Best Console Summing amp/Master Buss comp Ideas....

Hi
I really enjoy all you people and your ideas, so I thought I would pose a question and get opinions.

*Warning very subjective and opinion based matters follow! Bring it on*

I want to take out the Master channel summing makeup gain ne5534 opamps and build an out board sum amp.

I'm thinking building it out into a seperate box because even trying to put some DOA in place of were the ne5534 chips were seems cheap.

there will be about 20 db max of make up gain, and I want to build a compressor in with it, while I'm at it.
what seems to be the most diverse kind of compressor? Vari-mu? of course there is the gssl but I want to try somthing different.
Also I think the option of DOA or tube make up gain would be very sweet...
any Ideas welcome....

abechap024 said:

Hi
I really enjoy all you people and your ideas, so I thought I would pose a question and get opinions.

*Warning very subjective and opinion based matters follow! Bring it on*

Click to expand...

*Warning objective and experience based answer.

I want to take out the Master channel summing makeup gain ne5534 opamps and build an out board sum amp.

Click to expand...

Physical layout and routing of sensitive buses can be problematic. IMO you are asking for trouble (hum + crosstalk) moving the sum amp out board.

I'm thinking building it out into a seperate box because even trying to put some DOA in place of were the ne5534 chips were seems cheap.

Click to expand...

Cheap is good... You can see quite a bit of improvement from a drop in replacement, if that (5532) is indeed the weak link in your signal chain.

there will be about 20 db max of make up gain, and I want to build a compressor in with it, while I'm at it.
what seems to be the most diverse kind of compressor? Vari-mu? of course there is the gssl but I want to try somthing different.
Also I think the option of DOA or tube make up gain would be very sweet...
any Ideas welcome....

Click to expand...

Compressors are subjective and while I have designed compressors I will not proffer an opinion, other than that most popular music is toooo compressed these days.

JR

Hmm Thanks john
Its not that my console sounds bad.... the ne5534 sound pretty darn good. But I have this horrible voice in my head that says "...but it could sound sooo much better" so this will basically be for fun/learning.  and your right about over-compression. thats why I was thinking the vari-mu because of the sublety. I could always patch in an ssl when that is the sound I want...
But thank you for the input. This is one of those projects thats on "if i get around to it" basis right now. so just formulating ideas...

Look at Jeff Steiger API summing amp kit, I like mine & haven't heard of anyone who was disappointed with theirs.

I want to take out the Master channel summing makeup gain ne5534 opamps and build an out board sum amp.

Click to expand...

Don't confuse an active summing amp with a passive summing network plus makeup gain, because it isn't. The inverting op amp input in an active summer is a virtual ground point and the input resistors are part of the feedback loop.

Physical layout and routing of sensitive buses can be problematic. IMO you are asking for trouble (hum + crosstalk) moving the sum amp out board.

Click to expand...

+1

Ahh yes I think those API DOA are very nice...

johnR said:

I want to take out the Master channel summing makeup gain ne5534 opamps and build an out board sum amp.

Click to expand...

Don't confuse an active summing amp with a passive summing network plus makeup gain, because it isn't. The inverting op amp input in an active summer is a virtual ground point and the input resistors are part of the feedback loop.

Click to expand...

Thanks for bringing that to my attention, looks like I need to do some more research on the subject. its very interesting. and looking at the circuit again I'm pretty certain that my board is just a passive resistor sum and op amp make up gain. would it be and advantage to convert it into an active summing system?

abechap024 said:

and looking at the circuit again I'm pretty certain that my board is just a passive resistor sum and op amp make up gain.

Click to expand...

That is extremely unlikely, since the component count is exactly the same and the performance so much better.

Yes, your right. Just updated myself with fred forssells http://www.forsselltech.com/downloads/design_discussions/summing_buss.pdf
again. and you are exactly right its active. and now I see why it wouldn't make sense to have the makeup gain out of the console. I'll probably stick with the ne5534s for the time being.

abechap024 said:

Yes, your right. Just updated myself with fred forssells http://www.forsselltech.com/downloads/design_discussions/summing_buss.pdf
again. and you are exactly right its active. and now I see why it wouldn't make sense to have the makeup gain out of the console. I'll probably stick with the ne5534s for the time being.

Click to expand...

IMO, there's room for improvement. The 5534 is optimised for medium Z (3-4kohm), and a mixing bus is low Z (ca. 250 ohms).
Replacing the 5534 with a VLN opamp (ca. 1nV/sqrt Hz) could bring a seriuos improvement in noise and clarity.
Unfortunately, most of the VLN opamps (LT1028, LT1115, AD797, OPA847,...) exist only as singles, some only in SOIC package, so you would need some kind of adapter.
Extra caution must be taken with PSU decoupling and the current consumption is noticeably higher, about 2-3 times more than 5534, and of course, any less than perfect grounding of the reference point of the virtual earth ruins any effort.

Without getting too esoteric, another factor in bus combining is stability compensation. The 5534 is better than some unity gain stable peers, since it is under compensated for use above 10dB of closed loop gain. The ability to benefit from even less compensation depends on how the signals are switched onto the bus.. If the send are always attached you can live with less compensation, than for the case when all sends can be removed from the bus.

One technique I considered was using the other throw of bus assign switches to back ground the bus through a small value RC to provide the apparent closed loop gain only at very HF so stability was insured without audible noise gain when channels were de-assigned.

I will second that there are probably newer drop in parts somewhat better than 5534, and depending on the actual topology consider another non unity gain stable part, like the 5534.
 
JR

JohnRoberts said:

I will second that there are probably newer drop in parts somewhat better than 5534, and depending on the actual topology consider another non unity gain stable part, like the 5534.

Click to expand...

Old thread, but I've been doing a ton of reading about summing bus circuits this month, and thought you might find this Douglas Self article of interest:

http://www.edn.com/design/audio-design/4218273/3/Op-amps-in-small-signal-audio-design---Part-3--Selecting-the-right-op-amp

In which he says:

It has taken an unbelievably long time – nearly 30 years – for a better audio op-amp than the 5532 to come along, but at last it has happened. The LM4562 is superior in just about every parameter, but it has much higher current noise.

Click to expand...

Note that the LM4562 is also known as LME49720, and these are dual opamp chips. It's available in single opamp form as the LME49710.

leigh said:

Note that the LM4562 is also known as LME49720, and these are dual opamp chips. It's available in single opamp form as the LME49710.

Click to expand...

In the particular case of summing amps, the 4562 would not make a huge difference over 5534. It would over 5532, though.
Assuming a mixer with 48 channels routed via 10k resistors, the noise level would be:
5534: -89.3 dBu
5532: -83.9
4562: -91
An improvement of a little over 1.5dB, which is always nice, but nothing to write home about.
A more remarkable improvement would be using an opamp dedicated to low impedance sources, such as the AD797; in that case, the noise level would be -95.1, almost 6dB improvement over the 5534.

With only 16 channels routed, the difference would be even smaller, with:
5534: 97.4
5532: -95.4
4562: -98.5
The 797 would still be the champ with -100.6 dBu, only about 3dB better than the 5534.

I've neglected all other aspects of performance because I don't see any significant difference in terms of power bandwidth, output capability or THD.
There may be sonic differences but they aren't in the specs  .
The art of summing amps has been perfected about 20 years ago, with the use of hybrid designs (BJT + opamp); the new twist is that there are monolithics IC's that can bring the same level of performance. One cannot assume that it is now only a matter of swapping chips, because parasitics may introduce new problems, and anyway, real-life performance is never as good as theory.

abbey road d enfer said:

The 797 would still be the champ with -100.6 dBu, only about 3dB better than the 5534.

I've neglected all other aspects of performance because I don't see any significant difference in terms of power bandwidth, output capability or THD.
There may be sonic differences but they aren't in the specs  .
The art of summing amps has been perfected about 20 years ago, with the use of hybrid designs (BJT + opamp); the new twist is that there are monolithics IC's that can bring the same level of performance. One cannot assume that it is now only a matter of swapping chips, because parasitics may introduce new problems, and anyway, real-life performance is never as good as theory.

Click to expand...

Not to quibble but I wrote my console performance article in 1980 so almost 35 years ago.

I mentioned the Transamp as hybrid discrete-op amp, sum amp which has the added benefit of open loop gain increasing with closed loop gain for constant and superior loop gain margin.

I also spoke in broad strokes about my current source summing approach (current sources instead of resistors feeding virtual earth sum bus).

As I have said over the decades the noise floor is not the real benefit of higher performance bus topology, but reduced phase shift and/or lower distortion. I am not suggesting that phase shift and distortion are glaring shortcomings in analog consoles, just that there is more than noise involved in sum bus amp performance.

I also see little reason to bother with analog combining when digital can do it without any degradation at all (at least in theory). I really wish analog consoles were still worth developing because I have not finished perfecting my (analog) ideas, but at this point I see little market or commercial value in doing so.

JR

PS: I concede that many who read my article thought I was blowing smoke.... 8) While a few smart designers understood it, and one even told me years later that he used current source summing in a broadcast console he designed.

JohnRoberts said:

I really wish analog consoles were still worth developing because I have not finished perfecting my (analog) ideas, but at this point I see little market or commercial value in doing so.

JR

PS: I concede that many who read my article thought I was blowing smoke.... 8) While a few smart designers understood it, and one even told me years later that he used current source summing in a broadcast console he designed.

Click to expand...

Well I (and most certainly other too) would be happy to get a glimpse of that work beyond the mentioned paper. any rights refraining you from publishing some schematics and details? If the is no commercial value to you this would be even more so a reason to publish some details.....

- Michael

audiomixer said:

JohnRoberts said:

I really wish analog consoles were still worth developing because I have not finished perfecting my (analog) ideas, but at this point I see little market or commercial value in doing so.

JR

PS: I concede that many who read my article thought I was blowing smoke.... 8) While a few smart designers understood it, and one even told me years later that he used current source summing in a broadcast console he designed.

Click to expand...

Well I (and most certainly other too) would be happy to get a glimpse of that work beyond the mentioned paper. any rights refraining you from publishing some schematics and details? If the is no commercial value to you this would be even more so a reason to publish some details.....

- Michael

Click to expand...

It's not much of a secret..While I was tight lipped back in 1980, the much later AMR (Peavey) Production Series console owners manual came with complete schematics.

I can even tell you what I did wrong with that last design version back in the '90s (besides trying to sell a big split console while Mackie was sucking all the oxygen out of the room with his 8-bus).

I don't generally keep schematics for all my sundry designs over the years but a console owner posted a scan of the OM on Geekslutz a year or two back so it's around..  http://www.johnhroberts.com/AMR_manual.pdf (hint- The current sources are in the monitor modules hard wired to the L/R bus.)

I had an improved revision working in the lab and ready to go to production, when the powers-to-be killed the console due to weak sales. It was never a great fit for Peavey marketing/distribution, but the Mackie inline console for only $4k got all the love back then. 

I even talked with THAT engineers about repurposing the precision resistors in one of their I/O chip series to make a very nice current source, but they did not have enough protection circuitry on the chip to handle that with just a metallization layer change. They insist on protection circuitry for every external pin and I would have left one unprotected.

Sorry about the veer.... this really is old news now....

JR

PS: Please don't bust me over decisions I made in that console, there are still happy owners using them today decades later, so more right than wrong.

abbey road d enfer said:

In the particular case of summing amps, the 4562 would not make a huge difference over 5534. It would over 5532, though.

...

I've neglected all other aspects of performance because I don't see any significant difference in terms of power bandwidth, output capability or THD.
There may be sonic differences but they aren't in the specs  .

Click to expand...

Actually, there are some significant differences in the other specs as well. I have been focusing on comparing the single-amp packages, so it has been the NE5534A vs the LME49710 (vs some others as well). As you point out, these two are very close on their noise performance, when used in a low source resistance (< 1K) circuit like a summing amp.

Some other notable differences:


In every other spec, the LME49710 measures better or about the same, with the exception of Current Noise (on this it is 4x higher than the 5534)... but in this application, the contribution of current noise to the total noise figure of the summing amp is tiny.

cheers,
Leigh

leigh said:

Actually, there are some significant differences in the other specs as well. I have been focusing on comparing the single-amp packages, so it has been the NE5534A vs the LME49710 (vs some others as well). As you point out, these two are very close on their noise performance, when used in a low source resistance (< 1K) circuit like a summing amp.

Some other notable differences:

	NE5534A -----	LME49710
Open-Loop Gain (@1kHz)	100 dB	140 dB

Click to expand...

In a VE sum amp the op amp's open loop distortion (?) and open loop phase shift (90' above dominant pole for one pole compensation) is reduced by the ratio between open loop gain and closed loop gain aka loop gain margin. An extra 40 dB of loop gain margin will reduce phase shift and distortion a factor of 100x all else equal.

CMRR

100 dB

120 dB

Click to expand...

Not generally significant unless you run sum amp differential. And even then probably not audible.

PSRR

100 dB

125 dB

Click to expand...

This PSRR is reduced by the closed loop gain which could be significant in large VE structure so extra 25 dB of PS hum suppression is always a good thing.

Gain Bandwidth

10 MHz

55 MHz

Click to expand...

Probably not a huge issue in  20 khz bandpass but that GBW may be a consequence of the large open loop gain.

THD (@ 1kHz)

0.002 %

0.00003 %

Click to expand...

