Hi Dumpster,
No, I haven't got any figures to support that. I've only looked at Technics marketing spiel, which give figures similar to Quad, around 0.003% THD full power, full frequency range. 0.009% IMD. The fact that Technics use class AB (high biased 45mv), instead of Class C for the Current amp, would suggest that Technics would have a better performance at the lower power scale. Technics made Class AA into the mid 90's. They also used it in the preamp stages for CD players, cassette decks....etc.
You're right, CD is still being used in high end expensive items, such as Quad's Artera, Devialet Phantom. However, Class D seems to be the current leader in the HiFi world.
I've been using a Kenwood Pure Class A amplifier. It is only rated at 8w Pure class A, or, switched to Class AB 15W. THD is around 0.1% at full power. Prior to the Kenwood, I had been using a Quad 306.After a direct comparison the Kenwood upper frequencies had more 'air' and 'transparency', whereas the Quad had a 'strong', 'powerful' low end. Now I'm thinking the Kenwood power is a little too low and there maybe a fault with the Quad. I'm curious about the Quad's circuit, as the results are just as good, if not better than a Class A amp. I don't like it if a design has to use copious amounts of NFB to achieve the right figures. Quad's bridge seems to be very sensitive to component tolerances. My question is, how much difference does the bridge components make to the overall output measurements? Also, is the total Class A output really relevant? Would 500mw Class A output give the same result as 1w output?
No, I haven't got any figures to support that. I've only looked at Technics marketing spiel, which give figures similar to Quad, around 0.003% THD full power, full frequency range. 0.009% IMD. The fact that Technics use class AB (high biased 45mv), instead of Class C for the Current amp, would suggest that Technics would have a better performance at the lower power scale. Technics made Class AA into the mid 90's. They also used it in the preamp stages for CD players, cassette decks....etc.
You're right, CD is still being used in high end expensive items, such as Quad's Artera, Devialet Phantom. However, Class D seems to be the current leader in the HiFi world.
I've been using a Kenwood Pure Class A amplifier. It is only rated at 8w Pure class A, or, switched to Class AB 15W. THD is around 0.1% at full power. Prior to the Kenwood, I had been using a Quad 306.After a direct comparison the Kenwood upper frequencies had more 'air' and 'transparency', whereas the Quad had a 'strong', 'powerful' low end. Now I'm thinking the Kenwood power is a little too low and there maybe a fault with the Quad. I'm curious about the Quad's circuit, as the results are just as good, if not better than a Class A amp. I don't like it if a design has to use copious amounts of NFB to achieve the right figures. Quad's bridge seems to be very sensitive to component tolerances. My question is, how much difference does the bridge components make to the overall output measurements? Also, is the total Class A output really relevant? Would 500mw Class A output give the same result as 1w output?
About 14v/uS for the A21SE. This is feeding a quite highly reactive speaker load (simulated B&W 703).
I refer you to Jan's post 129 and Ed's 130. Jan makes it clear that the bridge needs to be balanced to achieve low distortion figures, whilst Ed points out that the actual slew of the 405 is somewhat better than what Google state.
Plus - Delphi, one can't come onto this forum, despite its name being "diyAudio" and quote figures from Google. You'll get shredded! Ha! Ha!
I can think of parts of the human anatomy that have a faster Slew than what Google quoted.
Plus - Delphi, one can't come onto this forum, despite its name being "diyAudio" and quote figures from Google. You'll get shredded! Ha! Ha!
I can think of parts of the human anatomy that have a faster Slew than what Google quoted.
Technics were known for publishing superbly low distortion claims on their brochures, surfing that wave of THD hype, how?
They made their measurements under cryogenic conditions. Yes they were getting those 'numbers', in Arctic conditions,
but not in summer with a room full of sweaty folk playing loud. Marketing interfering in technical specifications.
Thanks Earths - brilliant. It reminded me of the time I was called into the Company as a company had tendered to supply their Cavity Resonators/Combiner Networks for a National Mobile Telephone Network.
They submitted a prototype network and I was going to do the environmental testing in a large heated test enclosure. My initial tests concerned thermal drift, but the damned enclosure took ages to get up to temperature and so too the combiners.
Well it wasn't too long before the HP Network Analyser started to drift and it was a pig of a job to go into the enclosure to inspect the kit, and hours again before the temperature stabilised. It was winter and all this started with me wrapped up in warm clothing.
However, it just couldn't go on as tests were spilling into days. Suffice to say, several bottles of chilled water at hand, it seemed the easiest way to progress things was to leave the oven on and go in and do some checks and make some adjustments. I must have presented a pretty picture as the heat was such that final testing was done with me nipping in and out of the oven to do some measurements with the analyser, then diving back in having just swigged on a bottle of water. All this in my underpants.
I felt it only fair on the supplier to at least give them an idea why they failed to meet spec. As it turned out, their temperature compensating device was for the wrong range/environment.
Some wag from my department snuck in a caught me with a Polaroid Camera, the pictures doing the rounds for months.
They submitted a prototype network and I was going to do the environmental testing in a large heated test enclosure. My initial tests concerned thermal drift, but the damned enclosure took ages to get up to temperature and so too the combiners.
Well it wasn't too long before the HP Network Analyser started to drift and it was a pig of a job to go into the enclosure to inspect the kit, and hours again before the temperature stabilised. It was winter and all this started with me wrapped up in warm clothing.
However, it just couldn't go on as tests were spilling into days. Suffice to say, several bottles of chilled water at hand, it seemed the easiest way to progress things was to leave the oven on and go in and do some checks and make some adjustments. I must have presented a pretty picture as the heat was such that final testing was done with me nipping in and out of the oven to do some measurements with the analyser, then diving back in having just swigged on a bottle of water. All this in my underpants.
I felt it only fair on the supplier to at least give them an idea why they failed to meet spec. As it turned out, their temperature compensating device was for the wrong range/environment.
Some wag from my department snuck in a caught me with a Polaroid Camera, the pictures doing the rounds for months.
Yes, I know....Google search! 🙃
What is the slew rate for a 405? Or, a 306, 606 & 909, for that matter?
What is the slew rate for a 405? Or, a 306, 606 & 909, for that matter?
Here is comes in around 2.1v/uS over similar conditions to the Sugden. (I have quite a few amps as simulations in case you are wondering 😉)
This one is nicely controlled and this is into the reactive load.
This is the simulated speaker load which really is a tough test:
Correct me if I’m wrong, but R4 in the actual Quad is the inductor, right?
I believe it’s the inductor that’s responsible for the rapid “slewing” seen between the emitters of Q1 and Q2. Recall that inductors try to maintain their instantaneous current and change their voltage in an attempt to do so. (Think about the dynamics in a switching power supply when the switching transistor opens and the inductor swing to the catch diode.)
I’m not negating the analysis cited earlier, just offering a description of what I think may happen in the circuit. The switching transition between the emitters happens so quickly that that the circuit could be argued as continuous, thus justifying linear analysis.
Mooly is the wizard with LTspice. Maybe he can contrive a time aligned simulation of opamp output, Q1,Q2 emitter voltage, and inductor current. That might either confirm or put the lie to my conjecture.
Edit: some hiccup ocurred in my post. I was attempting a reference to Mooly post 105.
Last edited:
@T = zero, instantaneous voltage across an inductor is max, current = zero. We are going somewhere that I can't follow? Beyond what has come out of this exercise, I am happy to leave any in-depth analysis to others. I have neither my 405 or 909 before me, or test facilities to dig deeper. Moolys contribution has brought much greater clarity to this and I am happy to move on.
The finer design details are for those that wish to pursue them and I can understand why. I just want to get back to the music.
It has been fascinating to read everyones contribution - thanks
The finer design details are for those that wish to pursue them and I can understand why. I just want to get back to the music.
It has been fascinating to read everyones contribution - thanks
My understanding was that the resistor is lossy at high powers (efficiency suffers) and to get around that the resistor was replaced by an inductor and to maintain balance in the bridge is where C8 in the original circuit comes in (the 120pF).
In a simulation for example I can only guess at values such as coil resistance.
If only 🙂 I wish I was but its all been hard won. I've only scratched the surface of what LT can do.