Nominal Summing Buss Level

[rlaury]

Hi Group

I'm looking for information on a specification published by the Motion Picture Ind (SMPTE) on summing buss levels. This extensive study was done in the 1940's and supposedly concluded that the single channel loss through a summing network should be -32dB. Has anyone read this before? Basically, I'm trying to determine what the loss should be based on the number of sources summed. It will be an active balanced summing network, virtual earth type system. Any thoughts?

RonL

 

[SSLtech]

Ron,

Virtual earth is a special case, -no?

With a virtual earth summing amplifier, the gain changes as each source is added. (the successive paralelling of input sources reduces the source impedance, but the source impedance is simultaneously the attenuating leg of both each individual source and the single feedback leg.)

As a result, so long as the noninverting input of the amplifier is tied to ground potential, the inverting input will always be returned to ground potential.

Current flows "into and out of" the virtual ground point at the inverting input, but there's no measurable voltage present at that point.

The overall gain of the amplification stage increases, but the stage as a whole has fixed gain determined by the inverse ratio of source resistance to feedback resistance.

For unity overall gain, a Virtual Earth stage will have a gain equal to the number of inputs. If you increase the gain if the stage to recover additional losses incurred in things like passive pan circuits before the unity gain stage proper, the recovery gain will have to be added.

SMPTE standards from that far back: would they apply to virtual earth or just passive loss recovery summing?

While I can't be emphatic, I think that they might not apply to virtual earth. If anyone can correct me, I'd be happy to learn, since this is a long-held assumption on my part.

Keith

 

[NewYorkDave]

Ron,

In my extensive research of mixer technology of the tube era, over the last few years, I've never come across any reference to such a standard. I'm not trying to imply that it never existed, but it certainly wasn't cited in any documents I've seen.

SMPTE (then known as SMPE) published a paper by Miller and Kimball in 1944 titled A Rerecording Console, Associated Circuits, and Constant-B Equalizers. Here's an excerpt:

The basic console circuit has a transmission loss of about 23 dB. This is divided among the circuit parts as follows: The volume controls used are of the slide-wire type having a minimum loss of about 6 dB. Each of the two-position mixer transformers has a theoretical loss of 3dB, or an actual loss of about 4 dB. This makes 8 dB for 2 such coils in any circuit. The six-position mixer coil has a theoretical loss of 7.8 dB, or an actual loss close to 9 dB.

Granted, they're describing a mixer using hybrid coils, not the more conventional resistive mixing. Just about every mixer of the conventional constant-impedance type (circa '50s, early '60s) I've seen has a loss that is directly proportional to the number of channels. In other words, it varied from mixer to mixer. It was not unusual for the signal to be down nearly to mic level before entering the post-mixing booster amp.

Virtual earth (active) mixing was virtually unknown in audio in the 1940s. It was very uncommon until the '60s.

 

[rlaury]

Thanks guys.
I had this all figured out a few years back and I'm trying to get my head around it again. I guess what I'm after is, Consider 32 faders. But only one has a signal, for example a vocal panned center. I want to be able to push that fader up close to the top and still make +4 at the 2 mix out with the master fader setting close to "max". So, I guess I just need to work it backwards and come up with this figure. I'm thinking with the master fader setting wide open,, the input to the summing amps to 2 mix out should be unity gain. This would minimize noise gain. Does this make sense?

Thanks
RonL

 

[TedF]

I seem to remember back in the late 1950s reading about a recommended mix bus level of -32dBv; of course this ceased to be relevant when 'virtual earth' mixing was developed.

The first example I saw of this was a reference in a Mullard data book showing an ECC83 connected as a 'virtual earth' amplifier. (I can't imagine that it was that successful!)

In 1973 I developed a 'balanced virtual earth' mixing system that we used at 'Alice in the UK. It had the advantages of being free of all the dreadful ground problems that were endemic in simple 'VE' mixing in large frames, and it gave an additional 6dB of headroom. This arrangement was subsequently taken up by other manufacturers and is now the classic balanced mix design.
If you would like circuit info, pls email and I will let you have a current version (joke intended!) :grin:

 

[bcarso]

Welcome to the forum Ted.

 

[rlaury]

Hi Again Ted:
I tried to PM you. Not sure it was sent or not. Please PM the info.

BTW I met you at the NAAM show here in Nashville a few years back.
Dan Penn and I came by your booth. I think Dan bought a compressor from you. Hope your doing well.

RonL
http://www.nashaudio.com