SumWare Over the Rane, Bro

[WeAreAs1 at aol.com]

Hello listfolk,

As I have previously mentioned, I have a Yamaha recording mixer, model 
RM2408.  It was designed and built in the mid-1980's (first produced around 
1984).  I'm in the process of trying to upgrade it (rather than doing the 
sensible thing and buying a new mixer..).  Most of the opamps in it are 
4558's, so naturally, most of them are going to be replaced with more modern 
parts.  However, it also contains several Toshiba TA7322P opamps.  These are 
used specifically in the summing busses (8 submix busses and the main stereo 
mix buss).  These TA7322 amps are the actual summing amps.

Obviously, Yamaha chose this particular part for some reason.  At the time, 
it was probably a good reason (probably not for low cost - this was Yamaha's 
best recording board).  I'm pretty sure that that there are more modern parts 
that could be used to replace these old amps that would outperform them in 
most ways.  Nevertheless, I would like to know why Yamaha chose this 
particular opamp.  Such information might help me in making a wise choice for 
an upgrade part.  I was finally able to procure a data sheet for the part 
from Toshiba America, but I haven't yet been able to see what makes this part 
so appropriate for the summing buss application.  I'm hoping that some list 
members may be able to help me understand.

This opamp is referred to by Toshiba as a "High Voltage, Super Low-Noise 
Preamplifier".  It's a single opamp in a 9-pin SIP package.  The data sheet 
is dated July, 1980.  The spec says it can be operated at supply voltages up 
to plus/minus 30 volts DC!   however, Yamaha is powering it with a 
conservative plus/minus 16 volts, so they're not taking full advantage of its 
maximum output voltage swing.  The spec sheet claims a maximum output voltage 
of around 15 volts when the amp is powered at plus/minus 25 volts and about 
10 volts when it is powered at plus/minus 16 volts.   These output voltages 
are measured at 0.01% THD.  I would assume that larger voltage swings are 
possible, but with rapidly increasing distortion.

The part also has a few extra pins on it that are not seen in most modern 
opamps:  a "Bias" pin, and two "Phase Compensation" pins.  Only one of the 
"Phase Compensation" pins is used by Yamaha (as well as in the circuit 
examples shown in the data sheet).  The data sheet does not give any 
explanation about the function of either of the Phase Compenstation pins, nor 
does it explain how to use them.  Toshiba (and Yamaha) shows one of the Phase 
Comp. pins connected to the minus supply rail via a resistor and capacitor in 
series.  The other Phase Comp pin is simply left unconnected.

I'm aware that there are some older opamps out there that are referred to as 
"uncompensated" (such as the old LM301 found in so much ARP stuff), and that 
most of the opamps we use these days are "internally compensated".  Nowhere 
in the spec sheet does Toshiba specifically refer to the amp as an 
"uncompensated amp".  Do you think that these "Phase Compensation" pins are 
the same thing as the compensation pins found on other uncompensated opamps?  
If so, would someone care to offer a brief explanation of their purpose and 
function?  

The "Bias" pin is meant to be connected, via a series resistor, to the plus 
supply.  The value of this resistor is to be selected according to the chosen 
power supply voltages.  For example, most of Toshiba's circuit examples give 
a value of 91K, operating the amp on a plus/minus 25 volt supply.  They give 
a formula for selecting the resistor:

         Vcc-(Vee)-5.3
Rbias = ________________
                         I bias 

They recommend an I bias current of 500uA.  If we plug Toshiba's plus/minus 
25 volts and an I bias of 500uA into the equation, we end up with an Rbias 
value of 89.4k ohms.

Yamaha is using a 121K resistor at the bias pin (and operating at plus/minus 
16 volts), which means they are running the amp with a lower I bias current 
(about 220uA instead of 500uA).

I can't seem to figure out what adjusting the Bias current does to the 
performance of the amp.  According to an "Output Noise Voltage vs. Bias 
Resistance" graph that is given in the datasheet, it appears to have some 
effect on noise performance, but certainly not a huge effect.  I wonder if it 
also has some effect on available output voltage swing? (or available output 
current?)  Will reducing the Bias current work as a kind of "soft limiter" 
when output voltage limits are approached?  Such a thing might be useful in a 
hard-working mixer summing stage.  BTW, this Bias pin is not an input bias 
pin.  The simplified internal schematic in the data sheet shows that it is 
connected to an intermediate stage of the amp, after the differential input 
stage and before the final push/pull output transistors.

So, to sum things up:  

*  What is it about this amp that makes it so well qualified for the job of 
summing amplifier?
*  Do summing amplifiers have special requirements (as compared to other 
stages, such as gain stages, EQ blocks, output drivers, etc.)
*  What modern amps can exhibit the same qualities, but with better sonics? 
(I note that Mackie uses NJM2068 opamps in all of their main summing stages - 
they use NJM4560 opamps everywhere else in their boards.)  The new amp 
doesn't have to be a plug-in replacement.  I can make an adapter board if 
needed.  I would probably prefer to keep the gain structure and input 
resistors the same, if at all possible (16K input resistors on the inverting 
summing node with a 16K feedback resistor around the amp).
*  What kind of off-the shelf opamps are used as summing amps in some of the 
more high-end mixing consoles? (such as Neve, SSL, Harrison, Amek, etc.)  I 
realize that some of them may have custom-designed summing amps made from 
discrete components.  Would there be a great advantage to using an 
all-discrete amp at this stage (such as the Deane Jensen 990 opamp)?

Yeah, I'm rambling.  Sorry.

If anyone wishes to see the Toshiba datasheet (5 pages) or a schematic for 
the Yamaha summing amplifier stage, I can scan them and email them to you.

Best regards,
Michael Bacich