Any reason to not replace D1 with a buried zener precision voltage reference? It might lower the noise on the front end.
Well???????
I'm still waiting to hear/see pictures of a finished product.....
Has anyone actually scaled this thing up and lived with it long enough to give us an idea about its sound?
What snags did you run into when scaling it up?
Peter, how's yours coming??????????????
Thanks for any info!!!
I'm still waiting to hear/see pictures of a finished product.....
Has anyone actually scaled this thing up and lived with it long enough to give us an idea about its sound?
What snags did you run into when scaling it up?
Peter, how's yours coming??????????????
Thanks for any info!!!
Getting close on mine.
I just went with the 15 volt rails.
and made the PCB with the latest posted layout
with the suggested small grounding change.
Thank you all!!
I needed some verification on setting the DC offset,
Grey posted to set it between the speaker output and ground.
Is that "out+ and out-" or "out+ and gnd"??
I just went with the 15 volt rails.
and made the PCB with the latest posted layout
with the suggested small grounding change.
Thank you all!!
I needed some verification on setting the DC offset,
Grey posted to set it between the speaker output and ground.
Is that "out+ and out-" or "out+ and gnd"??
I think Grey was talking about the absolute DC voltage between either + speaker and ground, or - speaker and ground, which is set by adjusting the current source on the front end, and should be as close to 0V as is possible. The only offset a speaker would see is mismatch between sides. The reason you set the offset relative to ground is to keep as much voltage swing as possible. Anyone, feel free to correct me if I'm wrong (I know you all will).
NS
NS
I can't go through finishing the input board seemingly. Somehow the process of working on that amp is more enjoyable than having it done.😉 Maybe this weekend.
The DC offset should be checked both between +/- output and ground (absolute offset) and what is actually more mportant between + and - output ( because this is what your speaker see).
The DC offset should be checked both between +/- output and ground (absolute offset) and what is actually more mportant between + and - output ( because this is what your speaker see).
Active cooling
I was polishing off the design for mine yesterday, and I settled on active cooling. The heatsink I have on hand is an aluminum extrusion 13"x7.5"x1.3" with 21 fins, very similar to the R-Theta 66279. With passive cooling I cannot even do 25Wx2. The T<sub>j</sub> rises to 125°C. With passive cooling I can do 40Wx2 and the junction temperature is barely remarkable.
My design choices are +/-12V rails and 1A bias current through two devices per side per channel, for a total of 16 transistors and approximately 40W into 8Ω speakers. I'm using IRFP150 FETs for their superior R<sub>θjc</sub>
A Pabst 8412N[G]L fan can deliver 33m<sup>3</sup>/h at only 12dB(A), so I don't think fan noise will be a problem.
I'm very much looking forward to building it. Strangely this Aleph-X amplifier is less expensive than any of my previous projects, but stands to make the largest overall improvement in my listening enjoyment.
I was polishing off the design for mine yesterday, and I settled on active cooling. The heatsink I have on hand is an aluminum extrusion 13"x7.5"x1.3" with 21 fins, very similar to the R-Theta 66279. With passive cooling I cannot even do 25Wx2. The T<sub>j</sub> rises to 125°C. With passive cooling I can do 40Wx2 and the junction temperature is barely remarkable.
My design choices are +/-12V rails and 1A bias current through two devices per side per channel, for a total of 16 transistors and approximately 40W into 8Ω speakers. I'm using IRFP150 FETs for their superior R<sub>θjc</sub>
A Pabst 8412N[G]L fan can deliver 33m<sup>3</sup>/h at only 12dB(A), so I don't think fan noise will be a problem.
I'm very much looking forward to building it. Strangely this Aleph-X amplifier is less expensive than any of my previous projects, but stands to make the largest overall improvement in my listening enjoyment.
nobody special said:
Now when you mentioned it... I was planning on increasing the current through the input differencial to about 30mA and effectively I choose 300ohm drain resistors (for 5V drop). If I'm doing that, will I have a trouble with obtaining 0V absolute offset? I didn't expect that changing the front end current changes offset. If it does, how it works then?
As a matter of interest I have been doing some simulations on this design and changing the first stage current does effect the offset by quite a bit. However I have come up with a design change which allows 30ma of current in the first stage and with a 10mv offset across the loudspeaker.
By using active DC servo amplifiers on each side of the differential stage, this replaces the normal 4k7 resistor that Nelson uses and it also removes the need for the 33 to 100 Ohm resistors to ground. Although it does add quite a bit more component count to the design. However once this is incorporated into a PCB layout it shouldn't add to much more work to the construction.
I'm guessing some of you might be asking "Well how is this going to affect the sound quality?" Well I have used this technic in my own SE Class A design, with no loss of quality that I can tell...
know doubt I will hear what all of you think...
Cheers
Anthony Holton
🙂
By using active DC servo amplifiers on each side of the differential stage, this replaces the normal 4k7 resistor that Nelson uses and it also removes the need for the 33 to 100 Ohm resistors to ground. Although it does add quite a bit more component count to the design. However once this is incorporated into a PCB layout it shouldn't add to much more work to the construction.
I'm guessing some of you might be asking "Well how is this going to affect the sound quality?" Well I have used this technic in my own SE Class A design, with no loss of quality that I can tell...
know doubt I will hear what all of you think...
