in. My one looked like a battery charger, but with an output cord.
An old one, I presume.:)

GZAust - 11-7-2003 at 05:59

Hi new here. hehe and as your about to see, I'm lame-indescriminati-occasionalli!

I would love to make a variable power supply with rectifier.

But these are the things I would need for my sanity before setting out on this great project to be proud of:

- A good solid schematic before anything.

- Since it is high voltage, I think an analog or even better, a digital indicator of the voltage, amperage and wattage at the outputs.

- A 2-position switch to change from AC-DC and DC-AC.

- A beep and flashing ALERT of some sort when working at higher power.

- EDIT: An appropriatly placed fuse for this baby.

- A good-looking box, a used mini HiFi box perhaps, with a custom face piece.


What?? You want to make it worth keeping don't ya?

Ramiel! I love you enginuaty and initiative, but don't you think there would be some appropriate schematics around that are tried and true? you wouldnt like to make any mistakes in your calculations.
Is digital display a dream?

Organikum! What kind of power to PC power supplies put out in DC and at what amperage?

What kind of places would old electronics junk be traded? What to look for? Transformers are expensive, as are all High power stuff I think, isnt it?:D

[Edited on 11-7-2003 by GZAust]

kryss - 15-7-2003 at 13:22

If its cheap transformer kit your after I'd look on ebay , theres plenty of old stuff for sale - amps etc with rectifying circuitry in - just make sure you dont end up electrocuting yourself!

Organikum - 16-7-2003 at 00:05

PC powersupply:
- old AT style: 12V, 5V, -5V, -12V rated at about 200Watt most I knew (tower, not desktop)
5V ~ 15Amp
12V ~ 7Amp
-5V forget it
-12V ~ 2Amp
Data is written on the PSU itself. Old AT computers are everywhere for free (386/486/PI). Just ask around. Scrapyard with electronics scrap of course.

Tricky detail: It is not good to try to run full power output on 5V and nuthing on 12V and other way round. Connect a load to the not used voltage, a light, something pulling at least one third of the maximal output on this voltage.
Voltage regulation in certain limits by the "sense" line - do a websearch for your special model connectors layout this differs sometimes. And the so called "power good" line must get grounded.
By attaching an resistor to the senseline the 5V power can get cranked up to about 7V and by attaching the sense line to 12v instead 5V and an appropriate resistor you can go down to 2,5V mostly.
Yes! a varistor and on the 12V line makes it adjustable over the whole possible range. For the answer on what resistor/varistor to use please consult Mr. Ohms law. ;)

All output is DC and in the safe range. (with serious effort it might be possible to kill oneself this way but I see no way this can be done by accident). Don´t open the PSU. Thats all. Get two or three and throw away defect ones - repair is not feasible.

I had a PSU from an old INTEL server (286 based, multiprocessor network server ~ 40kg - outch) this was rated at 80Amp on 5V. Electroplating easy. :D
Here actually is true: The older the better built are the PSUs regarding the new use.

[Edited on 16-7-2003 by Organikum]

Ramiel - 20-7-2003 at 00:54

If you pull some 2A from the 12V as you say organikum, won't the resistor get really warm? I'd reccomend a heat sink. 25 V.A is quite a bit for one little resistor. Or what about a few 30 ohm resistors in parallel, that ought to solve the problem.
I only mention this because 30 Watt resistors are notoriously difficult to locate ;)


Organikum - 20-7-2003 at 05:33

There is no resistor for 30 Watt in my layout. There is a load, a light (icandescient lightbulb) to attach to the not used power output. Lightbulbs are perfect.
The resistor on the sense line can be any by hand as there is is nearly no current going through - it´s a sense line.


tryptamine - 20-7-2003 at 20:49

A rewound MOT is the way to go here(better yet a bank of rewound MOT's), you want high amperage. Using any other transformer anything you are trying to accomplish will take forever and a day. Searh online for a guy that made a huge ac welder using 8 MOT's. If this doesn't convince you of the suitability of these transformers nothing will.

Rectify your transformer output with a bridge of high amp diodes or an SCR/pulse generator combo.

There is no other practiucal way of doing this short of going out and dropping several hundred on a heavy duty rectifier.

I'm building a power supply like this right now in fact, I'm probably going with the SCR setup cause I might be able to get some enormous Toshiba's or Fuji's from work, plus it makes it easy to vary the load.

voltage concentration etc.

fogus - 17-10-2003 at 21:43

i am absolutly new to this sight, i hope that this works (i have had worse luck with other sights)
While working with electrolising potassium perchlorate I have herd that a good electrolising solution is one which uses a one third concentration of salts (potassium cloride by weight) and a low voltage of 3V is most efficient. I got this from a smart person and i was wondering if this was true or not.

i have been trying to make potasium perchlorate for a long time now and i might have my first batch today:D i am waiting for it to dry. i am hopeful that it electrolised properly. if it didnt i wont give up. this will only be one more thing that doesnt work.

old AT power supply problems

blazter - 19-12-2003 at 14:05

I managed to scavange an old AT power supply from a scrapped compaq computer. It doesn't seem to have a wattage rating, but it does claim to be able to supply 14A at +5v. This power supply would seem to be perfect to resurrect my chlorate cell which ate up a 25A bridge rectifier after about an hour of operation. With any luck i'll have a better, voltage regulated supply this time around.

