Need a current regulator? (colloidal silver generator)

[Serano]

I have tried voltages up to 36 with 15 mA output, but not for long as I noticed the transistor getting hot, I actually only want 5mA and that was what the last one was set to, but it seem unstable when I went back to check it at a lower voltage (9v) it seemed to show 10mA so I tried resetting the pot (live) and it just went to overload, I changed to 10 A mode and it appeared to be reading 1.3Amps :-o then it started to smell.  I don't know what type of variable resistors they are, just cheepies off e-bay, so I'm highly suspicious of these.  I do have some 1 watt standard resistors on order but they are taking a long time to arrive.  My aim is to use 5mA with 36  volts to minimise the required electrolyte.  My transistors are wired 180 degrees to yours, I found a way to determine which is the emitter and collector.  I don't think it would have ever showed an output if either of them had been around the wrong way, correct me if I'm wrong on that.

P.S there are also data sheets showing the opposite pin out, not sure who made mine.

[Kephra]

Wired backwards, the transistors overheat and burn out.  Check the datasheet pdf I referenced in the above post.

Also, if you are going to use a variable sense resistor, I would recommend putting a fixed resistor of 39 ohms in series with it to prevent going over 15 or 16 ma.

[Serano]

Good tip about the "saver" resistor.  Now you got me wondering again about the transistor orientation, do you remember our previous discussion about the first time you wired these circuits and how the transistors were around the other way.  The method I found is to measure on the multimeter diode setting between the base and the collector/emitter, apparently the emitter gives a slightly higher reading.

[Kephra]

Yes, I remember, and I know the technique you are speaking of, and it should work.  However, the absolute safe way is by the manufacturers specification.

I think the source of confusion is that the TO-92 package is used by a lot of different transistors, and their pin-outs vary.  So if you look up a TO-92 package, you will see different emitter-collector pinouts.  But all 2n2222 transistors should be the same.  A Fairchild PN2222 is backwards to the 2N2222 though according to its data sheet, and I would have thought they would be the same. 

[Serano]

Yes I'll bear that in mind, if I get a similar problem with the standard resistors I'll try reversing the transistors.  It'll either do nothing or go bang...

[Kephra]

OK, I went back and looked up my order from my supplier.  I ordered ON Semiconductors P2N2222AGOS-ND.  This particular part DOES have the pinouts reversed from a standard 2N2222.  So your orientation with the backwards D is probably correct.  Go figure... I bought mine from Digi-key, and the only ones they sell are backwards to standard.  I made my circuit boards to match the parts I bought, so mine work fine.  In the future, I think I will get MPS2222As to avoid this confusion. 

[Serano]

I finally received the standard resistors.  I did try replacing the transistors again with the variable resistors still in place, initially ok but then after a short while "bang" and blew the top clean off the bottom transistor.  I replaced all of the components except the diode (which has survived everything I have thrown at it) I used a 47K resistor for R1 and 100 ohm for R2.  This gave 5.5 ma when output shorted with 9v supply and 5.7 ma with 18v supply.  I decided with the higher voltages to use a load so with 27 and 36 volts and a 4.7K load I saw 5.7ma. A 10k load at 36 volts, as anticipated,  showed  a reduction in output current to about 2.5 ma as the resistance moved out of the controlling region.

I was aiming for about 5ma and this circuit now seem to be exactly what I expected. :-)   I think the variable resistors were probably causing the earlier problems they may only be rated for low voltage or may change due to heating etc. (the were only unbranded cheap and cheerful ones) Also, I think its a bit dodgy shorting the output with anything approaching 10ma, fine with a load but without one the top transistor (Q1) always seems to rapidly heat up. 

With a 5ma supply I can produce 1mg of silver per 3 mins so I can make 1ltr of 20 ppm in an hour or half a ltr of 40 ppm in the same.

What I intend doing is the second option:- I will make half a litre of 40 ppm using table salt as the electrolyte and Cinnulin, extracted from Cinnamon as described by Kephra, except I used 2 oz of Vodka rather than 4, (which seems to produce a more concentrated deeper red extract) I like to drink my vodka ;-)

If you do make the above circuit, a few tips:-

1) DONT use variable resistors.

2) Avoid pushing the circuit too hard i.e testing by shorting the output with high values of voltage and current,  15ma at 36 Volts for more than a few second is risky, keep you finger on the top transistor if it starts to warm rapidly disconnect.  If it gets too hot to keep your finger on it's probably already too late :-(

3) you may well find, as I did, that your transistors have to be fitted 180 degrees rotated to that shown in Kephras photo.

[Kephra]

1) DONT use variable resistors.

Good advice.  But if you do, put a 39ohm fixed resistor in series with the variable sense resistor. 

2) Avoid pushing the circuit too hard i.e testing by shorting the output with high values of voltage and current,  15ma at 36 Volts for more than a few second is risky, keep you finger on the top transistor if it starts to warm rapidly disconnect.  If it gets too hot to keep your finger on it's probably already too late :-(

At 15ma and 36 volts, the power dissipation of the upper transistor is just at its limit for continuous operation with a shorted output.  I do test my units with only the milliammeter connected, but I do not leave it connected longer than required to get a reading.  Also, the 2N2222A should be used (which is the one I used) as the 2N2222 has less power handling capability.

3) you may well find, as I did, that your transistors have to be fitted 180 degrees rotated to that shown in Kephras photo.

Yes, always check the manufacturers data sheets.  I am totally mystified why this is not standard.

[rumblefrost]

Donated.

[Kephra]

Sent by first class mail on Dec 22.  Let me know when you get it.

[rumblefrost]

Thanks Kephra. I'll be putting it to good use.

[Kephra]

OK, I have no more of the regulators.  If I decide to make any more in the future, I will post a note here.

[mraluma415]

I donated $30 for your continued dedication to helping people achieve optimal health through self-healing.
Keep up the good work Kephra!

BTW I forgot to enter my user ID..but here's my name

Jarex

[Kephra]

I donated $30 for your continued dedication to helping people achieve optimal health through self-healing.
Keep up the good work Kephra!

BTW I forgot to enter my user ID..but here's my name

Jarex

Thank you very much!  If I make any more regulators, I will see that you get one.