Simple power supply

David F. Spencer dspencer at is.dal.ca
Mon Sep 22 14:36:19 EST 1997


In article <5vub4d$d5v at cpca3.uea.ac.uk>, AndrewLeech at see.sig.for.email
(Andrew Leech) wrote:

> In article <5v74tp$qdi at news.cuny.edu>, kang at msvax.mssm.edu says...
> >
> >Dear colleagues;
> >
> >I recently developed new power supply which is totaly different from the
convent
> >ional one in that it does not use the core and coil and condenser for
transformi
> >ng AC to DC. Instead I used bridge diode just to change the AC to
directional pu
> >lse current. 
> 
> It's been done. We have a gel electrophoresis system called "Mupid-2", made
> by Cosmo Bio Co Ltd of Tokyo, Japan.
> Although the circuit diagram is not given in the manual, its characteristics
> are almost exactly the same as yours, and it does say "no transformer is
> used"...

And the Mupid is an extremely dangerous rig.  We have a Mupid (not a
Mupid-2) in the lab that I would never use and I have warned those who use
it that this design is nothing short of idiotic.  I just verified what I
already knew and that is that the negative output (in the case of the setup
we have, labelled 'BLK') is at full line voltage (here 115-120 AC) with
reference to any ground (in the UK, "earth") and that means air or gas
taps, sinks and faucets, and the metal cases of essentially any electrical
lab equipment which is grounded, in other words virtually everything sold
in the past 20 years in North America.  The current output of this "power
supply" would be more than enough to kill.

These cheap, simple designs are hardly novel and certainly not even worth
paying anyone to get a circuit diagram.  In 1964 the journal Ann. N. Y.
Acad. Sci. (vol. 121) devoted a whole issue to the then very young field of
polyacrylamide gel electrophoresis. In an article by Baruch J. Davis there
are presented two designs for power supplies, and one labelled "Simple 150
volt supply" is a circuit diagram obvious to anyone with any electronics
knowledge, a bridge rectifier (full wave) with a filter capacitor and a
couple of resistors on the line side of the diodes. It is called a 150
volt, 1 ampere power supply; 1000 mA should allow a fairly effective
electrocution.  Of course such cheap "power supplies" don't have any meters
because they are more expensive than transformers.

> >I made the beta version of this apparatus and wanat to sell with 40 dollars.
> 
> You might want to check it hasn't been patented first.
> 
> Then,
> In article <wqIErKAYO6F0Ewki at genesys.demon.co.uk>, duncan at genesys.demon.co.uk 
> says...
> >
> >This would be 'illegal' in the UK/EEC due to lack of safety and would
> >not be allowed to be sold. If one of the rectifiers fails short circuit
> >you can get full mains AC on the output with lethal consequences.  With
> >no current limiting this would kill. Given that a transformer and
> >condenser are literally only a few dollars surely it makes sense to use
> >that route?

This type of design would never meet standards in either the US or Canada
nor I'm sure in any western European country.  Given that electrophoresis
setups are used with conductive water solutions and that there is
frequently liquid spilled around the gel setup (which can leak as well) and
you've got the makings of a disaster.  And don't be lulled into any false
sense of security that there must be a component failure to cause a
problem; even in a full wave rectifier the output is only one diode from
the input, and the voltage drop across a silicon diode is only about 1.5
volts.  What you get hit with is a pulsed DC rather than a true AC but that
will give you a nasty bite none the less.

Indeed the cost of a basic transformer is trivial, even a simple 120 volt
to 120 volt .5 amp isolation transformer. Be warned though that variable
autotransformers (in the US and Canada sold under the names Variac and
PowerStat) have one line as a straight feed through and thus are not fully
isolated.

> I wasn't in the lab when it arrived, so where it came from...? Anyway it
> is well fused and has a cutout switch in the gel tank to stop you frying
> yourself.

Fuses protect electronic circuits not humans; a 250/500 mA fuse is totally
irrelevant when you are the load on the circuit.

> Ironically though, because the mains supply in the UK is 230V, it has to
> be supplied from a small step-down transformer, so it is isolated.

So they sell this cheap setup with a stepdown transfomer for use in Europe
where the line voltages are 205-250?  That's brilliant marketing, having an
external transformer worth more than the standard power supply.

> Personally I don't like it - but that may be because the gel trays are
> rather small and fiddly for cutting out bands, and of course you only have
> "slow" and "fast".

No one who has any knowledge of/respect for electricity would use a stupid,
cheap, live output "power supply". What amazes me is that we don't hear
about lab personnel being killed by such treacherous setups.

Dave Spencer

-- 
David F. Spencer, PhD
Dept. Of Biochemistry
Dalhousie University
Halifax, Nova Scotia
Canada

dspencer at is.dal.ca
dspencer at rsu.biochem.dal.ca



More information about the Methods mailing list