Any experience with negative impedance?

The requirement is to design a specialist one-off audio amplifier to
drive a loudspeaker at the end of a long cable for experimental
purposes. The voice coil is nominally 15 ohms with a pure resistance of
10 ohms. The loop resistance of the cable will be somewhere between 2
and 4 ohms. It would be desirable to have a damping factor of 10 or
better, so I need to reduce the effect of the cable resistance in some
way.

I have considered the use of line transformers or a four-wire feedback
system, but neither of them is practical in this particular case. As
this amplifier will form part of a fixed installation and will only ever
be used to drive the one loudspeaker, I am now looking into the
possibility of giving the amplifier 3 ohms of negative output impedance.

The simple solution is to use an off-the-shelf unit for the power
amplifier, the TDA7295 should be more than adequate. With a 0.5 ohm
resistor in the return loudspeaker wire, I can measure the output
current and derive a suitable voltage for adding to the input signal.

The practical situation generates at least two problems which the simple
theory does not take into account:

1) The capacitance between the conductors of the long loudspeaker
circuit will give the effect of a falling loop impedance as the
frequency increases, leading to ultrasonic instability.

2) The long unscreened run of loudspeaker wire, which might be in close
proximity to mains cables and dimmable lighting circuits, may pick up
interference by capacitive coupling, which will then be injected into
the feedback point and amplified.

Both of these might be overcome by a capacitor across the 0.5 ohm
resistor, the value being chose to only take effect above the highest
wanted audio frequency.

The problem is that I have to construct this equipment without access to
the loudspeaker or wiring, and it then has to be certain of working
correctly the first time it is installed. For this reason I would be
very grateful if anyone has had any practical experience of anything
similar and could let me know of the snags they have encountered and the
solutions which worked.

 

Google werner steiger "negative output impedance"

Back in the '60's a popular term was "motional feedback"
I built a speaker system for my truck using negative output impedance
to damp the speaker/cab resonance and boost the lows. I can't find
any documentation on it other than the paper referenced above.
I never had oscillation problems, but
this was ~40 years ago when transistors weren't very fast.

 

2) is not an option: This is an historic loudspeaker and there will
have to be several extra safety circuits to integrate the amount of
power delivered in various frequency bands and trip out the amplifier
immediately if there is any danger of causing damage. We are being very
careful indeed - any mistake will be the last.

Not high impedance; the exact opposite.

No audience, it will be the subject of tests by a group of audio
historians, museum curators and researchers.

 

This loudspeaker is much older than that, it is one of the first ever
moving coil types.

Those old Philips sets gave much better quality than many of the British
ones of the same age and price range. I used to listen to symphony
concerts on one in the 1950s and to Radio Netherland's shortwave
programmes in English during the 1960s. I still have one and, the last
time I used it, the quality was far better than any modern
self-contained A.M. radio.

The servicemen used to hate them because the circuits were too 'clever'
and managed to squeeze 5-valve performance out of only 4 valves. Nobody
seemed to know how to mend them properly.

Phiips was one of the first companies to produce magnets strong enough
and cheaply enough to make permanent-magnet loudspeakers available in
domestic sets. That initially made a separate smoothing choke necessary
(previously the loudspeaker field coil had served as a choke) but then
Philips started tapping the HT into the primary of the output
transformer and drawing off a current to feed the RF stages which
balanced the output valve current. This reduced the hum and removed the
DC magnetisation from the output transformer core.

Philips were a very innovative company right from the late 1920s
onwards. They were technically years ahead of most of the other
manufacturers, sometime to their cost.

We have a bit of a problem with that, we know the output stages of the
'modern' amplifier were push-pull UV845s with 1Kv on the anodes, capable
of giving about 40 watts; but there was a response-correction network
between the output transformer and the loudspeaker which threw most of
it away. The earlier amplifier was far worse.

We are talking about very early experimental technology here - and my
job is to allow it to be tested without damage.

 

http://www.poppyrecords.co.uk/other/images/Philips.gif

The ratio of the two halves of the output transformer primary was chosen
so that the current in one half produced equal core magnetisation to the
current in the other half. The value of R was chosen so that the ripple
it passed was the same as the ripple through the output pentode, so
there was effectively a push-pull action at audio frequency. There was
inevitably some audio power lost in R.

The transformer inductance and the resistor also assisted with smoothing
the H.T. supply to the earlier stages. The H.T. smoothing capacitors
were around 16 microfarads, so no A.F. found its way into the early
stages.

You can imagine why the traditional English serviceman was baffled when
he met a fault on this sort of circuit.

 

There are many protection circuits to make sure that any excessive
power, no matter what the cause, will shut down the whole system.

We have enough information to know that reproducing the original circuit
will only complicate the problem of measuring the performance of the
loudspeaker. The idea is to drive it from a well-controlled modern
source and make the necessary measurements, not to try to replicate the
original operating conditions (however much fun that would be - with
mercury arc rectifiers and a motor-generator set in the power supply).

This pre-dates Black's patent, there was no external feedback around the
valves (only the unavoidable internal feedback due to the anode-cathode
potential gradient of a triode).

