Well, not sure if somebody may be interested in such thing...
About a year ago, when I just got my Audiolab 8000x7 power amp and connected it to Onkio digital processor (having a good hardware media-pleer as a main source), I had significant switching-type-of... well, not even "clicks"... explosive bangs from time to time while switching something at the processor or starting playback at the player (Audiolab works starting from DC).
So I decided to build a high-quality subsonic filter to address this issue. 5-th order Chebyshev was designed, Burr-Brown OPA627 were selected as active elements.
Here goes the sad part of the story 😛 3 OpAs per channel * 7 channels = 21 OpA.
Local store gave me the price of $30 per OpA. Well, what could I do? Right, order at ebay... but then I would have to wait. I went for a faster route 🙂
The result is simply outstanding. No more "Booms", no even "clicks".
The filter if fully transparent above 18Hz, it virtually has "no THD" - I can's see it, I can's hear it, I can't measure it. There's only noise sobewhere at the level of -120db. No compromise device 😉
Well, for 2 channels the cost is much lower... only 2/7-th from my design 😀
I have attached some pictures. The graphs are taken from the device itself - no modelling.
I still use it and love it. Will also use it with PowerAmp-pio I am building now.
If somebody is interested in more info - I will publish the schematic and well-designed PCBs. Nothing special - devil is in details 🙂
Have fun,
Valery
About a year ago, when I just got my Audiolab 8000x7 power amp and connected it to Onkio digital processor (having a good hardware media-pleer as a main source), I had significant switching-type-of... well, not even "clicks"... explosive bangs from time to time while switching something at the processor or starting playback at the player (Audiolab works starting from DC).
So I decided to build a high-quality subsonic filter to address this issue. 5-th order Chebyshev was designed, Burr-Brown OPA627 were selected as active elements.
Here goes the sad part of the story 😛 3 OpAs per channel * 7 channels = 21 OpA.
Local store gave me the price of $30 per OpA. Well, what could I do? Right, order at ebay... but then I would have to wait. I went for a faster route 🙂
The result is simply outstanding. No more "Booms", no even "clicks".
The filter if fully transparent above 18Hz, it virtually has "no THD" - I can's see it, I can's hear it, I can't measure it. There's only noise sobewhere at the level of -120db. No compromise device 😉
Well, for 2 channels the cost is much lower... only 2/7-th from my design 😀
I have attached some pictures. The graphs are taken from the device itself - no modelling.
I still use it and love it. Will also use it with PowerAmp-pio I am building now.
If somebody is interested in more info - I will publish the schematic and well-designed PCBs. Nothing special - devil is in details 🙂
Have fun,
Valery
Attachments
Last edited:
Your design looks good and neat however I can't but help think it's a little pointless in your situation? I would say the processor is faulty or not designed correctly - it should disconnect the output if it's probable DC transients will occur when switching something?
And $30 for an opamp!?
And $30 for an opamp!?
Well, Onkyo TX-NR709 has got a great processing and DAC parts, and no capacitors at the line output, which I actually like. However, in some cases of internal re-configuration (like switching from 2-channel to 5-channel mode for example), some DC offset comes to the output in some channels. You don't notice it if your power amp has got some limit at the lower end of the frequency range, like 10-20 Hz, but really notice if it works starting from DC. That is actually why many "old school" power amplifiers had a built-in subsonic filter that you could engage/disengage. I have checked the Onkyo schematic - it has got transistor keys, shortening the output during certain operations, but DAC output may have some offset, which they ground and then release again - that is when the clicks may be audible. By the way, there are no clicks at all when the internal power amps are used, but Audiolab is definitely better as a power amp.
The software is actually a software part of a Velleman PCSU200 PC Oscilloscope & Signal Genetator, also having a hardware part, connected to the PC via USB cable.
And yes, $30 for an OpAmp... well... for a good one 🙂
The software is actually a software part of a Velleman PCSU200 PC Oscilloscope & Signal Genetator, also having a hardware part, connected to the PC via USB cable.
