I found a replacement for the AD797 too but in it´s day it was a visionary design and you still see it in High End Audio Equipment like the new MBL DA converter. That Accuphase uses NE5534 in their expensive CD players is a mystery to me though. Still they get phantastic reviews here and people that buy it are happy with the sound.
I realy whould love to hear the Constellation one day. I talked to the guys on the High End show. I think the guy actually was the ex. Audio Alchemy owner and guess what: the booth was totally empty for hours. I then told Holger Barske he should look there so they got at least some recocnition. I think the stuff is tremendoesly build and looks really cool if a bit overblown for my european taste. Although i studied Furniture Design i personally do not care much about the shape that things have if they do the job, but what a phantastic team of electronics designers you have ! This is something to watch in the future.
I realy whould love to hear the Constellation one day. I talked to the guys on the High End show. I think the guy actually was the ex. Audio Alchemy owner and guess what: the booth was totally empty for hours. I then told Holger Barske he should look there so they got at least some recocnition. I think the stuff is tremendoesly build and looks really cool if a bit overblown for my european taste. Although i studied Furniture Design i personally do not care much about the shape that things have if they do the job, but what a phantastic team of electronics designers you have ! This is something to watch in the future.
John:
I'll make a measurement with milliHertz resolution and really good sources. I just need a test device (or enough details and an opamp) to get it together. I would love to have you observe the process and review it to be sure its not inaccurate. And another observer if you think its necessary. After Thanksgiving?
Such is high end, taste of reviewers and happy owners. Some, on the other hand, do not like Accuphase sound. That is the reason why it is more easy to compare measurements rather than sound impressions, in case that people are several thousand miles distant and cannot make a listenning test to compare their tastes and impressions
.To me, the reviews have zero predicative value.
John, you catched my interest here. I had a nice small talk with Peter Madrick at the Munich High End show. Can you describe some hints of the Constellation phono stage?
Another point I'm not sure if it's practical, or better will be accepted by the audiophiles, is to implement the gain of the line stage into the phono stage. You have then the same signal voltage like a digital source and you'll need only a line stage with gain 0dB, in short a buffer.
Any comments here?
Unless you are in marketing
jan didden
That's it, but I hoped we were trying to find scientific proof and evidence. To me, those old images are not the proof. I have no problem with audible differences perceived and described by listeners, but I do have problems with pseudo-scientific attempts and hypothesis to explain such differences. Because such explanations sound just like marketing, rather than proof.
Tough nuts, PMA. Redo the test yourself, if you are so inclined. All you have to do, that we DIDN'T do at the time, was measure and record EXACTLY what frequencies are in the mix, especially the 4 major UNKNOWN frequencies. Then, PROVE that they have an EXACT mathematical correlation to some REASONABLE combination of the fundamental frequencies. 2f2-27f1 for example, will NOT CUT IT.
As far as old material, I just love the Bell System journal that was linked here. I am looking at magnetic materials in articles written in the 1930's. Just because it is 'old' doesn't make it less readable or interesting.
Now, why do I continually return to the same graph? Because it is there, for virtually all to see, AND it appears to show a problem that is easily explained with FM sidebands being mixed with AM sidebands that we were testing for.
There may be 100's of similar examples out there, old and new, but nobody appears to be looking at the DETAILS, just the overall distortion.
I also PERSONALLY worked with the equipment making the tests and did many, many, maybe 100's of almost identical tests myself, when I was with the VTT Lab in Finland.
HP even then, made good test equipment, that had to be accurate, and calibrated. This is not the world of Heathkit and Eico. This is a government research lab.
People's prejudice against old measurements is unwarranted, and IF you actually worked with the exact test equipment used, you would know this. This is fundamentally a prejudice against ANALOG vs the DIGITAL TESTING promoted by 'marketing', just like 'perfect sound forever' was to vinyl records compared to CD.
As far as old material, I just love the Bell System journal that was linked here. I am looking at magnetic materials in articles written in the 1930's. Just because it is 'old' doesn't make it less readable or interesting.
