Adrian C wrote:

I'm not having a go at him being autistic - heavens no, sorry if it
looked that way. But in the context knowlegabilty of the subject of
electronics, some of the symptoms of that may actually be a benefit
once his "I/O noise" misunderstandings have subsided.


I'm not reading his posts anymore. But near the start of the thread he did
make some intelligent points, which I responded to. His big problem with me
seems to be that I don't respond to the...er...other stuff he seems
compelled to include.

And when they do, meds or whatever, Phils participation is welcome to
me - he is obviously a smart chap ...


Agreed. I'm not a regular in here but I sometimes lurk. Some of his stuff is
good.

And here's worthwhile campaign currently broadcasting around the UK.
http://www.time-to-change.org.uk


Yes, mental health is a serious subject. But it is human nature to respond
to inappropriate behavior by laughing.

"Adrian C"


Using the word "autistic" incorrectly is a technical error of Phil's,
which I'm merely pointing out.


** I use the word perfectly correctly - all you are " pointing out " is
your own massive ignorance of the topic.

Autistic people are all around you, I bet you are one too.

Autism is more precisely called ASD for Autism Spectrum Disorder - it is
not metal illness and cannot be treated.

It is a metal disability or genetic syndrome.




...... Phil

TonyL wrote:

Here's a problem that is causing me some grief. I'm rusty
with electronics these days, perhaps the experts in here
can give me some pointers ?

I want to combine the audio outputs of an unspecified
number of op-amps, say 2-20. A simple mixer would do the
trick...except I need the resulting summed audio level to
be constant as outputs are connected to and disconnected
from the system.

I know a virtual ground is the standard method to mix
audio but there's another wrinkle here. I also need to
send the summed audio signal to an unspecified number of
outputs, between 2 and 20. In other words, the common
point where all of the audio signals are present needs to
be level insensitive to changes in numbers of audio inputs
*and* outputs.

Is this possible ? I've tried some breadboarding and find
that each added input results in a level drop at the "star
point". Also, adding a second virtual ground output
completely kills the audio. I can, of course, add as many
outputs as I wish using the output op-amps as high
impedance buffers, which leaves me with the input summing
problem.

Note...the electrical "star point" is not a physical
point. In practice the inputs/outputs are distributed
along a common cable, of unspecified length, which can
change.

Comments appreciated, other than "design yer own f******
stuff", of course :-) Just pointers to info would help.

Just passing, thread is long so someone may already have
said this.

You have a fundamental problem in your thinking. Your mixer
bus sums current, and has no voltage signal on it. This is
useful because currents sum in parallel, rather than
voltages, that sum in series. If you want to take a number
of outputs from the bus, then they must also be arranged in
series. You are trying to use a number of outputs in
parallel. Clearly they will share current, so the level of
each will drop. In the extreme, parallel virtual earths will
short each other out completely. Er...does this make sense?

So you must add outputs in series. That is, rather than
connecting each output between bus and ground, you must
daisy-chain them. The current through each will then be
equal, and they won't short each other out.

Will this work if you try chaining virtual earths?

In principle, you could create a voltage bus, by summing its
inputs in parallel. Then you could add high-impedance
outputs in parallel. I assume this arrangement would be
problematic, which is why mixers use virtual earths.

However, breaking into the bus each time a new output is
required also has its problems. I guess you could have all
outputs constantly active...as many as you will ever
need...and select from them further down the line.

Or use a separate output bus

Ian

I wrote:

...In principle, you could create a voltage bus, by
summing its inputs in parallel....

Er...series, sorry.

Ian

"Ian Iveson Notorious ****wit "

You have a fundamental problem in your thinking.


** For sure.

But this utter imbecile has no idea what the error is.


Your mixer bus sums current, and has no voltage signal on it.



** So to get a usable signal voltage, use a low Z buss feed by high value
resistors.

Summing still occurs without interaction and there is a source for driving
outputs.


( snip whole pile of putrid, spew inducing drivel )




..... Phil

Ian Iveson wrote:

Comments appreciated, other than "design yer own f******
stuff", of course :-) Just pointers to info would help.

Just passing, thread is long so someone may already have
said this.


Ian, the good folk in here have helped me out but I'll respond anyway. I've
snipped a few items for brevity.

I'm now using a "collector" bus with a virtual ground mixer feeding a
"distribution bus". The project, BTW, is for a group intercom whereby
multiple users can connect and disconnect at any time. Number of users and
total connecting cable length is to be variable.

You have a fundamental problem in your thinking. Your mixer
bus sums current, and has no voltage signal on it.

Yes, as was pointed out to me in this thread.

each will drop. In the extreme, parallel virtual earths will
short each other out completely. Er...does this make sense?


Yes, that is exactly what happened when I tried to use more than one virtual
ground mixer in parallel. Seems obvious now ;-)


Will this work if you try chaining virtual earths?


The 2-bus method avoids the issue entirely.


However, breaking into the bus each time a new output is
required also has its problems. I guess you could have all
outputs constantly active...as many as you will ever
need...and select from them further down the line.


The problem there is that the system is to be extensible, numbers of
inputs/outputs can be changed at any time and the audio bus length is also
liable to be changed as well...up to a limit. I will supply the bus cable in
sections. Users will add "intercoms" where the cable sections connect to
each other.

Or use a separate output bus


Yes !

"Tony LIAR "



Ian, the good folk in here have helped me out but I'll respond anyway.
I've
snipped a few items for brevity.


** ****heads like you SNIP everything that does not suit them.



I'm now using a "collector" bus with a virtual ground mixer feeding a
"distribution bus".


** Huh ????????????

Got any old " Double Decker " busses in there too ?????

you ****ing **** HEAD !!


The project, BTW, is for a group intercom whereby
multiple users can connect and disconnect at any time. Number of users and
total connecting cable length is to be variable.


** ****ing 100% ABSURD that was not in your VERY FIRST post -

you ****ing **** HEAD !!



The 2-bus method avoids the issue entirely.


** Shame how YOU have missed the buss entirely -

you ****ing **** HEAD !!

----------------------------------------


Get off this damn NG NOW

or I will ****ing kick you off.

You LYING ARSEHOLE.




....... Phil

In article ,
Phil Allison wrote:
Get off this damn NG NOW

or I will ****ing kick you off.

Crikey. He thinks he owns it now...

--
*Change is inevitable, except from a vending machine*

Dave Plowman London SW
To e-mail, change noise into sound.

"Dave Plowman (News)" wrote in message
...
In article ,
Phil Allison wrote:
Get off this damn NG NOW

or I will ****ing kick you off.

Crikey. He thinks he owns it now...

Should we get Admiral Sir John Jellicoe to
send a gun-boat?

Dave Plowman (News) wrote:
In article ,
Phil Allison wrote:
Get off this damn NG NOW

or I will ****ing kick you off.

Crikey. He thinks he owns it now...


I don't know the history of Phil too well. My initial reaction to his
current outburst was amusement but it seems that he really means what he
posts...which is rather sad.