This is a meaningless spec without a schematic. While if all else equal,  better is better. Some of these op amps have to run high closed loop gain just to make measurements, while it should translate for VE application.

In every other spec, the LME49710 measures better or about the same, with the exception of Current Noise (on this it is 4x higher than the 5534)... but in this application, the contribution of current noise to the total noise figure of the summing amp is tiny.

cheers,
Leigh

Click to expand...

Have you A/B'd them yet?  Even with those GOP (good on paper) specs they may sound similar.

JR

Thanks for weighing in on the significance of these specs, John.

JohnRoberts said:

THD (@ 1kHz)

0.002 %

0.00003 %

Click to expand...

This is a meaningless spec without a schematic. While if all else equal,  better is better. Some of these op amps have to run high closed loop gain just to make measurements, while it should translate for VE application.

Click to expand...

Attached is the Trident S65's summing amp schematic, annotated. It is obscured by the shadow, but after R133 the signal runs to the remix insert send, so this is the VE summer and then the insert send driver. So, that first stage is 12k summing resistors into an amp with a 4k7 feedback, and a 100p compensation cap. What does that tell you about THD performance?

JohnRoberts said:

Have you A/B'd them yet?  Even with those GOP (good on paper) specs they may sound similar.

Click to expand...

Haven't yet gotten to A/B... finished my spreadsheet of opamps last week, and been working on other (grounding) issues since then. 5534's are on order, and will get to A/B them once the grounding work is finished.

cheers,
Leigh

leigh said:

Thanks for weighing in on the significance of these specs, John.

JohnRoberts said:

THD (@ 1kHz)

0.002 %

0.00003 %

Click to expand...

This is a meaningless spec without a schematic. While if all else equal,  better is better. Some of these op amps have to run high closed loop gain just to make measurements, while it should translate for VE application.

Click to expand...

Attached is the Trident S65's summing amp schematic, annotated. It is obscured by the shadow, but after R133 the signal runs to the remix insert send, so this is the VE summer and then the insert send driver. So, that first stage is 12k summing resistors into an amp with a 4k7 feedback, and a 100p compensation cap. What does that tell you about THD performance?

Click to expand...

The schematic I was talking about is the schematic those distortion measurements were made with. An op amp will have X % THD operated open loop, and when you wrap NF around it that open loop distortion is reduced by the loop gain margin (ratio between open loop and closed gain.) I will ASSume the distortion numbers were made with comparable circuits. 

While I don't like to critique other console designs, I have a couple quick observations.  The 12k input resistors and 4.7k  feedback is reducing the worst case closed loop gain several dB which is "improving" the loop gain margin by those same several dB.  So this helps reduce phase shift and distortion in the first stage if/when a lot of stems are assigned. The following stage with 4.7k input and 12k feedback restores the gain to nominal unity and is relatively easy lifting for a decent opamp. There is no free lunch  so there is diminishing return from putting too much gain in the second stage, but for modest amounts this makes sense, especially for old school op amps. The phase shift and distortion added back from the second gain stage will be less than the amount of improvement from running the first stage with less gain. . 

This gain sharing or distribution could make a difference especially if the first stage op amp is marginal.  Note: the 5534 is only stable down to a gain of 10 dB (3x) so if those 12k input resistors are not always assigned a 5534 could be unstable without the extra compensation cap. The 49710 is unity gain stable so that could be useful in that circuit configuration.
 

JohnRoberts said:

Have you A/B'd them yet?  Even with those GOP (good on paper) specs they may sound similar.

Click to expand...

Haven't yet gotten to A/B... finished my spreadsheet of opamps last week, and been working on other (grounding) issues since then. 5534's are on order, and will get to A/B them once the grounding work is finished.

cheers,
Leigh

Click to expand...

Don't A/B it for me... the 49710 looks like a superior part...

JR

And just to add to Guru JR's comments, you need the highest standards of Decoupling, Grounding & Layout to get anywhere near the potential of 5534.

There's probably only a literal handful of people on this forum capable of doing this and even less if the system is large & complex ... like a big mixer.

The same numbers also applies to Mixer manufacturers.

In most designs, the yucky stuff introduced by poor Decoupling, Grounding & Layout, including spurious, marginal instability & THD which comes & goes, swamp any differences between OPAs .. except that most uber OPAs are even MORE FUSSY than ol' 5534 and often give worse performance in old gear.
________________________________

If you are versed in the art, you will look at the performance you get now from 5534 and compare with what your have calculated.  If they are far out, the problem isn't to do with 5534 ... and using an uber OPA will likely make things worse.

Can you actually measure the THD of your Trident to these levels?  And do serious work on stability?