$10 per board is half of what Olimex would charge, are you sure you are quoting for a double sided PCB?
There is no "standard" Phonoclone board: variants include the RC, the BE, and the 3. The clearest way to refer to exactly which revision you are talking about is to state the filename of the Eagle files: pc28c, pc23k, and pc31j respectively.
/Richard
There is no "standard" Phonoclone board: variants include the RC, the BE, and the 3. The clearest way to refer to exactly which revision you are talking about is to state the filename of the Eagle files: pc28c, pc23k, and pc31j respectively.
/Richard
So the phonoclone 3 lives!!
I stuffed the boards and fired it up tonight. All went uneventfully thank god! Basically its done as per RJMs recommendation. I used 3nF and 768K/110K for the RIAA. R1 is set at 10R which suits my cart and I set R2 at 330R, just at the limit of the 30x of R1. It powered up and apart from the tiniest click when I hit the power switch, you would never know it was on. Absolutely quiet, no hum hiss etc etc.
I am aware that I am using a fairly low value for R2 - and hence low gain. So maybe with higher gain something might happen. If I get a chance I'll stick in a 1K or something for R2 and see do I hear anything. I did orignally have a bit of hum with the previous phonoclone until I brought R2 down to c. 500R so it will be interesting to see how this one does.
I left out C12 and 13 and I used a big wima 4.7 MPK4 for C3. To be honest I am not 100% happy with that choice. I think I'm gonna search for some blackgate Nx4.7 - and I'm nearly certain I have a pair in the VSPS which I don't use anymore. If I can't find any, thats where I'll get some.
Sonics wise - I can see what RJM is talking about (extra detail), but on inital power up it seems a little hard, or maybe etched. After an hour of listening this did ease out a bit, so maybe on extended burn in that will go. It took 3 weeks for the other phonoclones to settle down so it would be unfair to judge this one yet. I'll give it a little time and see how it fares out.
Fran..... off to hunt for some Nx - wish me luck!
I stuffed the boards and fired it up tonight. All went uneventfully thank god! Basically its done as per RJMs recommendation. I used 3nF and 768K/110K for the RIAA. R1 is set at 10R which suits my cart and I set R2 at 330R, just at the limit of the 30x of R1. It powered up and apart from the tiniest click when I hit the power switch, you would never know it was on. Absolutely quiet, no hum hiss etc etc.
I am aware that I am using a fairly low value for R2 - and hence low gain. So maybe with higher gain something might happen. If I get a chance I'll stick in a 1K or something for R2 and see do I hear anything. I did orignally have a bit of hum with the previous phonoclone until I brought R2 down to c. 500R so it will be interesting to see how this one does.
I left out C12 and 13 and I used a big wima 4.7 MPK4 for C3. To be honest I am not 100% happy with that choice. I think I'm gonna search for some blackgate Nx4.7 - and I'm nearly certain I have a pair in the VSPS which I don't use anymore. If I can't find any, thats where I'll get some.
Sonics wise - I can see what RJM is talking about (extra detail), but on inital power up it seems a little hard, or maybe etched. After an hour of listening this did ease out a bit, so maybe on extended burn in that will go. It took 3 weeks for the other phonoclones to settle down so it would be unfair to judge this one yet. I'll give it a little time and see how it fares out.
Fran..... off to hunt for some Nx - wish me luck!
Yeah - they're the 47R on the reverse side, right? Added those, but didn't add C0 yet.
The BG Nx seem to be really gone forever. I really hope that I used a pair in that VSPS!
I also have some GE film caps that I used recently in the output of a DAC and I think they might be ok too - certainly worth a try.
Will have another look at this tomorrow.
Fran
The BG Nx seem to be really gone forever. I really hope that I used a pair in that VSPS!
I also have some GE film caps that I used recently in the output of a DAC and I think they might be ok too - certainly worth a try.
Will have another look at this tomorrow.
Fran
Food for thought:
http://www.edn.com/contents/images/178097.pdf
Mr. Mancini is one of the few people who really seems to know what he's talking about when tackling the trickier issues of regulator stability.
http://www.edn.com/contents/images/178097.pdf
Mr. Mancini is one of the few people who really seems to know what he's talking about when tackling the trickier issues of regulator stability.
Board wiki - anyone wanting any PCB's, even if you have allready emailed me please sign up here.
http://www.diyaudio.com/wiki/index.php?page=phono+clone+3+mini+gb
http://www.diyaudio.com/wiki/index.php?page=phono+clone+3+mini+gb
Ok, vey lucky me.... I pulled the 2 blackgate Nx 4.7uF from my unused VSPS and now have them in the phonoclone.
I also put in a 882R for R2 to look for hum at higher gain levels. There is some hum there now, but only when I have the preamp up to full volume. to be honest I don't need that gain - and my system is fairly low gain throughout. (preamp 15dB, 10WPC power amp)
More listening to follow. I feel now that I'm comparing fairly with the kit phonoclone (ie one of the special 10) - once burn in is complete.
I also added sockets for C3, C12 and C13 so I can easily drop them in.
Please don't take my post last night as negative. I am comparing with a fully burnt in, run in and polished best-you-can-do super-1-of-the-10-kit phonoclones so at first turn on and listen the phonoclone 3 really did quite well.
Its important I think to note that. Can't wait to get back and listen more now!!
Fran
I also put in a 882R for R2 to look for hum at higher gain levels. There is some hum there now, but only when I have the preamp up to full volume. to be honest I don't need that gain - and my system is fairly low gain throughout. (preamp 15dB, 10WPC power amp)
More listening to follow. I feel now that I'm comparing fairly with the kit phonoclone (ie one of the special 10) - once burn in is complete.
I also added sockets for C3, C12 and C13 so I can easily drop them in.
Please don't take my post last night as negative. I am comparing with a fully burnt in, run in and polished best-you-can-do super-1-of-the-10-kit phonoclones so at first turn on and listen the phonoclone 3 really did quite well.
Its important I think to note that. Can't wait to get back and listen more now!!
Fran
pc31k is a "let's suppose" exercise.
Let's suppose that putting the pass transistor inside the op amp feedback loop significantly reduces the stability of the gain block.
Let's suppose that while it does not oscillate, as originally configured there is a lot of ringing on transient loads. The exact level of the ringing will depend on the components, so the circuit becomes hypersensitive to tweaks, break-in etc.
Adding R21,22 would certainly help the situation, but its a relatively minor step. The real solution would be to dramatically kill off the regulator bandwidth using a compensation capacitor between the op amp output and the inverting input.
That's what 31k has provision for.
I don't know by how much the bandwidth must be reduced to completely damp out the ringing, but the uncompensated bandwidth is only 250kHz to start with.
I'm going to mod my boards and give 120 pF a shot, this will drop the bandwidth down to about 40 kHz. Just to see what will happen.
Let's suppose that putting the pass transistor inside the op amp feedback loop significantly reduces the stability of the gain block.
Let's suppose that while it does not oscillate, as originally configured there is a lot of ringing on transient loads. The exact level of the ringing will depend on the components, so the circuit becomes hypersensitive to tweaks, break-in etc.
Adding R21,22 would certainly help the situation, but its a relatively minor step. The real solution would be to dramatically kill off the regulator bandwidth using a compensation capacitor between the op amp output and the inverting input.
That's what 31k has provision for.
I don't know by how much the bandwidth must be reduced to completely damp out the ringing, but the uncompensated bandwidth is only 250kHz to start with.
I'm going to mod my boards and give 120 pF a shot, this will drop the bandwidth down to about 40 kHz. Just to see what will happen.
Trying to please everyone: 31m has both the resistors and the capacitors, for maximum tweakability.
I tried C18,19 120 pF and R21,22 zero (short) this evening. It was a fascinating if esoteric display of how the different parts of the power supply affect the sound.
120 pF considerably lowers the frequency at which the op amp signs off and passes its duties to the bypass caps, compared to simply having 47 ohms in series with the op amp output. Unfortunately I don't have hard numbers for you, but anyway, I suppose as you'd expect, this makes the whole thing sound more like the previous Phonoclone versions.
There is a loss of transparency, a loss of precision, a loss of control ... but at the same time whatever negatives might be ascribed to the regulation circuit (strain, etch, edginess) is gone too. As I told Fran in an email earlier today, it's the sonic equivalent of the sharpness parameter in digital photography. Too much is bad, not enough is bad. Exactly how much is "enough" is largely a matter of taste.
Personally I think that 120 pF essentially threw the baby out with the bathwater, but it's nice to know there is an effective kill switch (that still keeps the regulator working perfectly well). I'm thinking perhaps 33 pF might be better, perhaps leaving the 47 ohm resistors in there as insurance.
31m lets you play around to your hearts content. Rhysh, you can go ahead with this layout as far as I'm concerned, as I'm out of ideas as for other things to try.
