Thanks.
So LM4562 is good for U3?
Any suggestions or comments regarding the other parts? I'm especially unsure about the capacitors.
So LM4562 is good for U3?
Any suggestions or comments regarding the other parts? I'm especially unsure about the capacitors.
For the ceramic caps C0G and for the film caps PPS (Panasonic for example) might be a good choice.
For U3 the TL072 is fully appropriate.
For U3 the TL072 is fully appropriate.
I thought it surprising to see such an ancient op-amp recommended for a recent project. I had assumed the TL072 was put out of production long ago. However, in quickly scannng their current datasheet they perform much better than I had recalled, particularly for noise. I've no idea whether the circuit design has been updated over the decades, but the wafer fab process certainly has been, which probably accounts for much if not all of any noise improvement.
Many thanks for your recommendations Boris!
The 100p caps in the BoM already are C0G (NP0), so that's good.
However I'm having difficulties finding the 10u, 220u and 100u caps as PPS. Panasonic only offer them as SMD 0805 if I'm correct, while in Bruno's layout these are cans. (The Panasonics I have in the BoM are Aluminium Electrolytics). The only PPS in that build format I could find are Kemet SMR275226, but these are way too big to fit on the board. Any suggestions...?
What about using ELNAs or Nichicons for these values, would that make sense?
Thank you,
Robert
Last edited:
values of 10uF and above will probably have to be electrolytic
There are some ceramic smd 1206 @ 10uF, but the voltage rating is likely to be low and NP0/C0G will probably not be available, only the high K variety.
As coupling capacitors electrolytics are pretty good. It is filtering duty where they don't perform as well as film types.
There are some ceramic smd 1206 @ 10uF, but the voltage rating is likely to be low and NP0/C0G will probably not be available, only the high K variety.
As coupling capacitors electrolytics are pretty good. It is filtering duty where they don't perform as well as film types.
Hypex resort to using 22uF 63V electrolytics as decoupling caps for LM4562 in their modules, so if they're good enough for Hypex's reference design, they're good enough for me 🙂
As Hans so aptly put it: 'This preamp refuses to distort';-)
BTW, this weekend I was on a presentation of another one of Bruno's creation: the fully active Kii speaker. Can you imagine Joachim, a speaker being directional down to 40Hz?? Goodby room modes!
Ahh the power of DSP...
Jan
how does being directional lower affect the room modes? Reflections yes but modes are surely still modes?
Robert, for the 10μ caps in the servo circuit you can use electrolytics. There are unipolar ones shown in the schematic, but you could also use polarized ones, I believe.
The 100 and 220μ caps have to be electrolytics.
The 100 and 220μ caps have to be electrolytics.
Yes but when you are more directional you throw much less energy into the room which means the room modes are much less excited.
Jan
Ah ok. Not entirely sure that follows as you will just turn it up higher if it remains clean 🙂
I hope the speakers are as good as the tech suggests. Previous attempts to do this sort of thing have tended to end up compromised, but I know Bruno wouldn't do that.
I hope the speakers are as good as the tech suggests. Previous attempts to do this sort of thing have tended to end up compromised, but I know Bruno wouldn't do that.
Thank you Boris.
As 'ELNA fanboy' I have reworked this capacitor part of the BOM and chose ELNA RFS for 22u and 1000u as well as ELNA RBD for 10u. Hope everybody is ok with this? Here is the final BoM: http://1drv.ms/1W0T7dR
Will now get final quotes to set up the Group Buy.
Last edited:
Regarding the relays for the input switching, are there any preferred ones that are more immune to noise than others? It's one of the few parts without a specific part number.
I haven't searched in detail nor elsewhere on the forums for suggestions.
I haven't searched in detail nor elsewhere on the forums for suggestions.
choose a signal relay for signal duty.
The pin outs vary so you need to check that your chosen relay fits the PCB.
If you rigorously use twisted pairs for all lines inside the box and use a screened/shielded box there is not much left to interfere with the loops in the relays.
The pin outs vary so you need to check that your chosen relay fits the PCB.
If you rigorously use twisted pairs for all lines inside the box and use a screened/shielded box there is not much left to interfere with the loops in the relays.
I am not going to comment on the sound quality, what with my reputation for demanding double blind listening and knowing Bruno personally - no way I could be objective.
Yet, at the first demo at the first evening of the show for 16 pro-audio and studio people, 9 sets were sold before the evening was up. Next evening where I was present with some 12 consumer/audiophile types, 3 were sold during the demo. And these are not el cheapo's. Despite myself, I was impressed.
Jan
Signal relays like these neither pick up noise easily, nor generate any. I have added the TE Connectivity MT2-C93402 to the BoM, we should be good with this.
Jan, the DSP in the Kii is very much like what we did in Essex in the early 90th.
FIR for frequency response and IIR for phase. I think Bruno even uses 24dB Linkwitz in the phase linear variant.
We found that steeper filters have pre echos and othe artifact, so BOOM, here goes your 100dB/oct filter.
I know ways to make electronic cardioids too.
In fact i am preparing a pattent with an american partner.
Anyway, i really admire Brunos productivity.
FIR for frequency response and IIR for phase. I think Bruno even uses 24dB Linkwitz in the phase linear variant.
We found that steeper filters have pre echos and othe artifact, so BOOM, here goes your 100dB/oct filter.
I know ways to make electronic cardioids too.
In fact i am preparing a pattent with an american partner.
Anyway, i really admire Brunos productivity.
Yes he is almost as prolific as you 😉
Maybe with class D prices dropping and performance improving, CAD technologies and cheap DSP, this may slowly become a trend, while with you guys in the 90-ies (I guess that was with Malcolm) it was still something outlandish definitely not mainstream. You were maybe too early!
Jan
Last edited: