Jokes apart: thank you, Marco, you know I value much your opinion.
But If I may choose, I think I would rather be a Janitor. And even at that, I could only be a Vice-Janitor, given that this is Dario's thread and achievement, principally.
I'm serious about this Janitorship: while I had always enjoyed enormously the beautiful,democratic and creative atmosphere in this thread (due to all of you, and the righteous governance of Dario..), still
technical projects like this, where things can explode and take out paper cones... need some less democratic and more technocratic maintenance, appunto, a Janitor
Ciao, George --- alias F. Joseph K
(hmmm.. does it not look like an F word??)
George,
If we ( me for sure ) are reached this point, is thanks to Dario and his FE.
His continue input gave me the spark for continuing believe in this project!
Ciao
Marco
I'm serious about this Janitorship: while I had always enjoyed enormously the beautiful,democratic and creative atmosphere in this thread (due to all of you, and the righteous governance of Dario..), still
technical projects like this, where things can explode and take out paper cones... need some less democratic and more technocratic maintenance, appunto, a Janitor
I think the great strength of this amplifier is that it has received attention of a number of different people with different methodologies. The result is it has outstanding technical specifications as well as much care in the selection of parts. Many amplifiers are optimized in just one of these areas.
Thanks, interesting reading. Great to learn new things!
How true do you think it is to say that using opamps such as the ADA4627 and OPA827, we are as close to perfection as we can get with this topology? Where are the remaining weaknesses?
Wov, Dan.. I think this is a question appropriate for mr. Caldwell himself..
But I can point out little things: not all parameters of these amps are perfect; In fact what regards bandwidth & precision, there are already today opamps that are more performant (but not FET input). Noise vise bipolar opamps are the king for low impedance applications; then there are opamps with monster output drive capability (like OPA1622 - also with enhanced stability for capacitive loads)
Also integrated circuits today are going in a direction of miniaturization and power economy; this is not properly the direction most desired for us poor audio-piles.
Even if.. even if the new model of the muse opamp series had just been released, so we are good for the next week..)
But really, I would gently prod You to ask this same thing from Johnc124. Would be very curious of his response..
But I can point out little things: not all parameters of these amps are perfect; In fact what regards bandwidth & precision, there are already today opamps that are more performant (but not FET input). Noise vise bipolar opamps are the king for low impedance applications; then there are opamps with monster output drive capability (like OPA1622 - also with enhanced stability for capacitive loads)
Also integrated circuits today are going in a direction of miniaturization and power economy; this is not properly the direction most desired for us poor audio-piles.
Even if.. even if the new model of the muse opamp series had just been released, so we are good for the next week..
)But really, I would gently prod You to ask this same thing from Johnc124. Would be very curious of his response..
the only "weakness" that I can think of is perhaps the max output power and, even more so, the max current. This is a relatively serious limitation for the FE, while for the "Full-Evo" with double pump is much less of a problem.
Looking at it that way, I don't disagree, but it's only a weakness for low sensitivity speakers and in the lowest two octaves. Most speakers 20 cm (8") or less are fine with the FE power output. Even bigger speakers are fine as long as they have moderate sensitivity.
Still, it is a weakness and the double pump is a good answer.
Jac
The highest current demands of a crossovered two (or more) speaker come when very fast changing transients occur.
During these fast stopping and/or fast starting moments the current demand can exceed three times that of a resistive load equal to the rated impedance.
Because it's fast transients that cause this increased current demand, it follows that it is not the lowest two octaves that are our problem.
Maximum current demand is probably in the top three octaves from 4kHz upwards.
Maximum sustained power is probably the concern in the bottom two octaves.
Thank you very much George, that's all i wanted to know. I'll try to cope with DC
My words were shallow...
I switched my 1.5 boards to Evo Rev. A configuration and replaced LM318 using ADA 4627 opamps. I followed George's instruction exactly and removed C9s.
I put a 10k attenuator just after inputs. C3s are ok and running.
When i switch my amplifier on, relais do click and everything seems ok. But if i measure DC offset on the outputs i find very high values, around 200 mV and slightly higher, slowly increasing in time. Before the mods, with LM318s and C9s, i had something around 10 mV. I tried removing attenuator but no change at all.
I tried to drive my speakers anyway - using great caution - and i had no problem, amplifier works great and SQ has improved at least one level, especially using it as an integrated amp! Very happy about that!
But should i worry about the high DC offset on outputs? What could i do to lower it to normal levels? Did i do something wrong?
My words were shallow...
I switched my 1.5 boards to Evo Rev. A configuration and replaced LM318 using ADA 4627 opamps. I followed George's instruction exactly and removed C9s.
I put a 10k attenuator just after inputs. C3s are ok and running.
When i switch my amplifier on, relais do click and everything seems ok. But if i measure DC offset on the outputs i find very high values, around 200 mV and slightly higher, slowly increasing in time. Before the mods, with LM318s and C9s, i had something around 10 mV. I tried removing attenuator but no change at all.
I tried to drive my speakers anyway - using great caution - and i had no problem, amplifier works great and SQ has improved at least one level, especially using it as an integrated amp! Very happy about that!
But should i worry about the high DC offset on outputs? What could i do to lower it to normal levels? Did i do something wrong?
The 200 mV is enough to be concerned. Trying putting C9 back and if this lowers offset.
Mmmmhhh... that's exactly what i would like to avoid...
There's no other way to lower offset? What if i lower R10 value?
Last edited:
Luca, may I ask You: did You remove ! and shorted out C9, I hope?
Is the high offset happening in both channels?
The fact of drifting tells like there is no reference for the DC.. or something similar.
Anyway, it can not be left like this: something is wrong in the setup.
The offset cannot, must not be more than a couple of mV, like 2-4mV max. I regularly see below, closer to zero!
So we have to find out where is the problem.
Photos would help a lot!
Ciao, George
Is the high offset happening in both channels?
The fact of drifting tells like there is no reference for the DC.. or something similar.
Anyway, it can not be left like this: something is wrong in the setup.
The offset cannot, must not be more than a couple of mV, like 2-4mV max. I regularly see below, closer to zero!
So we have to find out where is the problem.
Photos would help a lot!
Ciao, George
Yes ,i firstshorted C9 leaving it in place; later i removed it, but obviously there was no change. Sorry i can't understand what do you mean about removing "!"...
The high offset is on both channels, but slightly higher (around 10% higher) on the right channel. What i noticed is that offset starts below 200mV but then slowly rises in time, continously, very slowly...
I will post some pics as soon as i can get my hands on it.