Anyways, I found some pinouts to the motherboard connectors, and it seems to check out when they are metered. This power supply appears to run without a load with no problems. The problem which I run into is that the voltage drops dramatically when the electrodes are immsersed in the salt solution. It appears to go as low as a fraction of a volt. This effect seems to depend on how far the electrodes are immersed. When only a few mm below the surface things seem to go well with the voltage slightly below 5v and many bubbles.

In an attempt to fix what appears to be a regulation problem, I shorted the power good lead (pin #1 orange) to a ground. In addition all the grounds seem to be common as they all give continuity to the chassis. This seemed to do nothing at all in regulating the voltage.

The other thing that should be noted, is that based on the amount of gas evolved by the test cell would indicate that there is no where near 14A going through the outputs. Just a wild guess, but its probably less than 1A based on the amount of bubbles produced (more on this when I hook up an ampmeter).

In an attempt to slightly load the +12v and +3.3v lines, I stuck small 12v pilot type bulbs on them. The funny thing is that the one on the 12v line seems to glow brighter as the voltage decreases on the 5v line.

Also, if I can ever get this working, I plan to hook up a 60w 120v lightbulb in series to the cell. The idea here is that the bulb will limit the amount of current which the cell can draw (hopefully) protecting it. In theory, running at 5v the bulb should allow at most 12A to flow through it.

Any suggestions on how to get this thing to supply the current which it claims to would be greatly appreciated. If all else fails, or it goes up in smoke, theres another AT power supply waiting :)

unionised - 19-12-2003 at 16:20

You can get a 35Amp bridge rectifier for £1.29
How much current do you need?

the more the merrier

blazter - 19-12-2003 at 17:24

For this chlorate cell which consists of salt solution in a 1 gallon glass jar, the more amps the better. Just getting 12A from the power supply which I have on hand would be just fine.

I'd really like to use these AT power supplies since they are on hand doing nothing, and are free. Some others have reported success with them, and I would be very interested in any secrets to getting it working properly.

link to super do it yourself electronics

Mr. Wizard - 19-12-2003 at 19:42

This guy has a site with all sorts of ingenious do it yourself power supplies. He seems to specialize in redoing older items that are available:
he has no connection with Mad Science

Saerynide - 29-12-2003 at 06:41

I hooked up the 12v at 8 amp pins on an old computer power supply and got no luck... It doesnt seem to work unless its plugged into the comp... I tried pluging in a cdrom and floppy drive and still no luck :(

Any hints?

I am a fish - 29-12-2003 at 09:24

PC power supplies will only deliver power to peripheral devices when instructed to do so by the motherboard. With ATX power supplies, this can be done by shorting the green PS-ON lead to any of the black ground leads.

Saerynide - 5-1-2004 at 02:25

Wow, this is strange. After shorting the green to the black, the case of the power supply is now electrified :o Is that supposed to happen??

And why is there no negative lead for 3.3 volts? On the key, there is no negative compliment for "+3.3 v 14 amp", only something that says "+5 Vsb" Does anyone know what that means?

Btw, my power supply works great for electrolysis now, but it cant hurt to have an extra voltage/amperage choice :D

Star - 8-1-2004 at 16:38

After you have got your hands on a AC power supply[which basically is a transformer ] ,you must be able to regulate the output right?
The law of Ohm :I=U/R

So if you want to have the right amperage you must know the internal resistance of the electrochemical cell right?[whats the resistance of a NaCl solution?]

I am a fish - 9-1-2004 at 01:45

Originally posted by Saerynide
Wow, this is strange. After shorting the green to the black, the case of the power supply is now electrified :o Is that supposed to happen??

I don't think so. It's possible that your power supply is faulty and that one of the previously switched off power lines has shorted to the case.

Saerynide - 10-1-2004 at 08:46

I had 4 different ppl touch the power supply box to see if it was electrified. 2 said it was, 2 said it wasnt and that it was only the vibration of the fan :o

I think I'll just stay clear of the case irregardless of its ability to conduct electricity :D

CommonScientist - 3-3-2004 at 20:27

I used a 3A 12V battery charger. I have a IBM PC pwer supply, pics of my cell and power supplies

The only problem with it is that it ran in cylcles, on and off. It would shut off, turn on , so I dont know how long it actually ran. I have some RC car battery chargers. TYCO R/C one is 11.5 V 1.085A, which is nothin really.