You could do the Devialet thing, empty claims that can't possibly be substantiated. This guy looks to have applied the Dyson vacuum cleaner concept to speakers. You aught to at least challenge the "0" distortion.
Ed I'm surprised, you're in the middle of the Burson Audio add. Their discrete VFA is a drop in for the AD844 CFA really, surprise, surprise for some I to V folks.
Last edited:
It looks like I have two stadiums and a performing arts center booked for next year. So I probably will stick to reality or at least close to it.
But if I ever want to make extraordinary claims I could try more folks listen to my systems than JC's, Nelson's and Charles' combined. Of course any of theirs' sound better, use less power but don't go as loud.
I am however going to finish my power supply article before the end of this year.
But if I ever want to make extraordinary claims I could try more folks listen to my systems than JC's, Nelson's and Charles' combined. Of course any of theirs' sound better, use less power but don't go as loud.
I am however going to finish my power supply article before the end of this year.
Last edited:
More details
Back to out-of-band signals in audio ---- But not RF this time.
Some years ago, i saw a lot of HF idle tones out of a DVD player's audio output. Today I looked at a audio 'mastering' CD machine. It too has an idle tone which only shows up when a signal is being processed.
Now, in this one's level it is low and at 93KHz. But not all are so nice and to preamps and power amps with wide bandwidth, can there be distortion caused by attempts to amplify such out-of-band stuff?
D.Self showed that such freq coming in on the ac line and thru the ps to the amp did raise the measured audio distortion level.... and this CD player output is a direct path into the audio amplification chain. So? Better watch what is coming and going outside our 20KHz hearing range.
Design a fast, low noise and low distortion wide BW amp and roll-off the signal at the I/O's. And, have a very well filtered ac line on each component.
THx-RNMarsh
Back to out-of-band signals in audio ---- But not RF this time.
Some years ago, i saw a lot of HF idle tones out of a DVD player's audio output. Today I looked at a audio 'mastering' CD machine. It too has an idle tone which only shows up when a signal is being processed.
Now, in this one's level it is low and at 93KHz. But not all are so nice and to preamps and power amps with wide bandwidth, can there be distortion caused by attempts to amplify such out-of-band stuff?
D.Self showed that such freq coming in on the ac line and thru the ps to the amp did raise the measured audio distortion level.... and this CD player output is a direct path into the audio amplification chain. So? Better watch what is coming and going outside our 20KHz hearing range.
Design a fast, low noise and low distortion wide BW amp and roll-off the signal at the I/O's. And, have a very well filtered ac line on each component.
THx-RNMarsh
Last edited:
Funny, the peak is almost exactly the 6th (actually 5.97) harmonic of the PAL TV line frequency. Are you sure this is not coming from a multi system DVD player?
I read this article. As I understand, this emc-lab did some measurements with quite high Radio frequencies that showed that Bybees do some damping of these, similar to unexpensive ferrite clamps.
The measuring setup is obscure, lacking precise information.
Last edited:
RNM
I am of the OPINION that one of the advantages of FETs and tubes is that they are much less affected by out of band signals. In my large scale systems rf energy shows up as an increase of base noise level.
ES
I am of the OPINION that one of the advantages of FETs and tubes is that they are much less affected by out of band signals. In my large scale systems rf energy shows up as an increase of base noise level.
ES
Scott
Some years back I did a review of their discrete opamp. It measured quite decently but not as pretty as the monolithic ratings champ (an AD something back then, sure you know about it. 🙂. ). It did however sound better when a modified CD player was compared to a stock unit. (Before and after mod.)
I attribute that to the FET input resisting the noise floating around the circuitry.
At the levels of THD from either chip or discrete I doubt that was the issue.
Some years back I did a review of their discrete opamp. It measured quite decently but not as pretty as the monolithic ratings champ (an AD something back then, sure you know about it. 🙂. ). It did however sound better when a modified CD player was compared to a stock unit. (Before and after mod.)
I attribute that to the FET input resisting the noise floating around the circuitry.
At the levels of THD from either chip or discrete I doubt that was the issue.
Last edited:
But isnt it just good engineering practice to limit the bandwidth to only true frequencies of interest? Why would one want an amp that is still strong out at 90kHz+, other than for marketing numbers? Open the window wider, more garbage comes in.
How do you limit the bandwidth of energy coupled into the circuit? For serious use around transmitters the practice used to be shield, shield and when in doubt shield. A more recent approach is to use devices that are still working at the EMI frequencies and make them part of an active low pass filter.
Some stray energy will always come in. Filtering it out has well known limits. If you restrict bandwidth, where do you do that? Are the sub circuits that still are affected?
Last edited:
RC filtering is about as old tech as there is in electronics, pretty easy to add a pole on the input at 30kHz.
And if the EMI comes in the output, the power supply or even through the case directly? What do you calculate for the maximum attenuation of a simple filter? The resonance of even a simple capacitor?
Zobel? Sensible engineering? (What's that?)
There's plenty of perfectly decent lab equipment that directly behind where my butt is parked on this laptop that has to manage quite a lot of RF noise (and wider bandwidth than audio!). Mountains, molehills and all that.
There's plenty of perfectly decent lab equipment that directly behind where my butt is parked on this laptop that has to manage quite a lot of RF noise (and wider bandwidth than audio!). Mountains, molehills and all that.
You forgot my point: why have bandwidth unmanaged? 90kHz from a cd player to the input is not a problem to manage.
Bandwidth to 100KHz or more is usually not controlled in either preamps or power amps. It is almost impossible to add a filter without adding significantly more complexity, even another stage. Think it through. TRY to be flat to 20KHz, and then many decades down at 90KHz. Show us how to do it, easily and without extra phase shift and distortion.
You mean i can't tell the difference between what equipment I just ran the test on?
There is no PAL in USA/California.
Thx-RNMarsh
Cool, CA.
If you dont measure this stuff and remove it, you might have audible consequences when amplifying it X times by pre and PA. However, it would be better not to use such equipment in the first place. But, you dont know what you are doing unless you look at/for it.
THx-RNMarsh
THx-RNMarsh
Last edited:
No system is truly flat to 20kHz. You have to decide how unflat you can tolerate. Most people can't hear 20kHz anyway. -3dB at 30kHz or 40kHz (or 50kHz for people who are awkward) should suffice. Phase shift is unavoidable - causality demands it. I am not sure why distortion suddenly appeared in the sentence; passive filters need not add non-linear distortion. I suspect the real problem is specmanship: -3dB at 150kHz sounds so much better in ad copy than -3dB at 50kHz, yet the latter is better engineering and both are probably indistinguishable from -3dB at 30kHz.john curl said:
You probably don't need many decades down at 90kHz. A low level tone coming in on one input won't need much attenuation to make it harmless; if necessary you could notch it. RF coming in at other ports is likely to be at much higher frequencies so easier to filter - until you get to microwaves which get harder to filter.
If you use an RC filter on the input that drops the 20,000 Hz by 1 dB what do you think it drops 90,000 Hz? 10dB? Not really effective.
Yes marketing over engineering because people do not understand science but love MAGIC. -3 db at 150khz in many cases may in fact be worse than -3 db at 30khz . Take for example the bias frequency for a tape deck at 160k hz in the 150khz system if un-filtered it using lots of power reproducing stray signal that reduces the power available for desired signal . Where the -3 db 30khz is down with far less power going to the error signal. You do make some fine points.
- Status
- Not open for further replies.