|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
71.178.181.244
In Reply to: RE: Updated information posted by pbarach on January 25, 2014 at 05:13:34
I am saying that the phenomenon reflects a problem unique to the Beveridge direct-drive amplifiers. They are unusually prone to oscillation due to the circuit design. In this case, only one of two amplifiers goes into oscillation, the one driving the R channel speaker. I am not saying that there is anything at all wrong with the Ayre or Pioneer units. Indeed, I think the Ayre C5XeMp is the best one-box off-the-shelf CD/SACD player I have ever heard.
It is not easy to work on the amplifiers, because when in use they are built in to the base of the speaker and are totally inaccessible. On the other hand, when you take them out on to the workbench, their behavior is "different", because they are not coupled to the ESL panels inside the Bev speakers. Then too, the output stage develops +/-1600V (3200V from pos to neg), which is a bit scary on the workbench. So it's not easy to trace down the cause, but we are going to try, first by investigating why one amp tolerates the SACD output and the other does not.
If I'd had my druthers, I would have deleted this thread, but once there was a response, I was not allowed to do so. Nevertheless, I welcome bright ideas.
Fortunately for me, I have another more conventional system in which I can implement the Ayre. And it's also possible that using the balanced outputs from the Ayre will ameliorate the problem with the Bevs.
Follow Ups:
From your enhanced description, it seems possible that the noise shaping hump produced by direct DSD-analog conversion is troubling to one of your amps. It is curious that one of your pair is different, however. This ultrasonic noise hump appears in the analog output after conversion from DSD and it is not usually an audible problem in most systems. But, it appears to be destabilizing to one of your amps.
...when I saw yours. The frequency range possible from SACD goes out in the ultrasonic range to ~50kHz. Some amplifier topologies appear to have trouble with ultrasonic frequencies, causing distortion and oscillation. I remember reading a caution when I had the Sony SCD-777 SACD player that one shouldn't use the switch on the back in the "full range" position (or something like that) unless you were sure that your amps behaved nicely with the ultrasonics present in the output. I never had any problems running "full range" with the tube amps I was using during the time I had the SCD-777. This reason also explains why RBCD is fine - brick wall filters take out the ultrasonic info.
At least Sony is willing to admit there is a 'potential' problem. I use an outboard low pass filter and I sleep well.
-reub
I can't solve the problem of the Ayre making the Beveridge amplifier oscillate, can you say where you got the low-pass filter or if you built it yourself, can you supply instructions? Thanks.
My low pass filter is homemade, and the basic schematic is on page 11 of the linked application note. A 'full' schematic can be seen in the gain vs frequency plot on page 15, 'with RC'.
I'm no guru, but after repeatedly beating my head against the wall I came up with a filter that works for me. If you want to try one like it, send me an email for more info.
best regards,
reub
I think I can build that with the info provided. What cut-off frequency did you choose?
50KHz - Enjoy the journey!
I and my local guru pretty much agree with your explanation(s). He also mentioned the VHF noise that comes with DSD. We took pains to make the amps stable prior to setting up the Bev system, and the first thing I need to do is to figure out whether we treated the two amplifiers differently in any way. It's quite possible that that is the case. For example, I built a shield out of "TI Shield" that runs between the input and driver stages of each of the two amplifiers (which is all solid state) and the very high voltage tube output stage. Each of the two amps has the shielding, but maybe one shield is working better than the other, due to better grounding or to slight differences in how the shields were located. Similarly, there are ferrite beads here and there on leads that connect the solid state circuit to the output stage.
In retrospect, I had been experiencing moments when the R channel would sound a bit lower in gain compared to the left, but it is a very momentary phenomenon, and nothing else elicited the white noise. Possibly those fleeting changes in gain characteristic represent an episode of oscillation as well.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: