I don't feel that setup/adjustment is a particular problem, if the right documentation is available, with a clear, well laid out and marked up board. A well written construction handbook, proceeding in a logical manner through the process would certainly ease the dificulties. In fact, as I can't help much with the theory side, I will volunteer to write such a manual if the project gets off the ground.
pinkmouse said:
I already wrote up a way of setting up the unit elsewhere in this thread, and I must insist that it is not easy. If you want to make it relatively simple, you need to unsolder components etc.
My personal next version may well end up with a scheme for maintaining correct bias in input stage, especially if it is Zen'ed.
MBK said:
I like what you said about average levels and peak levels, it sounds spot on to me. Unfortunately, my speakers aren't biampable or triampable, and have a nominal impedance of 4 ohms. I'd like an amp that can run up to about 20 watts class A able to hit peaks of 500 to 600 watts class AB. I think this should be doable without too big of heat sinks.
Update on isolatin washers and grease
In an earlier post, I wrote about isolating washers.
http://www.diyaudio.com/forums/showthread.php?postid=841
This now seems to be superceeded. I no longer prefer the Bergquist isolators (even though they are probably still very good) - I have since converted to Aluminum Oxide washers as they are now quite common. To mount, I now use heatsink grease intended for computer industry. I don't know if it is any better, but I find it relatively easy to apply with a knife edge to these hard units. The grease I use is called Arctic Silver Ceramique and is non-conductive (http://www.arcticsilver.com/ceramique.htm)
Since I don't produce a lot of units, the cost of the grease is largely irrelevant, and to say the truth I must admit that other "fancy" greases are painful to employ. Too much like clay to get it "knifed" of "squeezed" into a thin layer so it probably does more harm than good.
Your mileage may vary.
Petter
(I never was able to employ BeO units due what was claimed to be European import restrictions, but that is probably just as well)
In an earlier post, I wrote about isolating washers.
http://www.diyaudio.com/forums/showthread.php?postid=841
This now seems to be superceeded. I no longer prefer the Bergquist isolators (even though they are probably still very good) - I have since converted to Aluminum Oxide washers as they are now quite common. To mount, I now use heatsink grease intended for computer industry. I don't know if it is any better, but I find it relatively easy to apply with a knife edge to these hard units. The grease I use is called Arctic Silver Ceramique and is non-conductive (http://www.arcticsilver.com/ceramique.htm)
Since I don't produce a lot of units, the cost of the grease is largely irrelevant, and to say the truth I must admit that other "fancy" greases are painful to employ. Too much like clay to get it "knifed" of "squeezed" into a thin layer so it probably does more harm than good.
Your mileage may vary.
Petter
(I never was able to employ BeO units due what was claimed to be European import restrictions, but that is probably just as well)
Depends on what You consider as not too big.....
For that much of max. power You`d need lots of supply voltage.
Although 20W Class-A appears little, the required bias to achieve it, together with this high voltage supply to get the max. power You aim for, it would mean quite some dissipated heat.
I think what You describe pretty much would equal a Pass X-350.......
I was thinking something more along the lines of an X-250.cocolino said:
Depends on what You consider as not too big.....
For that much of max. power You`d need lots of supply voltage.
Although 20W Class-A appears little, the required bias to achieve it, together with this high voltage supply to get the max. power You aim for, it would mean quite some dissipated heat.
I think what You describe pretty much would equal a Pass X-350.......
GregD said:
I am with you. I have been searching for a solution like this for a while. I am starting Aleph 1.2 now, but would be quite happy to jump on the Class A/AB with max 500 - 600 W in AB. I like cappability of easy scaling power up and down based on personal needs, such as Aleph amps - add or deduct amount of Mosfets and rail voltage and pick your power.
Was there ever any cross-fertilization between the "X-100 backengineered", and Uli's design here (schematic and PCB's existing, and has been built as well)? As I understand the idea is the same, but the circuit is quite radically different, esp. the front-end:
Uli's "normal X-amp"
MBK
Uli's "normal X-amp"
MBK
X100 is a scalable power amplifier reverse engineered (not through printed circuit board, but using techniques such as looking at photographs etc etc) with the Pass Labs X600-1000 as motivation/goal. I only needed 100W so that was the aim and hence the name, but the principle is the same - the original design is very scalable - with suitable selection of components you can get 1000W or more as Nelson Pass originally did.
A couple of people built something on this including myself, but most others wandered off to other threads, AlephX seemed as a major path most folks took.
This thread has limited activity, but there is still a lot of insight from smart DIY Audio members and thus a good place to learn.
The smaller X amps are not designed the same way, be that positive or negative.
This thread was to my knowledge the first time that NP came out on DIY Audio. Now he is responsible for motivating users in a separate forum on this site. Kudos!
Petter
A couple of people built something on this including myself, but most others wandered off to other threads, AlephX seemed as a major path most folks took.
This thread has limited activity, but there is still a lot of insight from smart DIY Audio members and thus a good place to learn.
The smaller X amps are not designed the same way, be that positive or negative.
This thread was to my knowledge the first time that NP came out on DIY Audio. Now he is responsible for motivating users in a separate forum on this site. Kudos!
Petter
Thanks Petter. The schematic is very interesting: there is no feedback from the output stage, only from the cascode stage output. Do you use any feedback from the output stage?
The cascode gain stage should have lower distortion than the differential stage used in Aleph X. I wonder why it is not used in Aleph X.
The cascode gain stage should have lower distortion than the differential stage used in Aleph X. I wonder why it is not used in Aleph X.
ctong said:
I don't use feedback from the output. The original patent discusses this topic as well. I believe NP likes to have low output impedance as a function of a low output impedance output stage rather than fixing it with global feedback. Global feedback can also be a problem since speakers typically send stuff back to the amplifier as they are rarely purely resistive devices.
Petter
Petter,
I read that you set a high bias to make the amp class A/B. Do you need a dynamic bias to achieve that? From the schematic it appears that the bias is fixed.
Do you experience a DC offset at the output terminals since there is no feedback to correct it?
I also wnder whether you have had a chance to measure the performance of your x amp.
Thanks.
I read that you set a high bias to make the amp class A/B. Do you need a dynamic bias to achieve that? From the schematic it appears that the bias is fixed.
Do you experience a DC offset at the output terminals since there is no feedback to correct it?
I also wnder whether you have had a chance to measure the performance of your x amp.
Thanks.
Peter, do you have the Service manual for the X 600 and X 1000? If oyu did, it should be quite easy to plug the schematic into a circuit board design software and there you have your PCB no???
Is it that difficult to get this SErvice manual??? Wouldn't that answer an aweful lot of questions??
C
Is it that difficult to get this SErvice manual??? Wouldn't that answer an aweful lot of questions??
C
CT: fixed bias. No issues with offset - this needs to be tuned out anyway - but it not all that easy to do (I have given instructions for how I do it). I have not had the chance to do any serious measurements - no balanced measurement equipment available at the time. My unit is now in pieces, I have aquired a worthy cabinet so the plan is to complete a better variant in that cabinet - where I incidentally employ switched mode power supplies as well. There is another project standing in the way for now - a new DAC I am fiddling with.
Clarkcr: I don't think there are any major secrets anymore, but copying wholesale would help. The biasing for example relies on constant current drive to be a constant voltage. How valid that assumption is, I don't know.
Petter
Clarkcr: I don't think there are any major secrets anymore, but copying wholesale would help. The biasing for example relies on constant current drive to be a constant voltage. How valid that assumption is, I don't know.
Petter