Now, why do I continually return to the same graph? Because it is there, for virtually all to see, AND it appears to show a problem that is easily explained with FM sidebands being mixed with AM sidebands that we were testing for.
There may be 100's of similar examples out there, old and new, but nobody appears to be looking at the DETAILS, just the overall distortion.
I also PERSONALLY worked with the equipment making the tests and did many, many, maybe 100's of almost identical tests myself, when I was with the VTT Lab in Finland.
HP even then, made good test equipment, that had to be accurate, and calibrated. This is not the world of Heathkit and Eico. This is a government research lab.
People's prejudice against old measurements is unwarranted, and IF you actually worked with the exact test equipment used, you would know this. This is fundamentally a prejudice against ANALOG vs the DIGITAL TESTING promoted by 'marketing', just like 'perfect sound forever' was to vinyl records compared to CD.
>
Another point I'm not sure if it's practical, or better will be accepted by the audiophiles, is to implement the gain of the line stage into the phono stage. You have then the same signal voltage like a digital source and you'll need only a line stage with gain 0dB, in short a buffer.
<
Unless you need some gain for other sources.
What we did was have the selector switch on the output,
rather than on input of the line amp.
Another point I'm not sure if it's practical, or better will be accepted by the audiophiles, is to implement the gain of the line stage into the phono stage. You have then the same signal voltage like a digital source and you'll need only a line stage with gain 0dB, in short a buffer.
<
Unless you need some gain for other sources.
What we did was have the selector switch on the output,
rather than on input of the line amp.
Scott,
I found an interesting issue with the ADA4898-1. I thought it was THE OPAMP. Yes, it is very, very good.
BUT - when it gets a kind of input impulse (short time wave) overload, the output saturates near to +V supply rail. I could have lived with that, and filter the input signal. But same happened when shock came thru power supply, like contactor turn-off transient. That's a big problem. If this did not happen, I would like the opamp very much.
Regards,
I found an interesting issue with the ADA4898-1. I thought it was THE OPAMP. Yes, it is very, very good.
BUT - when it gets a kind of input impulse (short time wave) overload, the output saturates near to +V supply rail. I could have lived with that, and filter the input signal. But same happened when shock came thru power supply, like contactor turn-off transient. That's a big problem. If this did not happen, I would like the opamp very much.
Regards,
For everyone to understand this, I do not see an EXOTIC combination of IM frequencies to be useful in explaining an unknown frequency. This is because, at some point, resolution will suffer, just because of the test equipment limitations, and if you take virtually EVERY combination of frequencies, perhaps 100 different pairs or triples, you will probably find a close fit, perhaps even to 1 Hz or so.
This is TRYING to fit the data into a KNOWN reality, e.g. IM distortion. This is done to attempt to eliminate the existence of FM distortion, and sometimes one goes too far.
It reminds me of the example of the 'planet' tracking us, behind the Sun, in order to 'explain away' Einstein's relativity, back 100 years ago. It gave essentially the same results with Mercury, but you must admit that it was a bit far fetched. Same here.
This is TRYING to fit the data into a KNOWN reality, e.g. IM distortion. This is done to attempt to eliminate the existence of FM distortion, and sometimes one goes too far.
It reminds me of the example of the 'planet' tracking us, behind the Sun, in order to 'explain away' Einstein's relativity, back 100 years ago. It gave essentially the same results with Mercury, but you must admit that it was a bit far fetched. Same here.
John, I guess you really don't get it. The frequencies that FM are at are the same ones that IM are at, THERE IS NO OTHER POSSIBILITY. There is no mechanism to create or destroy the extra cycles. No, this reminds me more of crop circles and UFO's.
PMA does it stay stuck? If so that is bad and I will gladly contact the designer. I have used them in a hot plug-in application and have had no latchups.
Yes, unfortunately. Close to +V rail, not exactly +Voutmax. I have to switch off the PSU to cure.
In case you do not want do discuss this here, please send me an e-mail through PM. Just e-mail, I do not have direct PM option allowed. I could provide the designer with application details. Vdiff transient problem?
Very nice circuit, indeed. Just this trouble.
- Status
- Not open for further replies.