Hello all QQ people,
Rolvkarsten, The amplifiers of the CD-4 era were not bad. Homotaxial base and triple defused base high speed bipolar power transistors had been developed and everyone was using them. These transistors were quite fast for power transistors and would yield low distortion (THD, IM, TIM) in a typical amplifier.
PNP power transistors were harder to make, so they cost a lot more than the equivalent NPN type. The universal amplifier circuit was the complementary symmetry op amp configuration. This used a complementary output transistor pair (NPN,PNP). Since the PNP transistors were so expensive an engineer at RCA semiconductor came up with a modified design known as the quasi-complementary symmetry op amp configuration. In this design, there are two Darlington pairs of transistors, with one side being a virtual PNP. In the schematic that I have included with this posting, you can see how the virtual transistors are connected. This configuration makes up 98 % of the analog audio amplifiers. Otala's amplifier is the same basic configuration with the exception that there is an extra differential gain stage. This makes the overall open loop gain higher, making feedback lower for higher gain. Otala set the output stage mode to class A, which automatically lowers the amplifiers TIM and IM distortion. The THD however, did not improve because of the beta collapse of the power transistors.
Beta is the dc current gain of a bipolar transistor. As the current through the transistor increases the beta decreases. This decrease is called beta collapse. Most solid state amplifier designers assumed that Darlington output configuration would correct beta collapse. Darlington beta is measured by multiplying the beta of the first transistor by the beta of the second transistor.
At lower current levels the individual betas may be something like 100. That would give a Darlington beta of 10,000, which is very good. At high current the individual beta of the second transistor can decrease to 3. This is because the second transistor drives the speaker directly. The Darlington beta becomes 300, which is quite poor. This results in rising THD as the output power rises.
The solution to the problem came a few years later when the triple Darlington configuration was employed. With this configuration, at high power, the first two transistors have a beta of 100 each and the output transistor has a beta of 3. The Darlington beta equals 100 X 100 X 3 which is 30,000. This is excellent and will reduce the THD by 20 to 30dB. Other distortions also improve and the amplifier mode can be changed from class A to class AB2 which is much more efficient.
Quadfather,
It is soap box time again! During the original NQRC field trials, we set up two independent double blind tests to determine the average listeners ability to perceive point source imaging, phantom imaging, frequency imaging, and position imaging verses channel separation. One study was done in San Francisco and the other was done in Washington D.C. Double blind protocol means that no one involved in the test knows any information about any of the test. That includes the administrators and the auditors.
The results were that the average auditor could easily determine point source and phantom source position regardless of frequency. Yes, in fact bass is directional. Prove it to yourself. Next time you go to a fireworks display, you will notice that when a mortar explodes to the left or right of you, the bass does not come from a center position. It comes from the location of the mortar explosion.
So how did 5.1, 6.1, 7.1 etc come about. Not by some high technology means. It is technology being driven by business. Speaker manufactures were worried about getting 4 full range loudspeakers in the home. Film sound mixers were worried about mixing dialog using well know phantom techniques because of the training of the personnel in the theaters in setting up the playback system as well as the home users ability to set up a home system.
From this came 4 small satellite speakers with only midrange and tweeter speakers, a sub-woofer, and a dialog speaker. Let me state to all of the Quad people, you can do everything with a discrete Quad playback system that can be done with 5.1, 6.1 and 7.1 and do it better. First you must have 4 full range loudspeakers. If you want low bass emphasis then you want a 4.4 system. That is four full range loudspeakers(30 Hz to 20 KHz) and 4 real sub-woofers(15Hz to 50Hz). A dialog speaker is not needed! If an actor is on the left side of the screen, then that is where his(her) voice should come from. If that same actor walks across the screen the sound should follow the picture, not just come out of the center. All of this is easily done by using point source and phantom imaging. I have mixed many Quad pieces, both music and stories with actors and have had excellent results.
One additional item is playback setup. You all know how time consuming 5.1 etc setup is. The band aid circuits in 5.1 receivers are overwhelming. Quad receivers need none of this. Just four sources, four amplifiers, and four full range loudspeakers. Discrete Quadraphonic sound provides more accurate performance then any 5.1, 6.1, 7.1 systems with simpler hardware and software.
kfbkfb,
Kirk, Why? Discrete Quad FM does everything 5.1 does, only better!
QuadroAction,
Dietrich, Compound work came after, Q540 was born about 6 months later.
Malcom2010,
Malcom, it is a custom chip I had made for another project. It's number is XY4240.
winopener, I still use the "QSI record Cleaner - For Superior Sound" (That was our slogan) to clean records and CD/DVDs.
Well that is it for now,
Lou Dorren