And yes, $30 for an OpAmp... well... for a good one 🙂
Why wouldn't series single capacitors that gave you say 1Hz response have worked instead ? It would have saved you a LOT of $ & time etc 😉
Can you do 5-th order (see attached) with unity gain without active elements (like OpAmps)? 😉
The task is not only arranging AC coupling, the task is removing all unwanted signals below 20Hz.
Attachments
Last edited:
@ vzaichenko
Well of course it won't do a 5th 😀 But there shouldn't be any issues to deal with in the first place, & as Boscoe mentioned, there must be something wrong upstream. I would rather find the cause of that & try to cure it, instead the band aid approach 😉 No offense intended !
Well of course it won't do a 5th 😀 But there shouldn't be any issues to deal with in the first place, & as Boscoe mentioned, there must be something wrong upstream. I would rather find the cause of that & try to cure it, instead the band aid approach 😉 No offense intended !
I'll try to explain 🙂 There's nothing wrong upstream. This is just a specialty of particilar setup. The source is a media player, digitally connected to DSP via HDMI interface. The player actually acts as a transport, sending a bit-stream to DSP. DSP, and DAC behind it, are actually a front part of Onkyo receiver. I take the signals from the line outputs of the receiver.
When receiver changes decoding mode, it connects / disconnects / reconnects some of its 7 outputs. Like, when the player is in "stop" mode, receiver senses it as Stereo, having only 2 channels connected. You start playing a 5-channel DTS material - receiver switches to 5-channel configuration, connecting additional 3 channels. This is when some step signal appears at the line output.
With some AC-coupled amplifiers it is not so noticeable (though it is there). Audiolab and some of my DIY prototypes are true DC amplifiers. Being directly connected to the receiver, they produce substantial "click" at the output. This is not a problem of particular Onkyo - I tried the other one from the same product line and one of top Pioneer receivers. One way or the other, they all "click" when changing the channels configuration. Apart from that, Onkyo's sonic performance is brilliant, so I don't want to change it to something else so far.
With my low-cut filter those clicks are just not there.
I just forgot they ever existed.
It's like a sort of "sound conditioner" for the receiver.
For sure, some cheaper OpAmps could be used. But this was an exclusive 1-instance project for myself, plus I wanted to see what 627s sound like for that price 😉
When receiver changes decoding mode, it connects / disconnects / reconnects some of its 7 outputs. Like, when the player is in "stop" mode, receiver senses it as Stereo, having only 2 channels connected. You start playing a 5-channel DTS material - receiver switches to 5-channel configuration, connecting additional 3 channels. This is when some step signal appears at the line output.
With some AC-coupled amplifiers it is not so noticeable (though it is there). Audiolab and some of my DIY prototypes are true DC amplifiers. Being directly connected to the receiver, they produce substantial "click" at the output. This is not a problem of particular Onkyo - I tried the other one from the same product line and one of top Pioneer receivers. One way or the other, they all "click" when changing the channels configuration. Apart from that, Onkyo's sonic performance is brilliant, so I don't want to change it to something else so far.
With my low-cut filter those clicks are just not there.
I just forgot they ever existed.
It's like a sort of "sound conditioner" for the receiver.
For sure, some cheaper OpAmps could be used. But this was an exclusive 1-instance project for myself, plus I wanted to see what 627s sound like for that price 😉
Hi, well whatever it is/was, you've dealt with it. But it shouldn't have been happening in the first place. That's why i thought you might want to actually discover Exactly what the problem was, rather than paste over it, so to speak 😉
Why choose a topology with a stop band (Bode plot, image#5 of post#1)? Why not let the very low frequency response continue to fall at -100dB/decade?
It would be great if it would be possible 🙂
When developing active filters, generally a normalised circuit is considered which produces an “ideal” frequency response having a rectangular shape, and a transition between the pass band and the stop band that has an abrupt or very steep roll-off slope. However, these ideal responses are not possible in the real world so we use approximations to give us the best frequency response possible for the type of filter we are trying to design (see attached - the picture shows low-pass, for low-cut just mirror it).
Here is a good article on how it works: Boston University - Active Filter Notes
As usual, you have to choose the right compromise between the pass band / stop band ripple and stop band attenuation.
Bode plot at post #1 is a real analyzer screenshot, showing the live performance of my filter.
Attachments
- Status
- Not open for further replies.