This is the item in which we trade tips for finding parts and such.

33 responses total.

A few quick browses didn't find any items with source, or whatever, at least
for what I'm trying to find.

I'm trying to find a public address 70 volt speaker line transformer.  I'm
building a PA amp to connect to speakers in a building, and I need a 50+ watt
transformer that goes from 4 or 8 ohms to a 70.7 volt line (all the speakers
then have little transformers to go the other way, standard sort of setup).

Any ideas or sources?  Even Purchase Radio didn't know, although they could
find a 35W transformer attached to a wall plate and volume knob.  Maybe a
couple of those linked together...

Did you try the yellow pages?

Why not use one of those x-formers backwards?

Wanted: 8-36 (NF) screws (just 4).

re: #3:  60watts > 10 watts

To be less cryptic, I'm looking for a transformer with a high power capacity.
No problem running one "backwards" if I could find one in the 50+ watt range.

The Yellow Pages don't have much there in the sections I was checking out.
Doesn't seem to be an "audio transformers" section either.

These are a common PA system item.  I think Radio Shack used to sell them.

I thought you meant you *had* a 50 watt, 70-volt line to 4-8 ohms. But
I think I don't understand what a 70.7 *volt* to *4 to 8 ohm* transformer
is. They are usually rates in ohms to ohms, and power. Please describe
the x-former you seek more exactly, and I may find one in a catalog.

Well, the trouble is that "everybody" knows what "70.7 volt line" is, so
nobody bothers to describe it.  It basically takes the usual 8 ohm speaker
input/output and converts it to one with higher voltage and lower current in
order to reduce line loss in building wiring.  It requires a transformer at
each speaker, and usually at the amp (although I now suspect that amp
designers these days just found a way to drive the line directly rather than
pay for a heavy transformer).  Maybe 70.7 volts for 1 watt at 8 ohms?

The difficulty is that with reduced use at the amp end, the higher power
transformers are getting hard to find.  I was able to find a 100 watt
transformer 15 years ago when I did something similar at Pioneer High School,
but so far I'm only getting blank looks.  10 watt transformers are easy to
get, but the big one for the amp is another story altogether so far.

My mark of distinction is that I am not "everyone"  8^}. Actually, it is "in
group" jargon. 

Last week I visited a site where they did the same trick with transformers.
They generated power at 56,000 volts and kicked it up to 500,000 volts for
distribution (the Raccoon Mt. Pumped Storge Facility, outside Chattanooga).

I found an "audio line matching x-former" in Mouser for "4 to 8 ohm to 70 volt
line". It is tapped for ".625, 1.25, 2.5, 5 and 10W". Is W watts? Since
all a tranformer can have are turn ratios, winding and mutual inductances,
and winding resistances, I'd would like to know what these are for this "4-8
ohm to 70 volt" thing.

Right, my Mouser catalog has the same part (imagine that!).  W == watts.  I'm
curious about what "70 volt line" really means, too.  One interesting thing
is that the speaker side transformers often have multiple taps on the line
side, to allow different amounts of power to be tapped off (i.e. the speaker
in the warehouse gets 10 watts, in the Inner Sanctum the speaker runs off a
measly 1 watt).

Maybe it's 70.7 volts when driven at 1 dBm, or some such?  I'd guess 70.7 is
the voltage at some standard reference level, especially given the
suspicious number.  (One half the square root of 2 is 0.707, and as I recall
that's the scale factor for converting from peak to RMS voltage for a sine
wave.)

You could run some AC into the input of one of the 10w speaker-side
transformers and see what voltage you get on the output; this will tell
you the turns ratio.  You could either be brave and run 110vac direct
into the input (if it's designed for 70vac audio, it probably has some
margin of safety), or you could run the transformer in series with a
resistor -- a 100w incadescent lightbulb would make a fine resistance,
and measure the voltage on both sides of the transformer.

Once you know the turns ratio, you can experiment with experiment with
ordinary 60hz "power" transformers, or you could wind your own
transformer.  Do audio transformers use the same magnetic material as
power transformers (silicon iron?)

They don't worry about operation at 10 KHz for power x-formers. The
core must be very low loss at high frequencies, which requires very
thin lamina (orignally) or special core materials. 

If this is a PA, hi frequency fidelity may not be important.  But if it
is, then you probably want to get something better.  Other things to try
would be a ham swap meet, or an electronics distributor.

good idea to try measuring with AC.  But I'd guess that even if I don't care
about 10kHz operation, I'd be pretty unhappy without 5kHz.  

Try it and see.  I believe telephones cut out around 2kHz.  Music is
more likely to have obvious problems than speech.