 

We are trying to make objective measurements in addition to subjective
listening tests, so it is worth going to some trouble to remove the
effects of any external equipment from the equation.

 

Adrian Tuddenham a écrit :

Well, he should have had a cap of T...

 

Two reasans:

1) There is nothing to equalise.

2) A 'graphic equaliser' doesn't equalise anything, it is a sound
effects unit which has no place in objective audio measurement.

I am trying to measure how a loudspeaker responds to a known voltage
across its voice coil, unfortunately this one cannot be taken anywhere
for testing and the only practical way of making a connection to it is
by means of a very long piece of wire. I want to remove the effect of
the wire on the source impedance.

 

Thanks, that is an excellent idea because, even if I don't make it
oscillate accidentally, someone else might feed in signals at an
excessive level and cause damage before the protection circuits can
operate. Two back-to-back yellow LEDs work a treat across a 600-ohm
circuit at about 0dBm.

 

He'd probably have a cuppa and a fag, then nip down the frog to the
bookies and put a couple of bob on the gee-gees. After that he might
nip back and see if his oppo has sorted it while he was AWOL.

 

The important thing is whether the impedance pf the external circuit
drops below the negative impedance of the amplifier at any frequency. I
agree with your assessment that the loudspeaker impedance will always be
more than the rated resistance, so I should have nothing to worry about
from that cause.

I have heard of acoustic reflections and resonances giving lower than
normal impedances at some frequencies and the possibility of stored
energy being dumped back into the amplifier. For this reason I intend
to make the feedback adjustable, so that it is only sufficient to
compensate for the circuit resistance and not for the resistance of the
voice coil.

There are practical difficulties with getting the wires to where the
loudspeaker is located. Ordinary house-wiring cable will have to be
used for both the voice coil and for the field coil supplies because it
may share a trunking with other mains wires. If unsreened and untwisted
wire is use for the sensing circuit, it could pick up a lot of
interference and I have to supply this system so that it can be
installed and made to work immediately without any further alterations.

[...]

That is a possibility. I must try to ensure that there is only one
dominant pole in the response.

 

and then you want to add the uncertainty that you negative impedance
might
have a effect?

why complicate things when a thicker wire and or moving the amp closer
will do
it ?


-Lasse

 

After reading the other responses and your added details ("legacy"
speaker, wiring constraints, etc.) I'm wondering why you don't install
a suitable wireless amplifier at the speaker and feed it power with
the available wiring.


Mark L. Fergerson

 

The loudspeaker circuit contains three major sources of resistance, the
resistance of the voice coil, the apparent output resistance of the
amplifier and the actual resistance of the connecting wires. For the
purposes of this application, I want the resistance of the voice coil to
be the controlling factor, so I would like the other two to be
negligible by comparison.

The apparent output resistance of the amplifier is very low, so low as
to be negligible, because it has a large amount of local negative
feedback. The wiring between the amplifier and loudspeaker has to be
long (for reasons outside my control) and has a resistance which is not
negligible compared with the drive coil resistance.

By using the amplifier in a feedback circuit which makes it appear to
have a controlled degree of negative output impedance which is equal to
the unwanted resistance of the wire, I can reduce the loop resistance of
the whole output circuit loop to just that of the loudspeaker voice
coil.

I am not trying to counteract the loudspeaker resistance, just that of
the unavoidably long wiring.

 

I don't want to, but it is the least difficult of all the methods I have
considered so far.

The loudspeaker is hanging from chains in the roof of a very tall
building and the only possible wiring run to the control position is
hundreds of feet long and must be insulated to mains standards.

 

Perhaps I should have mentioned earlier that this loudspeaker is hanging
from chains in the roof of a tall building.

 

I am hoping that the inductance of a long run of 1.5mm T&E will not be
too high, but it is something to look out for. Perhaps it will need a
small inductor in series with the current-sensing resistor to balance it
(perhaps the current-sensing resistor should be made of copper, so as to
counteract ambient temperature effects).

By earthing the central 'earth' conductor, I ought to be able to reduce
the capacitive coupling between cables by a significant amount.

The idea goes back to the 1890s, when tramway generating stations used
negative impedance boosters to counteract the I*R drop on long DC lines
to remote parts of the system. It wouldn't surprise me to find that
steam and hydraulic engineers were using the same principle in their
domains fifty years before that.

 

* Speakers NEVER have "pure resistance",except at one or a few points in
the audio spectrum: Voice coil versions are inductive near the low
end,tend to become resonant in the middle, with a widely varying
_impedance_ thereafter.
* I suggest you read up on "damping"...

* Use of either line transformers or a four-wire feedback system is
strongly recommended.

* Twisted wire can do wonders..

* No access? "Must work" first time? Forget it.

 

I didn't want to complicate my original question with lots of detail,
but the reason for the long wires is because the loudspeaker is hanging
from chains in the roof of a very tall building. As far as I know there
are no commercially-available wireless amplifiers which will offer
sufficient protection to the loudspeaker in the event of a fault or
mis-use.

I also have to supply the field coil with power, preferably from the
same unit as the amplifier, but that is another story.