Cheers
Anthony Holton
🙂
The Saint said:
So how does it affect the offset and what has the major inluence? Current alone or the resistance which adjusts it?
Hi Peter
Current adjustment from the current the source has the most influence on offset from what I have seen from the sims.
and to a lesser degree the load resistors in the diff stage.
Also in the original design 220k was use in the Global feedback.
100k seem to work alot better.
Cheers
Anthony
Current adjustment from the current the source has the most influence on offset from what I have seen from the sims.
and to a lesser degree the load resistors in the diff stage.
Also in the original design 220k was use in the Global feedback.
100k seem to work alot better.
Cheers
Anthony
Voltage Offset
I'm not sure which version of the circuit you are referring to, since Grey's original seems to use 100k feedback resistors. However, I thought it might be worth reporting my findings with respect to what influences DC offset most.
I've removed R11, V1 and R33, V3 entirely (numbers taken from Grey's circuit) from the Aleph current sources and increased the values of R25 and R40 to compensate. This makes the Aleph current source much more stable with DC voltage variation at the output. With this change, the sensitivity of the DC offset adjustment is considerably reduced without having to resort to a servo.
Ian.
I'm not sure which version of the circuit you are referring to, since Grey's original seems to use 100k feedback resistors. However, I thought it might be worth reporting my findings with respect to what influences DC offset most.
I've removed R11, V1 and R33, V3 entirely (numbers taken from Grey's circuit) from the Aleph current sources and increased the values of R25 and R40 to compensate. This makes the Aleph current source much more stable with DC voltage variation at the output. With this change, the sensitivity of the DC offset adjustment is considerably reduced without having to resort to a servo.
Ian.
Ian
I think you are looking at a different ccT to mine. the designators you are referring to don't match up to mine.
From what I have seen of Greys ccT he has always used 220k for the Global feedback, perhaps you are referring to the 100k resistors which are in the input to ground.
What function do these resistors play in the circuit you are using?
Anthony
I think you are looking at a different ccT to mine. the designators you are referring to don't match up to mine.
From what I have seen of Greys ccT he has always used 220k for the Global feedback, perhaps you are referring to the 100k resistors which are in the input to ground.
What function do these resistors play in the circuit you are using?
Anthony
There is only one circuit posted by Grey and it's here:
http://www.diyaudio.com/forums/attachment.php?s=&postid=32776
http://www.diyaudio.com/forums/attachment.php?s=&postid=32776
Thanks Peter
This is indeed the circuit to which I was referring.
As aside, I note that Nelson has also dropped the aforementioned resistors from the Aleph current source as used in the recent Zen variations.
Ian.
This is indeed the circuit to which I was referring.
As aside, I note that Nelson has also dropped the aforementioned resistors from the Aleph current source as used in the recent Zen variations.
Ian.
I was considering this circuit for current source. Do you think it would bring improvement? More description here: http://www.diyaudio.com/forums/showthread.php?s=&postid=70176#post70176
Attachments
Alternative current source
Improvement in what exactly? Sound quality, lower drift? Harry's current source is indeed better than the simpler one used by Grey, but the question to my mind is whether this will result in an audible improvement. Personally I doubt that it will, as the circuit does not seem particularly sensitive to the type of current source used in my experience. As regards lower drift, I suspect that this will be improved but I guess that Harry is in a better position to comment on this aspect.
Ian.
Improvement in what exactly? Sound quality, lower drift? Harry's current source is indeed better than the simpler one used by Grey, but the question to my mind is whether this will result in an audible improvement. Personally I doubt that it will, as the circuit does not seem particularly sensitive to the type of current source used in my experience. As regards lower drift, I suspect that this will be improved but I guess that Harry is in a better position to comment on this aspect.
Ian.
Peter,
The circuit works very well. The use of the current diode improves rail rejection by 20db according to Mr.Pass (see A40 article). I use this circuit quite a lot, there are a few ways to improve it but not by much.
Jam
P.S. By the way, how goes the construction of your amplifier?
The circuit works very well. The use of the current diode improves rail rejection by 20db according to Mr.Pass (see A40 article). I use this circuit quite a lot, there are a few ways to improve it but not by much.
Jam
P.S. By the way, how goes the construction of your amplifier?
Ian,
Did you read that: http://www.diyaudio.com/forums/showthread.php?s=&postid=72205#post72205
When I'm talking about improvements, I always mean sound quality.
Jam,
It's going slow now, but sooner or later it will be done. I was distracted by other projects. Maybe Mr. Pass should post more pics?😉
Did you read that: http://www.diyaudio.com/forums/showthread.php?s=&postid=72205#post72205
When I'm talking about improvements, I always mean sound quality.
Jam,
It's going slow now, but sooner or later it will be done. I was distracted by other projects. Maybe Mr. Pass should post more pics?😉
regulated PS
Hi all,
I have everything I need to start the buildup, except the power supply. I was think about a simple regulator, like the one used in Z4. Anyone has tried with your Aleph-X? Does it bring any benefit? or I should stick with CRC filters only.
Thanks,
niner
Hi all,
I have everything I need to start the buildup, except the power supply. I was think about a simple regulator, like the one used in Z4. Anyone has tried with your Aleph-X? Does it bring any benefit? or I should stick with CRC filters only.
Thanks,
niner