Re: connection of Q7, Q8
Actually, Bob, if the output stage is biased in Class B/AB, transistors Q7 and Q8 will also operate in Class AB, particularly with adverse loads.
Only devices Q5/Q6 operate in class A regardless of output loading, provided their emitters are cross-coupled by means of a sufficiently small resistor.
This is easily verified in practice and SPICE.
Bob Cordell said:
Actually, Bob, if the output stage is biased in Class B/AB, transistors Q7 and Q8 will also operate in Class AB, particularly with adverse loads.
Only devices Q5/Q6 operate in class A regardless of output loading, provided their emitters are cross-coupled by means of a sufficiently small resistor.
This is easily verified in practice and SPICE.
Hi Mikeks
---The temptation to take this ''idea'' [the feedback point is taken at the node between the resistor and the grounded capacitor of the ouput RC Zobel network] at face value must, in this context, be avoided at all costs, otherwise the RC network introduces a conspicuous pole in the major feedback loop that virtually guarantees instability.---
I recently used the idea when building a power supply based on an LM723 which seems to work fine. However your comment reminds me that I did not make any stability control.
---The temptation to take this ''idea'' [the feedback point is taken at the node between the resistor and the grounded capacitor of the ouput RC Zobel network] at face value must, in this context, be avoided at all costs, otherwise the RC network introduces a conspicuous pole in the major feedback loop that virtually guarantees instability.---
I recently used the idea when building a power supply based on an LM723 which seems to work fine. However your comment reminds me that I did not make any stability control.
Re: Re: Re: Re: connection of Q7, Q8
Resistor is between their emitters, it always loads both of them, no matter how big base currents outpit transistors consume.
Howeve, output transistors left alone with naked base - emitter junctions to fight against capacitive loads, and it is bad, but it is another story...
mikeks said:
Resistor is between their emitters, it always loads both of them, no matter how big base currents outpit transistors consume.
Howeve, output transistors left alone with naked base - emitter junctions to fight against capacitive loads, and it is bad, but it is another story...
moamps said:In my opinion, the biggest source of EMI today are GSM phones and wi-fi networks, rather than nearby AM or FM commercial transmitters.
On the other hand, I’ve heard some acquaintances of mine talk about very audible effect of wi-fi networks on the sound reproduced by their home audio systems. I’m guessing that any negative effects of wi-fi networks on home audio systems will be tracked to interconnects, rather than amplifier output.
So, although the amplifier-speaker connection has a very low impedance at audio frequencies, at higher frequencies it becomes a system with distributed parameters, where certain parts of the system can act as isolated reception antennas. Personally, I don’t believe that the attenuation of the signal from the amplifier’s output (with NFB) to the inverting input (loop gain) at audio frequencies works well for high frequencies too. Now, I’m sure that anyone out there who has ever had the misfortune of witnessing the sound of an amplifier-turned-oscillator needs no ghastly reminder from me of what a HF signal can do to an amplifier.
Regards,
Milan
Milan,
Thanks for taking the time to respond. In my experience all interference (unless the sensitive circuit is sitting next to a high power transmitter) can be dealt with through proper layout and effective grounding (and shielding). The rules are pretty straight forward but to get the hang of it takes a problem to solve and an open mind.
As for the amplifier-turned-oscillator scenario we've all experienced, you'd be amazed at what controlling the functional loops in your circuits will do to eliminate those pesky little instabilities. Has it never bothered anyone that there is no audible content above the 20khz range, yet the need is felt to design circuits that are stable from DC to light? Only with the advent of diamond tweeter domes have speakers even been able to produce sound above that freq. We place a lowpass filter on the input to the amplifier circuit, give it a high frequency pole and that pesky energy still seems to develop somehow and make our circuits ring.
In general, please excuse my attempts to rattle anyones' cage enough to spark a bit of conversation that wasn't steeped in the past and parroting the well documented truth. It's basically a bit of frustration on my part. Seemed like a good idea at the time.
And now back to your regularly scheduled programing, already in progress.
Regards, Mike.
The only place I've had RF enter the audio chain was back when Citizen's Band was popular. It came in through the mm-phono input even when the turn table was switched off. I assume it was the high sensitivity of the phono input compared to tape, tuner or aux. Select phono, turn up then volume and listen to "breaker, breaker good buddy". I thought this as likely as Aunt Tilly's claim of getting Radio Moscow on her bridgework until I actually heard it.
Speaking of RFI... 
About 30 years ago...
I remember horrible days when I was a student and there was an amateur station on the 9'th (the top) floor of our building... The roof was covered by sophysticated antennae, it was TIASUR...
We vere very creative and used to invent all possible and impossible ways to protect our tape recorders and turntables applying all knowlege we were getting on lectures and laboratory works, but nothing could help, because during contests they generated such power so electric bulbs in the whole building were dimming in tact with their distorted SSB speach from our Hi-Fi loudspeakers...
Once I've found a genuine solution. If it is impossible to fight against them passively, it may be possible to fight actively deafing them out.
I quickly made an oscillatir with one tunnel-effect diode tuned on 3.5 MHz frequency. It produced lot of harmonics, so all amateur bands (3.5, 7, 14, 21, 28 MHz) were taken care of. I used a varactor diode for frequency modulation, and one 2-base diode for triangle sweep generator. The whole thing including couple of 9-volt batteries was hidden inside of a slipper under the bed, thin wires were going to the metal frame of the bed and to the central water-heater battery.
After I switched on my device they continued traransmitting a little bit, then The Silence Come...
I laid down on my bed and switched on Deep Purple "Sweet Child in Time, You See The Line" on my turntable...
I enjoyed the music when a roommate who was a radio amateur come... He sat down into the chair, listening to the music, but his eyes were going around the room... Sometimes he looked in my eyes as if begging: "I know it is you, please stop!", but I kept demonstrating that I don't understand what happens... They won't understand that people want to listen to the music the same they want to speak to around the globe.
About 30 years ago... I remember horrible days when I was a student and there was an amateur station on the 9'th (the top) floor of our building... The roof was covered by sophysticated antennae, it was TIASUR...
We vere very creative and used to invent all possible and impossible ways to protect our tape recorders and turntables applying all knowlege we were getting on lectures and laboratory works, but nothing could help, because during contests they generated such power so electric bulbs in the whole building were dimming in tact with their distorted SSB speach from our Hi-Fi loudspeakers...
Once I've found a genuine solution. If it is impossible to fight against them passively, it may be possible to fight actively deafing them out.
I quickly made an oscillatir with one tunnel-effect diode tuned on 3.5 MHz frequency. It produced lot of harmonics, so all amateur bands (3.5, 7, 14, 21, 28 MHz) were taken care of. I used a varactor diode for frequency modulation, and one 2-base diode for triangle sweep generator. The whole thing including couple of 9-volt batteries was hidden inside of a slipper under the bed, thin wires were going to the metal frame of the bed and to the central water-heater battery.
After I switched on my device they continued traransmitting a little bit, then The Silence Come...
I laid down on my bed and switched on Deep Purple "Sweet Child in Time, You See The Line" on my turntable...
I enjoyed the music when a roommate who was a radio amateur come... He sat down into the chair, listening to the music, but his eyes were going around the room... Sometimes he looked in my eyes as if begging: "I know it is you, please stop!", but I kept demonstrating that I don't understand what happens...
They won't understand that people want to listen to the music the same they want to speak to around the globe. 
Hi, Mike,
The "junk" intrusion to feedback loop that I'm thinking is not until we can hear radio broadcast in the speakers, but the intermodulation is small enough that we can only hear it as "harsh" or "calm/clean" music (when there is no junk intermodulation) on the original music reproduction.
Do you think that the feedback loop system will suppress the junk picked by speaker cables/anything outside amplifier's output binding post? It will make no audible difference? Looking at an amplifier schematic, if there is anything HF picked by speaker cables, even it is small, it should enter the differential feedback system, if there is no output inductor or other LPF guard on amp's output that is directly going to base of inverting input.
The "junk" intrusion to feedback loop that I'm thinking is not until we can hear radio broadcast in the speakers, but the intermodulation is small enough that we can only hear it as "harsh" or "calm/clean" music (when there is no junk intermodulation) on the original music reproduction.
Do you think that the feedback loop system will suppress the junk picked by speaker cables/anything outside amplifier's output binding post? It will make no audible difference? Looking at an amplifier schematic, if there is anything HF picked by speaker cables, even it is small, it should enter the differential feedback system, if there is no output inductor or other LPF guard on amp's output that is directly going to base of inverting input.
Lumanauw,
feedback loop on RF is nothing against base-emitter junctions of output transistors that rectify the RF signal. And already rectified it goes stright to the inverting input of the amp and being amplified intermodulates the same RF signal shifting bias of B-E junctions of output transistors.
"Don't look for the black cat in the dark room if it is not there" (C)
feedback loop on RF is nothing against base-emitter junctions of output transistors that rectify the RF signal. And already rectified it goes stright to the inverting input of the amp and being amplified intermodulates the same RF signal shifting bias of B-E junctions of output transistors.
"Don't look for the black cat in the dark room if it is not there" (C)
Re: Pavel's amplifier
OK, I need to look at it again. As i said, it came out of my printer hard to read, and I thought the R values were off for a true Hawksford EC. But maybe I was wrong.
Jan Didden
Bob Cordell said:I reviewed the schematic of Pavel's amplifier, and I agree with Darkfenriz that his output stage is a near-exact copy of the design I published in HAES in 1984. Even some of the resistor values are unchanged. Too bad he compromized the design by using un-degenerated bipolar differential inputs pairs with old-fashioned Miller compensation rather than just copying the whole design
Cheers,
Bob
OK, I need to look at it again. As i said, it came out of my printer hard to read, and I thought the R values were off for a true Hawksford EC. But maybe I was wrong.
Jan Didden
A lot of sources of RF interference have been metioned here. Personally, the worst I've seen recently is the hash comming from CFL's. I had one in my bench lamp which was pulled low over the amp I was working on and I had some strange ringing on the output waveform - I turned the lamp off and the problem dissapeared. I'd be interested to here other peoples experiences here.
If one worries about RF signal coming from amp-speaker interface, than has to realize, that severe unmatching exists on both terminals of speaker cable. Therefore only standing waves seem to be a significant RF junk in there.
My thought is that a signal-ground resistor equal in value to characteristic impedance of the cable placed in the middle of the cable should damp most of RF signal.
Adam
My thought is that a signal-ground resistor equal in value to characteristic impedance of the cable placed in the middle of the cable should damp most of RF signal.
Adam