In ,
Arfa Daily typed on Sun, 30 Mar 2008 16:24:25 GMT:
"BillW50" wrote in message
. com...
In ,
Arfa Daily typed on Sun, 30 Mar 2008 14:23:08 GMT:
"BillW50" wrote in message
. com...
In ,
Arfa Daily typed on Fri, 28 Mar 2008 16:07:57 GMT:
"BillW50" wrote in message
. com...
In ,
Arfa Daily typed on Fri, 28 Mar 2008 12:52:58 GMT:
[...]
As for a transistor radio that uses PNP transistors, I haven't
seen one that uses transistors at all for many years, let alone
PNP ones, so I think you might be struggling to fit that into my
"modern" category, also.

Well Arfa... they still use transistors (both NPN and PNP types)
in modern equipment. The reason you don't see them anymore is do
to the magic of minturization. But they are still there, just
neatly packaged into what is known today as the intergrated
circuit (IC chip).

And I though I was pedantic !! Yes, of course ICs contain
transistors, and yes, I would accept that some of them may be PNP
types, depending on block function within the IC, but I don't
think, with the best will in the world, that this is the level of
transistor existence that William was referring to with his
"transistor radio using PNP transistors" scenario. In any event,
in the case of an IC taking a single polarity rail, it is
irrelevant whether the transistors inside are NPN or PNP or FETs
or whatever. PNP transistors are just used 'upside down', as are
discrete PNP transistors when used in any piece of single
polarity rail equipment. The ground is still (typically for
//modern// equipment) the "-" side of the power supply / battery.
Anyway, this is getting out of hand. The OP aked a simple
question, and I gave a simple answer. This afternoon, I was in a
friend's computer repair shop. Bear in mind that he deals with
monitors of all types and makes on a daily basis. I asked him how
he would go about determining the polarity of such a monitor, and
he said that he would stick one side of his ohm meter on one of
the D connector locking screws, and the other on each pin of the
DC connector. When he found the pin that read short to the
connector locking screw, it was his contention that he would have
identified the "-" side of the power supply. So that's pretty
much exactly what I said. He also frowned and shook his head, and
said that he couldn't remember how many years it had been since
he had seen a DC connector that had the pin as the "-".
Which is also pretty much what I said ...

Arfa

That would be okay if it were a negative ground system. But like
what William Sommerwerck mentioned, we don't know that. The way I
would do it is to ohm the the power in. And the lower resistance
would be the correct polarity. Although you would need another
meter to read the polarity of the ohm meter. As they are not
standardized on multimeters. You could also use a diode (or LED)
to learn of the polarity of the meter as well. --
Bill

JHC !!! Do you not understand the word "modern" ? Do you not
understand the phrase "... deals with monitors of all types on a
daily basis" ? I repair this stuff all day every day for a living. I
have done for over 35 years. I cannot remember the last time I saw a
piece of kit of any description, which employed a positive ground.
My friend, who owns a computer repair shop, and has done for many
years, cannot remember the last time - if ever - that he saw a
monitor with an external power supply, that was not negative ground
with the connector sleeve as the negative connection.
With so much interconnectivity between household items now, there
has had to be a degree of standardisation on this issue, and it has
evolved through a general concensus amongst manufacturers, that
negative ground will be the convention.

As for your method of determining polarity, it makes no sense at
all, unless you are assuming a series diode, which is quite
unlikely in most modern kit, as it represents a waste of power due
to its forward voltage drop. It may even have a shunt protection
diode, in which case, your 'test' will ensure that the polarity is
determined INcorrectly. Even if the device did have a series diode,
depending on where the supply first goes, there is still no
guarantee that there will be any reading at all on a standard
multimeter on ohms. If there is not any diode - series or shunt -
any reading of ohms obtained across the input socket, is unlikely
to reveal anything meaningful. What is your experience in
fault-finding, I wonder, to have caused you to have formulated such
a bizarre method, and believe that it would uncategorically give
you a correct result ? Arfa

Actually being an electrical engineer for 35 years, I could careless
how long your friend has been repairing computers. And the reason
why the ohm meter works is because all of the curcuits are in
parallel with the supply. Thus you will get a lower reading when the
polarity is correct. And you will get a higher reading when it is
not correct. Thus as all of the circuits are reversed biased.

That is unmitigated nonsense. If there is a shunt protection diode,
it will be FORWARD biased when the polarity is WRONG. Also, the fact
that my friend repairs this stuff all day, and as an electrical
engineer, you clearly do not, that makes him an expert, compared to
you ...

If there was a shunt protection diode, then both resistance measurements
would be forward biased. Thus you would know that since there was no
high resistance reading. But let's assume and use your plan for a
minute. There are plenty of examples where they don't ground the shield
but just let it float. Yes it sounds stupid I know, but it has been done
from time to time. And I worry about everything seemingly coming from
China nowadays. Which IMHO is only going to make things worse. And some
of this stuff from China, isn't even UL or FCC approved.

Whether or not all manufactures use negative ground or not, I have no
idea.

Well, as I repair this stuff all day as well, I *do* have an idea, so
that clearly also makes me more of an expert on this particular
subject, than you ...

That is indeed possible. I always said and believed that we can learn a
lot even from a child.

Although in all of my experience, I have learned to never assume
anything. And I have seen many strange designs. One of them had an
OP amp's output connected directly to ground. I was confused about
that one until I chatted with the designer. Then it all made sense.
--
Bill

In general, I would agree with you not to assume anything, but some
things are a matter of convention, and in recent years, based on my
direct experience of such things, I would stick my neck on the line,
and say that this is one, and that all modern kit, manufactured for
the domestic market, employs circuitry with a negative ground, to
which (most) external metalwork is firmly bonded.

Arfa

I also have seen grounds that wasn't really ground either. Shields that
were not connected to anything. Lots of weird stuff goes on in consumer
grade equipment. Most of it IMHO is done to save a buck. Some of it is
just ingenious! And some of it is just sloppy engineering. And sometimes
it was done as a last minute bandaid just to pass FCC radio emissions.


--
Bill