No, phones cut off above 5-6kHz.

And even with a 5-6 kHz cutoff, speech intelligibility is adversely
affected.  The reason you have such a hard time spelling things to people
over the phone is that some of the frequencies in the sibilant consonents
get cut off, making them sound more alike.

At any rate, even if frequency response weren't a problem, I doubt a power
transformer would have the proper impedance to match the amp output.

No, trying a power transformer was a good idea, and I may still try it.

actually phones clip anything not ni the 300-3400Hz range, but the highest
a human voice can go is around 5KHz, though that is only when singing. normal
voice is well within that range. 

Yes, though the 'white noise' in sibilent and explosive consonents goes well
above that.  I'm curious where you got the 3,400 Hz figure; I've never seen
it, though it sounds about right for communications radio.  (I think SSB has
a bandwidth of about 3 to 3.5 kHz.)

Bit of a breakthru:  Out of the 20+ electronics books I checked in the local
library, only 2 had any significant mention of these transformers.  One of
the books promised more detail in volume 2 (which of course was not
available), while a crusty old speaker design book mentioned the formula for
picking the speaker's transformer:  70^2 / (desired watts).  So a 10 watt
speaker tap would have a primary of 590 ohms, and a secondary of 8 ohms.  So,
if I want a 60 watt transformer for the amp, and the amp can drive an 8 ohm
load, then I need a transformer that has about 10 to 1 turns ratio.  Hmmm,
I wonder how a 120vac to 12vac transformer would do here?

This must have been more of a hassle in the old tube days, when you would have
to add up all your speakers by power tap and make the amp match.  Solid state
amps only need to have a maximum load, while tubes also needed a minimum load.

(oops, change that "590" to "490")

..and the new MCM catalog finally had something useful.  The trend now is to
wire your fancy new castle with many speakers powered by conventional stereo
equipment, so load balancing is an issue.  MCM has a variety of "speaker load
matching" transformers to hook up a number of 8 ohm speakers to a 8 ohm amp
without overloading it.  One promising unit I'm going to purchase has taps
for 8x, 16x, etc. impedance.  A little math shows that these values work out
closely to the kind of 100watt to 70.7 line matching I need.  Since this is
a transformer specifically designed for use with speakers I'll be getting good
frequency response.

re 22: i got that number from my MCSE course, though it might be a bit off.
"white noise" is generally caused from three diffrent sources:
1. imperfect twisting of wires (though it has to be fairly bad to have an
effect, and then you are in deep trouble if you try using the line for data
transmissins as you will be effectivly destroying signals when both recieving
and transmiting at the same time)
2. unshielded wires near EM emmitters (usually tri-phase lines and major
appliances such as refrigirators and dishwashers)
3. momentary disruptions such as power surges, lightening, brown-outs, etc.

Lots of other things cause "white noise".  Tape hiss, for instance...

I'd expect refrigerators & such to generate 60Hz hum, more.

Depends on your definition of white noise.  Typically white noise is defined
to be noise that has equal amplitude at all frequencies.  60 or 120 Hz hum
induced from a nearby power line would not be "white".

what is the definition of "pink noise," scott?  i can tell the difference by
ear (i think) but i don't know the frequency composition...

Pink noise has some of the high frequencies reduced a bit.  It has to do with
how ears work, or somthing like that.  "Red noise" has even less high end.

White noise has no amplitude at any frequency. It has, instead, a power,
or spectral, density. In order to measure an amplitude, one must measure
over a range of frequencies. Then the power spectrum is the
mean-square-value averaged over an interval of frequency one hertz wide.
For white noise, this is a constant. No real noise is "white" as all
noise has lower and/or upper cutoffs, but some noise can be considered
"white" over a range of frequencies. The power density spectrum can
therefore be normalized to the value 1.0. Brownian motion, which does
exist at all frequencies, has the power density spectrum 1/f^2. A
common form of noise is 1/f noise. It is observed (approximately) in
many real situations, but there is no fundamental cause for it that has
been proposed. Music, incidentally, averaged over long periods is very
close to 1/f noise. 

One can make analogies to the color spectrum, so "red" noise cuts off
high frequencies, and "blue" noise, cuts off low frequencies. I'm not
sure what "pink" noise would be.

I've always heard pink noise defined as something "like" white noise
except with a distinct frequency peak in it.  60Hz hum would probably
not qualify.  White noise filtered through a not particularly good
filter (for instance, road noise or wind heard through a pipe) should
qualify.

for my post above, s/white noise/line noise. also #1 is called crosstalk and
can be a big problam if the wires aren't twisted.

Response not possible - You must register and login before posting.

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss