"BillW50" wrote in message
. com...
In ,
Arfa Daily typed on Mon, 31 Mar 2008 00:37:00 GMT:
[...]
And Bill.
I will now explain why your contention that your method will work
under all circumstances, is not valid.
Sorry if you got that impression, but that isn't what I meant.
You are quite wrong with your
assumption that all of the circuits in a piece of equipment are
stacked up in parallel across the DC input socket. Whilst this might
have been the case some years back, the DC connector on "modern"
(infer whatever period you like from that word) equipment, usually
connects straight into some form of internal ancilliary power supply,
or a regulator or regulators, which are often switching types.
Yes I remember.
The reasons for this are manifold, but include the fact that most
modern equipment does not contain circuitry that runs just from 12v,
which is a typical 'standard' value for external power unit
equipment, and also efficiency, which dictates the regulators
typically being switchers. The various circuits contained within the
equipment, are connected to the back end of these regulators, and are
thus not connected to the DC power socket in any way.
Yes but the regulators are.
Often, the input to this regulator circuitry, contained within special
purpose ICs, will not produce any meaningful ohms reading, when
subjected to the low test voltage from a multimeter. So, you are just
as likely to read a virtual open circuit across the socket,
irrespective of which way round you have your meter. This, in itself,
will not help you to determine the polarity.
Yes... so if you have a high/low resistance readings, this is very useful.
If you have a low/low or a high/high, checking the resistance isn't useful
under these conditions.
But worse. If the unit
employs a shunt protection diode, when your meter is connected
//backwards// to the correct polarity, you will get a reading of 700
ohms or so, but when it is connected the //correct// way round, you
may well read infinity or near. By your definition of how your system
works to determine polarity, that would give you a clear indication
of which was the correct polarity, but would actually yield the
*INcorrect* polarity.
True, but it will not hurt anything. As the shunt will only allow about a
negative 0.7v to the rest of the unit. So unless the shunt blows, it
shouldn't ever hurt it. You can possibly damage the adapter, but it should
be protected against such anyway. But it is cheaper than the unit it is
powering anyway. ;-)
As for external metalwork not being connected to the internal common
ground, these days, that is rare. I do come across the situation
sometimes, on AV amps, where the RCA socket sleeves are floating with
respect to the chassis, but it is the exception rather than the rule,
and is done to help alleviate potential ground loop issues when
connecting to other equipment.
As far as the polarity of the ground goes, I am prepared to say that
in my considerable experience, on modern equipment designed for the
consumer market, it is always negative. And that really is about as
cast in stone as anything in electronic design ever is. I could of
course be wrong on this, but if anyone wants to correct me with
specific examples - remembering "modern" and "domestic", I'm
listening, and willing to modify my position on it.
Yes I agree. But thinking that way, you may fry every positive ground you
come across. If that is an acceptable risk for you and others... well what
can I say? While some risks are acceptable to me, frying something because
you had the polarity wrong just isn't one of them. LOL
I say again, that the OP asked a simple question, to which there was a
simple answer. I don't really believe that there was any need to
muddy the waters to the extent of all of this silly stuff that has
been put forward, but hey - ho. I guess it all makes for an
interesting life ... d;~}
I believe having enough information to make a wise choice is far better
than bozos telling you that you have to do it this way and there are no
becauses. 
--
Bill
Well, I guess we could go on dancing around this one for ever ...
With the circuit loads being connected to the back end of regulators that
are likely to have high input resistances with respect to the low test
voltage supplied by a multimeter, you really are unlikely to read anything
meaningful across the DC input socket. Trust me. I do this (very
successfully) for a living. Reading high-low, low-low, high-high, is
fundamentally useless to determining input polarity, unless you have a
schematic for the equipment to know what you are looking at. If you had a
schematic, you would not, of course, be trying to determine the polarity in
the first place ...
It is also not necessarily true that you will damage nothing if you do
arrive at a wrong conclusion as a result of applying your ohm-meter test to
an equipment which employs a shunt protection diode. Remember that the power
supplies for LCD TV sets and monitors, when these are external types, are
capable of supplying typically 2 to 4 amps. This is plenty enough to destroy
a typical 1 amp shunt diode, or to blow a pico or surface mount fuse, as is
typically found in such devices, or even to take out print, which is
sometimes deliberately 'necked' to provide a fuse function. Shunt protection
diodes seldom survive a reverse connection. Ask anyone who repairs CB
radios, or PMR radios, or plain old car entertainment radios.
Even if no shunt diode is used, there is still no guarantee that any
regulator device which has reverse polarity applied to it, will survive. I
have seen plenty that haven't.
I honestly don't believe that I am going to fry *any* positive grounds that
I come across, for the simple reason that on modern equipment, I just don't
come across them. They died out pretty much with germanium PNP transistors.
Obviously, if I was trying to determine the polarity of a piece of 30 year
old kit, I would take the trouble to employ different methods to do so, on
the off-chance that it might have a positive ground, but again, trust me,
positive grounds simply *aren't* encountered on modern equipment.
I hope when you refer to "bozos", you are not including me in that, as it
would cause me to take extreme offence at you. Having enough information to
make a wise choice is indeed a laudable objective, but discounting the
advice of someone who has more than 35 years declared experience in a field,
borders on stupidity. I certainly would not tell anyone that they *must* do
it this way, but if I believe, based on my considerable experience in the
repair field, that a particular method is likely to yield a correct answer
with a better than 95% certainty, then I am going to advise them of this,
which I believe is the way I approached the OP's original question, in the
first place.
Whilst there are always "becauses" as you put it, in this particular case,
their validity is negligible, for all of the reasons that I have
(painstakingly) explained over and over.
Arfa