OK I freely admit I don't have a grasp of the design issues involved. I just assumed it would work for my 80 Ohm headphones, plus putting the standard 50 Ohm output resistor, as long as the output driver can handle the current. On second thought at least the output stage bias would have to change. So far I get it. As for the influence on the open loop gain I am at a loss. Nevermind.
As for IC op amps, I tried OPA2134 as a headphone driver, it is clearly not enough. (I tried power IC as well, OPA548, less apparent detail than OPA2134+BUF634 but what a punch - and very natural sounding)
One thing that keeps baffling me is that even for preamp-out-to-poweramp-in use, IC op amp performance doesn't follow the spec sheet, so to speak. I had DRV134 outputs working into INA134 inputs as of TI standard configuration, worked just fine. Then I tried a pseudo balanced output with BUF634 - OPA2134 combo, and the dynamics and detail improved quite a lot. Why is that so? By the peak voltages and currents involved, the standard configuration should have had no problem whatsoever, more like it should have about 20dB headroom.
And - what kind of dynamics performance can we then expect from the Mo(p)amp?
As for IC op amps, I tried OPA2134 as a headphone driver, it is clearly not enough. (I tried power IC as well, OPA548, less apparent detail than OPA2134+BUF634 but what a punch - and very natural sounding)
One thing that keeps baffling me is that even for preamp-out-to-poweramp-in use, IC op amp performance doesn't follow the spec sheet, so to speak. I had DRV134 outputs working into INA134 inputs as of TI standard configuration, worked just fine. Then I tried a pseudo balanced output with BUF634 - OPA2134 combo, and the dynamics and detail improved quite a lot. Why is that so? By the peak voltages and currents involved, the standard configuration should have had no problem whatsoever, more like it should have about 20dB headroom.
And - what kind of dynamics performance can we then expect from the Mo(p)amp?
Well, I think we are entering the territory where sound quality is more than just a numbers game. BUF634+OPA2134 is probably better in terms of sound quality than DRV134+INA137, despite the much higher CMRR of the latter pair.
Nelson has said you are doing pretty well to get 20dB of real noise rejection. It is probably worth it to sacrifice 60dB of theoretical CMRR for superior sound quality.
By the same token, a diff. reciever or transmitter based on moamps' opamp will likely have poor measured performance with respect to the TI parts. (more distortion, much lower CMRR even with precision discrete resistors) It will probably sound a whole lot better too.
The main reason I brought up DRV134 in the group buy thread is that it is easy to use. For my stereo system, I won't let one get close. (maybe surround channels, ok) I'm currently using discrete BJT pairs to perform these functions, but I expect this design to be superior, mainly due to the superior matching of the SK389 dual JFET.
Nelson has said you are doing pretty well to get 20dB of real noise rejection. It is probably worth it to sacrifice 60dB of theoretical CMRR for superior sound quality.
By the same token, a diff. reciever or transmitter based on moamps' opamp will likely have poor measured performance with respect to the TI parts. (more distortion, much lower CMRR even with precision discrete resistors) It will probably sound a whole lot better too.
The main reason I brought up DRV134 in the group buy thread is that it is easy to use. For my stereo system, I won't let one get close.
(maybe surround channels, ok) I'm currently using discrete BJT pairs to perform these functions, but I expect this design to be superior, mainly due to the superior matching of the SK389 dual JFET.Tiroth,
you can simply use a pseudobalanced out with the mopamps. That saved 2 of the 3 op amps needed for a standard balanced line driver. That's what I do with the BUF634+OPA2134: Same output Z on the ground line as on the signal line, run into fully differential input. Noise cancels the same, just no 6 dB gain because the (-) line is ground. Theoretically CMRR should be the same. Me at least I didn't have any increase in noise or hum after switching from DRV134 to the pseudobalanced. The sound got better however.
Actually the funny thing is that I went to great lengths to rebuild my system to "industry standard" balanced (mainly bec. my active X-O is part of the speaker, with monoblock amps - so I chose to have long signal lines rather than long speaker cables). And it went from "no big hum or noise" (unbal) to "really no big hum or noise" (bal). So that either means I got it right either way, bal or unbal, or that CMRR is not that much of an issue in a typical home audio situation, all else being well implemented (grounding etc). PSRR I believe is more of a potential real world issue when switching from IC's to minimalist discretes. The lower PSRR of my Aleph Minis comapred to chip amps is noticeable. Then again, barely so, at 1" off the speaker at night when it's quiet.
I used DRV134 etc, because, I wanted to try the textbook application first, before the funny experiments.
BUF634+OPA2134 on headphones: extremely detailed, doesn't seem to run out of power, and yet, and yet... ultimately, not very "nice" to listen to. Maybe I should experiment with that ear canal resonance canceling notch filter that Siegfried Linkwitz recommends for headphones.
you can simply use a pseudobalanced out with the mopamps. That saved 2 of the 3 op amps needed for a standard balanced line driver. That's what I do with the BUF634+OPA2134: Same output Z on the ground line as on the signal line, run into fully differential input. Noise cancels the same, just no 6 dB gain because the (-) line is ground. Theoretically CMRR should be the same. Me at least I didn't have any increase in noise or hum after switching from DRV134 to the pseudobalanced. The sound got better however.
Actually the funny thing is that I went to great lengths to rebuild my system to "industry standard" balanced (mainly bec. my active X-O is part of the speaker, with monoblock amps - so I chose to have long signal lines rather than long speaker cables). And it went from "no big hum or noise" (unbal) to "really no big hum or noise" (bal). So that either means I got it right either way, bal or unbal, or that CMRR is not that much of an issue in a typical home audio situation, all else being well implemented (grounding etc). PSRR I believe is more of a potential real world issue when switching from IC's to minimalist discretes. The lower PSRR of my Aleph Minis comapred to chip amps is noticeable. Then again, barely so, at 1" off the speaker at night when it's quiet.
I used DRV134 etc, because, I wanted to try the textbook application first, before the funny experiments.
BUF634+OPA2134 on headphones: extremely detailed, doesn't seem to run out of power, and yet, and yet... ultimately, not very "nice" to listen to. Maybe I should experiment with that ear canal resonance canceling notch filter that Siegfried Linkwitz recommends for headphones.
Nelson,
Earlier in this thread Fred Dieckmann made a reference to the similarity to the XVR1, although I was not aware from his comment that the circuitry was literally identical. I have always given attribution to moamps because I thought this was correct.
I think I remember seeing a comment from you in another thread about cancelling your project due to the similarity of a design by either Gray or Moamps. If you canceled it because you were upset, then I guess I missed that fact--I thought you were just saying that the need for a design had already been met within the DIY community.
As the person that organized the MOX buy, it makes me feel bad that I may have inadvertantly encroached on your design without your permission. I still have a small number of boards available, but I will remove them from my site immediately if you wish.
Please feel free to email me if you wish to discuss offline.
Earlier in this thread Fred Dieckmann made a reference to the similarity to the XVR1, although I was not aware from his comment that the circuitry was literally identical. I have always given attribution to moamps because I thought this was correct.
I think I remember seeing a comment from you in another thread about cancelling your project due to the similarity of a design by either Gray or Moamps. If you canceled it because you were upset, then I guess I missed that fact--I thought you were just saying that the need for a design had already been met within the DIY community.
As the person that organized the MOX buy, it makes me feel bad that I may have inadvertantly encroached on your design without your permission. I still have a small number of boards available, but I will remove them from my site immediately if you wish.
Please feel free to email me if you wish to discuss offline.
I am not upset, and you are right - the project was cancelled
because I saw no reason to offer a duplicate.
I have not had an issue with Grey, although it would certainly
be the case that I did not have to concern myself with the
Aleph X for DIYers since he did a fine job of putting it out there,
and discussed it with me first, which I always appreciate.
I have not talked to Moamps, but I think that unless it is an
incredible coincidence, it would have been appropriate for him
to attribute the source.
Not putting schematics in owner's manuals gives me some control
over the release of information, allowing me to proceed on what
I perceive as an orderly path and avoiding wasted effort, and
let's face it - I do like to get credit for my own work. (Even if it's
not such an original piece)
because I saw no reason to offer a duplicate.
I have not had an issue with Grey, although it would certainly
be the case that I did not have to concern myself with the
Aleph X for DIYers since he did a fine job of putting it out there,
and discussed it with me first, which I always appreciate.
I have not talked to Moamps, but I think that unless it is an
incredible coincidence, it would have been appropriate for him
to attribute the source.
Not putting schematics in owner's manuals gives me some control
over the release of information, allowing me to proceed on what
I perceive as an orderly path and avoiding wasted effort, and
let's face it - I do like to get credit for my own work. (Even if it's
not such an original piece)
Hi Nelson:
So now you get around to claiming that Mox is IDENTICAL to XVR1. I do not think so.
If I remember correctly, a year or so ago you DARED people participating in the below thread to reverse-engineer your XVR1 based on its general schematic (with no values).
http://www.diyaudio.com/forums/showthread.php?s=&postid=275062&highlight=#post275062
Post#14
A few of us were competing at the time. When I came out with 'my' design Fred Dieckmann commented it wasn't even the same topology as XVR1.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=23522&perpage=10&highlight=&pagenumber=7
Post#66
Meanwhile, I naively expected you to confirm or deny that what I published was in fact XVR1 .… but you never did. Then I got sort of upset
http://www.diyaudio.com/forums/showthread.php?s=&threadid=23522&perpage=10&highlight=&pagenumber=8
Post#71
and received a private mail from Ian (macka) telling me to calm down because you couldn't publicly comment on XVR1 design for a number of reasons (Would you like me to send you Macka's mail too?). So I decided to forget everything about that little game of yours and let go.
As it happened, Mox started a life of its own. I published the complete project, together with the schematic, values and PCB, for the DIYaudio community to use and enjoy and did not get or expect a dime out of it.
Now, you aren't accusing me of ripping you off and not having given you any credit, are you? Again, please re-read the post#71 referred to a couple of paragraphs earlier.
Also, if you think MOX is your IP feel free to contact me and I'll give you all my personal details so you can file a lawsuit against me for breaching your IP rights. I'm not kidding.
Regards,
Milan
So now you get around to claiming that Mox is IDENTICAL to XVR1. I do not think so.
If I remember correctly, a year or so ago you DARED people participating in the below thread to reverse-engineer your XVR1 based on its general schematic (with no values).
http://www.diyaudio.com/forums/showthread.php?s=&postid=275062&highlight=#post275062
Post#14
A few of us were competing at the time. When I came out with 'my' design Fred Dieckmann commented it wasn't even the same topology as XVR1.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=23522&perpage=10&highlight=&pagenumber=7
Post#66
Meanwhile, I naively expected you to confirm or deny that what I published was in fact XVR1 .… but you never did. Then I got sort of upset
http://www.diyaudio.com/forums/showthread.php?s=&threadid=23522&perpage=10&highlight=&pagenumber=8
Post#71
and received a private mail from Ian (macka) telling me to calm down because you couldn't publicly comment on XVR1 design for a number of reasons (Would you like me to send you Macka's mail too?). So I decided to forget everything about that little game of yours and let go.
As it happened, Mox started a life of its own. I published the complete project, together with the schematic, values and PCB, for the DIYaudio community to use and enjoy and did not get or expect a dime out of it.
Now, you aren't accusing me of ripping you off and not having given you any credit, are you? Again, please re-read the post#71 referred to a couple of paragraphs earlier.
Also, if you think MOX is your IP feel free to contact me and I'll give you all my personal details so you can file a lawsuit against me for breaching your IP rights. I'm not kidding.
Regards,
Milan
Anyone who wants can read the threads, the XVR1 manual,
any other material and decide for themselves. I stopped reading
the posts after the first few pages, so I didn't get to post #71,
and yes, on page 8 you belatedly attribute the manual, so I
stand corrected on that point.
So I will re-state my position:
1) I am not upset about the MOX thread, nor would I accuse
Moamps of anything greater than a lapse in good taste.
2) I did discontinue my own crossover project based on that
thread, but since the MOX is virtually identical, I saw no point
to going forward with it.
3) I have reverted to an earlier policy about publishing
schematics before their time.
4) You (the DIY community) are perfectly welcome to reverse
engineer anything I build or tell you about, but it is proper to
label prior art as such, separating it from your own contributions.
any other material and decide for themselves. I stopped reading
the posts after the first few pages, so I didn't get to post #71,
and yes, on page 8 you belatedly attribute the manual, so I
stand corrected on that point.
So I will re-state my position:
1) I am not upset about the MOX thread, nor would I accuse
Moamps of anything greater than a lapse in good taste.
2) I did discontinue my own crossover project based on that
thread, but since the MOX is virtually identical, I saw no point
to going forward with it.
3) I have reverted to an earlier policy about publishing
schematics before their time.
4) You (the DIY community) are perfectly welcome to reverse
engineer anything I build or tell you about, but it is proper to
label prior art as such, separating it from your own contributions.
Nelson Pass said:
Mr. Pass:
Now, why would you want to write things like that? Not only that you did get to post#71 at the time but even replied in post#77 on the same page. Also, I did not post my schematics until page 6 in that thread so I do not see how page 8 could be belated as compared to page 6?? Please, don't do that.
Also, I do not understand what prompted your belated reaction as MOX is water under the bridge. MOX is not particularly original either as it is based on a generic design for discrete opamps and SK-type crossovers that can be easily found anywhere, not just in the XVR1 manual. May I just remind you of your own words in the MOX thread and then we can let everybody be their own judge, indeed:
"Fred, I always appreciate your looking out for my interests
and the circuit looks a lot like an XVR1, but there is hardly
any ground-breaking stuff there. Also, anyone who wants to
implement Sallen-Key filters and their variations doesn't have
to talk to me.
Can I do better? Guess I'll have to....."
Regards,
Milan
I propose to send my Mox "original design" with a Passlabs DIY stick on the front panel to Mr Pass for his verification.
No doubt this will be a good reason for a tasting session and may even throw a bottle of Aussie red in for good measure.
I thank both Mr Pass and Milan for their generous support in making this project possible. Without either, nothing would have happened and the prior crossover wars are hardly worth mentioning so lets not go there again .
Ian
No doubt this will be a good reason for a tasting session and may even throw a bottle of Aussie red in for good measure.
I thank both Mr Pass and Milan for their generous support in making this project possible. Without either, nothing would have happened and the prior crossover wars are hardly worth mentioning so lets not go there again .
Ian
Attachments
moamps said:
Well, my post on page 8 gives you a clue:
"If you post over in the PassLabs forum, you have a better
chance of getting a response, as I only check here sporadically."
I'll tell you what: I officially declare you right, and I'm wrong.
Ian, thanks very much. Just send the wine....
Hmmm...interesting turn of events.
In the Mini-A and Aleph-X threads, I believe I referenced Nelson's relevant patents and stated that all concepts were his intellectual property.
After much yik-yak from certain members here, I started a crossover thread which I called the Xenover. I didn't really have time that point in my life to 'maintain' such a thread, and crossover threads are pretty much hit-or-miss affairs, anyway, but there it is. The Xenover had certain topological similarities to the XVR1, but also certain omissions. In particular, I treated the crossover design as an exercise in Sallen-Key topology and did not put in a discrete opamp at all--at least not at first. I did somewhat later put in a "speculative" discrete opamp in which I purposefully used a slightly wonky resistor value, just to see if anyone was paying attention. They weren't; not one person commented on it, as I recall. I think I put a few references to Nelson's commercial product in that thread, too, although there aren't any patents involved that I'm aware of.
Somewhere back in the mists of time, I had a thread on the Aleph 2. At this time it's a bit vague in my mind, but I'm pretty sure I put the patent information in there, too.
For the record, the only thread that Nelson and I formally coordinated was the Aleph-X, and even that one had an element of surprise for me in that he had asked that I remain quiet for a while longer (I was glad to do so), but then he sprung the surprise earlier than I anticipated. Not a problem, really--just thought he and I were going to sit on it a bit longer.
I've been accused of a great many personality failures and weaknesses over the years, but I try to remember to give credit to Nelson, John Curl, Walt Jung, et. al. when I find myself on a path that they have blazed. Should I have neglected to do so at some point, I hope any offended party will accept my apologies.
Grey
In the Mini-A and Aleph-X threads, I believe I referenced Nelson's relevant patents and stated that all concepts were his intellectual property.
After much yik-yak from certain members here, I started a crossover thread which I called the Xenover. I didn't really have time that point in my life to 'maintain' such a thread, and crossover threads are pretty much hit-or-miss affairs, anyway, but there it is. The Xenover had certain topological similarities to the XVR1, but also certain omissions. In particular, I treated the crossover design as an exercise in Sallen-Key topology and did not put in a discrete opamp at all--at least not at first. I did somewhat later put in a "speculative" discrete opamp in which I purposefully used a slightly wonky resistor value, just to see if anyone was paying attention. They weren't; not one person commented on it, as I recall. I think I put a few references to Nelson's commercial product in that thread, too, although there aren't any patents involved that I'm aware of.
Somewhere back in the mists of time, I had a thread on the Aleph 2. At this time it's a bit vague in my mind, but I'm pretty sure I put the patent information in there, too.
For the record, the only thread that Nelson and I formally coordinated was the Aleph-X, and even that one had an element of surprise for me in that he had asked that I remain quiet for a while longer (I was glad to do so), but then he sprung the surprise earlier than I anticipated. Not a problem, really--just thought he and I were going to sit on it a bit longer.
I've been accused of a great many personality failures and weaknesses over the years, but I try to remember to give credit to Nelson, John Curl, Walt Jung, et. al. when I find myself on a path that they have blazed. Should I have neglected to do so at some point, I hope any offended party will accept my apologies.
Grey
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.