Is the BA-660 For You?
...Smooth,...Fat,...Warm,...Silky,...Analog
..Tres Violent,…Bad Ass, …Dangerous
The BA-660 is a tool. And like any tool, it has it’s uses and it’s abuses. This machine is not for everyone. The faint of heart should not hear-by apply. There are a number of concerns which are applicable to the BA-660;
We optimize the performance of each stage, for the widest possible dynamic range; i.e. the lowest noise and highest output. In order to do this we employ relay switching in the bias circuitry of the tubes. This upsets the DC balance of each stage, (Quite considerably in some instances!). With ~90db of gain, a slight imbalance of the input stage could be quite spectacular indeed! In a recording environment, this should not prove to be a problem, however, for live work, Oh Well????
The output stage is a DC coupled, Class “A” solid state current buffer, with DC servo to correct for offset. We can get rid of the pop, but at the expense of the low frequency response of the output stage. I firmly believe that the small pop is a small price to pay for the linear response down to 6 Hz. Like the mute, in a recording environment, this should not be unworkable. You adjust the gain, and then forget it while you record the performance.
The above mentioned output stage is balanced, truly. When you connect one side to ground with no additional compensation, it “barks” at you. This is normal. There is a recessed 3 position slide switch in the rear to accomplish the required compensation. In one position, it will sound the best, use it in that position.
This is a tube device. It takes a while to heat up and reach proper operating temperature. The devices that drive the meter are temperature sensitive. You should find that upon reaching terminal temperature, (5-15 minutes) the meter should rest at “0”. If it goes beyond, it needs tweaking. Use the recessed multi turn pot on the LEFT side of the gain reduction meter. Turn the threshold fully clockwise, with no signal, set the meter for “0”
After you have turned the pot on the right, in order to recalibrate the gain cell,;
A. Apply a steady signal,
B. Set the input monitor select to the “out” position
C. Put the unit in Bypass
D. Set the output trim so that the LED meter reads “0”
E. Turn the threshold fully clockwise
F. Take the unit out of bypass
G. Set the trim pot for unity gain. (The same level as bypass)
H. Switch back and forth between bypass and confirm that the level is the same.
In order to maintain the internal signal integrity and at the same time bring the side chain out for processing, I employ a hardwire relay bypass to the insert send and return buffering electronics. These buffers, when used, establish the maximum operating headroom at +22, Way too low for my comfort. So I bypass them with a relay when they are not being used. This allows for >+40 maximum headroom. Well, the sensor for this switch circuit take this long to charge up. It happens once, upon turn on, and then stays put. It is another indicator that the unit has reached temperature.
The system is truly balanced from the input to the output. The input impedance of the front end is on the order of 2meg ohms, reduced with a termination ‘R’ but still high. The shield is dropped on the input and carried on the output. The greatest likelihood exists that this is being driven from a single ended source, and the output shield is the only ground reference that the unit sees. This will introduce DC offset and is most easily cured by attaching a ground wire from the posts on the rear to the console chassis. We could cure this “problem” by carrying the input shield, but this in most instances would increase the overall noise floor, especially when properly earthed.
The above list summarizes the “warts” of the BA-660 Mic Pre-Comp/Lim as I know them. These issues are not going away. The “solution” to the above situations would compromise the sonic integrity of the box, and this is not acceptable to me. In some issues I am intractable, please forgive me.
Wither Tubes?
...Smooth,...Fat,...Warm,...Silky,...Analog.
These are just some of the terms used to describe the sound of vacuum tubes. A technology in resurrection. A boon for the listener and a bane for the designer. From an engineering standpoint there are many difficulties with tubes;
- Low gain
- High electrical noise
- High distortion
- Thermal instability
- Low mechanical stress thresholds
- HIGH VOLTAGE!...(DANGER)
- High output impedance
- Low drive capability
Faced with the question of why use tubes, aside from the obvious marketing advantages, we have to look at what a tube offers that semiconductors don't..... Headroom! (a by-product of above item 6). A little background is in order. Of the five senses, hearing has the widest dynamic range. On the order of 7 decades (140db), our ears rival the sciences to accommodate with fidelity the capturing of perceptible acoustic phenomenon. Because of this, I have set out to produce the world's finest Microphone Pre-amp with the widest dynamic range possible within practicable means (154 db) and couple it to the best limiter that can be built. This device can present to standard studio equipment a signal of exceptional purity within the confines of our current digital dynamic range window of 96db(16bit), [or 120db(20bit)].
HOW DO WE DO IT?
When examining 'classic' tube designs, the trade-offs made at the time of the design must be examined and the logic behind them explored. In the '50's and '60's the cost of components, (Caps, Precision resistors and tubes) was considerable. Most engineering effort went to squeezing the last ounce of performance out of every part used, then using less parts, at a subsequently lower cost. A four stage design was extravagant indeed! The audio benefits of such an approach are those of minimal signal path, but at the expense of stability, linearity, distortion and circuit flexibility.
Semiconductor technology has brought about enormous economies of scale. Mass assembly methods have reduced the expense of discrete transistors, resistors, capacitors and sockets. The cost of components has thus plummeted to a level where the sheer number of devices is no longer the primary bounding element in circuit topology.
Exploiting technology in the production of our tube Mic Pre/limiter line we have capitalized on the benefits of tubes by utilizing semiconductors for the stabilization and bias of the tube elements without the necessary sonic trade-offs, (Audio passes only through the tubes) and without the concurrent trade-offs of noise, distortion, thermal instability and high output/low load impedance that have previously plagued tube design. With careful attention to parts selection and biasing choices, the Mic preamp has a noise figure that is set by the losses of the input transformer. We achieve the exceptional noise performance of the limiter by utilizing the wide headroom margin internal to the Mic-Pre/complim(+40db), instead of requiring 25 to 40db of attenuation of the input signal to provide operating range.
The most significant tradeoff necessary in this implementation is that I have not used global feedback and have kept local feedback limited to DC bias. This means that the measured distortion figures are high (<.7%). The use of feedback (which is how all low distortion figures are achieved) brings a new set of problems which have adverse sonic effects. The dichotomy is that equipment with low distortion figures achieved via large amounts of feedback don't sound as good as gear that has high(er) distortion numbers but achieves fidelity through careful balance and bias.
Our patent pending methods of tube control yield performance heretofore unachievable and provide pleasing sounding equipment capable of handling any signal that current transducer technology can source, now and into the forseeable future. Have a listen.
Sincerely,
Greg Hanks