Defo diminishing returns now mate. Once you'd done all those small parts and the discrete output, and made everything run off it's own quality reg that's surely all you can do! And the proof is in the listening - no weak areas. I can still find faults with mine. Maybe due to op-amps and those polyester caps in the output filter. This will soon change though...
Weaknesses: last drop of refinement, last scrapings of detail, absolute naturalness/faithfulness. Only realised after listening to the fully modded one.
Weaknesses: last drop of refinement, last scrapings of detail, absolute naturalness/faithfulness. Only realised after listening to the fully modded one.
awpagan said:
And then you hear another and like the sound of,???????????????
aaaaand "what if" starts again
allan
Ps Fun isn't it?
Oh yes it's fun alright! I used to think CD sounded fuzzy and harsh but with extreme low noise psus etc. that isn't really the case. I'm not sure what's next after this. A few more mods and I'll be very happy for a while. Perhaps an investigation into SACD modding is next...
edit: by investigation I humbly mean copying others' mods from this great forum
Chivvp wrote:
The issue is NOT thet the raw psu and regs aren't up to job, but that this way you are getting return currents distributed in an uncontrolled way between the clock signal wiring and the other functions powered. The extra noise on the clock wiring (and the way this spreads digital noise around ) will ensure you're not getting the best you can from this setup.
Give the clock its own little transformer, or offload all the other functions back to the PSUs they came from. It will help out of all proportion ot the effort involved.
HTH!
I beleive that the reason you're not 'blown away' by teh imporvement you oght to be getting is simply that the raw psu is shared between clock and the other functions.
The issue is NOT thet the raw psu and regs aren't up to job, but that this way you are getting return currents distributed in an uncontrolled way between the clock signal wiring and the other functions powered. The extra noise on the clock wiring (and the way this spreads digital noise around ) will ensure you're not getting the best you can from this setup.
Give the clock its own little transformer, or offload all the other functions back to the PSUs they came from. It will help out of all proportion ot the effort involved.
HTH!
Hi Brent,
I have new psu for cd67 dac, opamps and reciever chip I think its called. In total 6 new supplies. I was wondering if I run a low noise finesse shunt and taking the 11V dc from the input to one of the 5 V regs would be ok? If you think about it, all of the audio chain is isolated from anything bad the clock might introduce. What u think?
martin clark said:
Hi,
Thanks for the advice Martin, I'd assumed the improvement would come from the tx being better than the standard one.
At the weekend I'll try feeding the psu to just the clock, see if that helps.
If it does, I still have the option to split the secondaries on the tx and duplicate the rectificaction / smoothing stages. then I could feed one to clock and one to Dac (max about 300ma each). Would that be ok or would it put me back to where I am now?
Regards
Pete
That would be great. The real issue is to avoid the clock signal and return wiring carrying anything other than clock signal, i.e. NO operating current(s). This can only be done by providing the clock with its own secondary winding, rectifier etc.
It's quite OK to have this dedicated secondary on the same transformer as other supplies - at least its a long, long way down the list of things worth worring about IMO
Hi everyone,
I've been putting together a parts list for Ray's discrete output stage, using Jaap's pcb design. I've been sitting on the pcb for a while, a combination of fear and lack of funds preventing me from building it up.
I've attached my parts list. If you're bored please take a look and tell me if I've got anything wrong. I've chosen top notch parts as I don't see the point of not going all-out on this. The 2x BG FK are for psu bypass, but I'm not sure how beneficial they would be. Another thing I'm not sure about is the inductors: 37DCR seems a touch high! Does this matter?
Unfortunately it's going to get even more expensive when I feed it with some kind of "super" regs.
I might power it with standard 7812/7912 to start with... 2 fewer expensive parts that can pop and/or smoke.
Simon
I've been putting together a parts list for Ray's discrete output stage, using Jaap's pcb design. I've been sitting on the pcb for a while, a combination of fear and lack of funds preventing me from building it up.
I've attached my parts list. If you're bored please take a look and tell me if I've got anything wrong. I've chosen top notch parts as I don't see the point of not going all-out on this. The 2x BG FK are for psu bypass, but I'm not sure how beneficial they would be. Another thing I'm not sure about is the inductors: 37DCR seems a touch high! Does this matter?
Unfortunately it's going to get even more expensive when I feed it with some kind of "super" regs.
I might power it with standard 7812/7912 to start with... 2 fewer expensive parts that can pop and/or smoke.Simon
6h5c said:
Cheers Ray. Is that a better choice? It looks like the leads may be annoyingly short. Lower DCR though, which is surely a good thing. And shielding is nice. But a ferrite core. Us audiophiles have always been told that's bad in speakers, but perhaps for low power it's not a bad thing??
Simon
The leads are shorter, but it's a radial component, so they are exactly long enough to fit on a 5mm pitch. If you're going to mount them on Jaap's PCB it should be fine. DCR is indeed lower (18.8R).
Ferrite doesn't have to be bad in audio, even in speakers. I use them in my Vivace '94. A good ferrite inductor has excellent high frequency properties and low distortion. The core has to be big enough for large current inductors (as in speakers). Mostly that means more expensive, so often cheaper or lesser material is used. But it is very well possible to make a good ferrite inductor for speaker applications.
The best core is an air core of course, but in the bass section it is often a trade-off between low DCR, size, and inductance. A ferrite can be very useful there. The mechanical distortion caused by the woofer is many times bigger than the distortion caused by the inductor's hysteresis. This Panasonic choke can handle 80mA by the way
Regards,
Ray
Ferrite doesn't have to be bad in audio, even in speakers. I use them in my Vivace '94
. A good ferrite inductor has excellent high frequency properties and low distortion. The core has to be big enough for large current inductors (as in speakers). Mostly that means more expensive, so often cheaper or lesser material is used. But it is very well possible to make a good ferrite inductor for speaker applications. The best core is an air core of course, but in the bass section it is often a trade-off between low DCR, size, and inductance. A ferrite can be very useful there. The mechanical distortion caused by the woofer is many times bigger than the distortion caused by the inductor's hysteresis. This Panasonic choke can handle 80mA by the way
Regards,
Ray
6h5c said:But it is very well possible to make a good ferrite inductor for speaker applications.
The best core is an air core of course, but in the bass section it is often a trade-off between low DCR, size, and inductance. A ferrite can be very useful there. The mechanical distortion caused by the woofer is many times bigger than the distortion caused by the inductor's hysteresis. This Panasonic choke can handle 80mA by the way
Regards,
Ray
Thanks for the info Ray, I will go for this Panasonic choke then I think. Hope to order soon.
I'm well aware of the trade-off in speakers as I recently turned my MTMs into 2.5way using a 3.9mH inductor. It has very thick wire and an iron dust core. It's heavy and macho and as part of the whole sounds superb. My other mid/bass coil is about 0.2DCR. How much do you rate those Kiwame resistors? I might try some in my speakers.
I'm worried now about fitting 2w resistors onto Jaap's little pcb. Is this madness??
Simon