[Note : even the least attractive of the Phonoclone 3 "sounds" I've so far experienced is at least as good as any of the previous Phonoclone versions, so don't get all disheartened on me. I'm just being a lot pickier this time around.]
I tried C18,19 120 pF and R21,22 zero (short) this evening. It was a fascinating if esoteric display of how the different parts of the power supply affect the sound.
120 pF considerably lowers the frequency at which the op amp signs off and passes its duties to the bypass caps, compared to simply having 47 ohms in series with the op amp output. Unfortunately I don't have hard numbers for you, but anyway, I suppose as you'd expect, this makes the whole thing sound more like the previous Phonoclone versions.
There is a loss of transparency, a loss of precision, a loss of control ... but at the same time whatever negatives might be ascribed to the regulation circuit (strain, etch, edginess) is gone too. As I told Fran in an email earlier today, it's the sonic equivalent of the sharpness parameter in digital photography. Too much is bad, not enough is bad. Exactly how much is "enough" is largely a matter of taste.
Personally I think that 120 pF essentially threw the baby out with the bathwater, but it's nice to know there is an effective kill switch (that still keeps the regulator working perfectly well). I'm thinking perhaps 33 pF might be better, perhaps leaving the 47 ohm resistors in there as insurance.
31m lets you play around to your hearts content. Rhysh, you can go ahead with this layout as far as I'm concerned, as I'm out of ideas as for other things to try.
[Note : even the least attractive of the Phonoclone 3 "sounds" I've so far experienced is at least as good as any of the previous Phonoclone versions, so don't get all disheartened on me. I'm just being a lot pickier this time around.]
RJM,
What about doing something with the Vref circuit for the opamps? If we could drop the noise/ripple on that it might make a difference. Don't know how exactly you'd make that work on the existing boards, but might be worth a look. I sent you a link the other night to one used elsewhere here. Did you see that?
I might be way off - really I'm only shooting in the dark here. Could also just try a precision low noise reference.
My one has evened out a lot. I put back in C12 and C13 and I think it improved the sound a bit. Other than that it would be very hard to tell the difference between it and the kit phonoclone.
RJM, another thought. I'm using a reasonably low gain set up - using 330R for R2 (10R for R1). I'm guesssing an output of 1V (which would match my DAC) If you are using higher value for R2, does it make the circuit more "twitchy", or at least more transparent to changes.
Whatya think?
Fran
What about doing something with the Vref circuit for the opamps? If we could drop the noise/ripple on that it might make a difference. Don't know how exactly you'd make that work on the existing boards, but might be worth a look. I sent you a link the other night to one used elsewhere here. Did you see that?
I might be way off - really I'm only shooting in the dark here. Could also just try a precision low noise reference.
My one has evened out a lot. I put back in C12 and C13 and I think it improved the sound a bit. Other than that it would be very hard to tell the difference between it and the kit phonoclone.
RJM, another thought. I'm using a reasonably low gain set up - using 330R for R2 (10R for R1). I'm guesssing an output of 1V (which would match my DAC) If you are using higher value for R2, does it make the circuit more "twitchy", or at least more transparent to changes.
Whatya think?
Fran
Thanks for the comments,
My gain is set to 60 dB with the cart I use. I don't think yours is any less than that.
I'll go hunt down that link...
The Vref part of the circuit is almost certainly not the problem, though. The way the vref filter works is higher frequency stuff is attenuated far more than the 120 Hz ripple, which is already more than 80 dB down.
Stability, the subtleties therein, is the "problem" if you want to call it that, but really, it's so subtle it can for the most part be considered a tuning exercise. Stability, the subtleties therein, probably accounts for why a good deal of audio equipment sounds the way it does.
I digress. Time for bed. Today I discovered the relationship between error amp bandwidth and sound is surprisingly reliable. I could hear it as the regulator ceded it's iron grip, replaced by the softer, murkier presentation of the electrolytic bypass caps. Progress was made.
My gain is set to 60 dB with the cart I use. I don't think yours is any less than that.
I'll go hunt down that link...
The Vref part of the circuit is almost certainly not the problem, though. The way the vref filter works is higher frequency stuff is attenuated far more than the 120 Hz ripple, which is already more than 80 dB down.
Stability, the subtleties therein, is the "problem" if you want to call it that, but really, it's so subtle it can for the most part be considered a tuning exercise. Stability, the subtleties therein, probably accounts for why a good deal of audio equipment sounds the way it does.
I digress. Time for bed. Today I discovered the relationship between error amp bandwidth and sound is surprisingly reliable. I could hear it as the regulator ceded it's iron grip, replaced by the softer, murkier presentation of the electrolytic bypass caps. Progress was made.
FWIW, I spent another couple of hours with the P3 tonight and it is and has been changing over the last few days. I spoke earlier of a sense of the vocals being a bit recessed, further back in the soundstage. Well that has changed quite a bit. Its kinda like as if the image started out being from the two speakers, and now the centre image is filling in nicely.
As regards detail, yes, I do think there is a bit more detail there. However some things aren't quite as pronounced yet. Eg on paul simon still crazy after all these years, theres a great ring to the drums on 50 ways. On old phonoclone that rings longer and is much more noticeable whereas right now on this one its not nearly so noticeable. Having said that I noticed other things in the recording that I hadn't noticed before. But as I mentioned above the most noticeable thing is that the vocalist has moved forward quite a bit and is now much more out in front of the band so to speak. The other thing is the bass - it seems much stronger than before, but maybe less distinct.
I have a vague memory of RJM saying a bit about burn in so I went and rooted it out:
That was here: http://www.diyaudio.com/forums/showthread.php?postid=1317390#post1317390
So what I'm hearing is pretty much exactly as predicted. Right now I'm at the day 2-3-4 bit. The extra detail might well be that extended treble and I've already mentioned the bass.
Its just gonna take time I reckon. Right now (unless RJM tells me otherwise) I'm going to leave it as it is for another week or two. the only thing I might do is add C0 on the opamps pins to match what I did with my phonoclone limited edition kit.
So thats my experience for now. Lets see how it settles down.
Fran
As regards detail, yes, I do think there is a bit more detail there. However some things aren't quite as pronounced yet. Eg on paul simon still crazy after all these years, theres a great ring to the drums on 50 ways. On old phonoclone that rings longer and is much more noticeable whereas right now on this one its not nearly so noticeable. Having said that I noticed other things in the recording that I hadn't noticed before. But as I mentioned above the most noticeable thing is that the vocalist has moved forward quite a bit and is now much more out in front of the band so to speak. The other thing is the bass - it seems much stronger than before, but maybe less distinct.
I have a vague memory of RJM saying a bit about burn in so I went and rooted it out:
That was here: http://www.diyaudio.com/forums/showthread.php?postid=1317390#post1317390
So what I'm hearing is pretty much exactly as predicted. Right now I'm at the day 2-3-4 bit. The extra detail might well be that extended treble and I've already mentioned the bass.
Its just gonna take time I reckon. Right now (unless RJM tells me otherwise) I'm going to leave it as it is for another week or two. the only thing I might do is add C0 on the opamps pins to match what I did with my phonoclone limited edition kit.
So thats my experience for now. Lets see how it settles down.
Fran
I broke my Black Gates in for about three days, so allowing I started at day 3-4 of in the schedule I wrote above, things are quite a bit different this time around.
The initial sonics are dominated - and marred - by this weird "skidding" off out of control on transients, more easily recognized as etch especially early on, when it is more or less continuous. It's so significant I could not believe it would be a burn in issue, especially as I'd never heard anything like it on previous builds.
Well, it seems that it is. I've gone back and removed the 120 pF caps and 47 resistors both, and found the sound completely altered from what it was before when I was at the same configuration. Simply put the etch is gone. All that's left is pure awesomeness.
I have a question: I can record up to 24bit/192khz with my Onkyo soundcard. If I recorded a .wav sample of the phonoclone 3 in action, would anyone know where I could temporarily host the fairly large file for people to check out, preferably involving no fees for yours truly?
/R
The initial sonics are dominated - and marred - by this weird "skidding" off out of control on transients, more easily recognized as etch especially early on, when it is more or less continuous. It's so significant I could not believe it would be a burn in issue, especially as I'd never heard anything like it on previous builds.
Well, it seems that it is. I've gone back and removed the 120 pF caps and 47 resistors both, and found the sound completely altered from what it was before when I was at the same configuration. Simply put the etch is gone. All that's left is pure awesomeness.
I have a question: I can record up to 24bit/192khz with my Onkyo soundcard. If I recorded a .wav sample of the phonoclone 3 in action, would anyone know where I could temporarily host the fairly large file for people to check out, preferably involving no fees for yours truly?
/R
Can you sign up with rapidshare or the like?
Only thing is that not everyone will have a decent enough playback chain to really tell. I would be very interested, although what I would need to do is burn the file to a CD and then listen that way.
My digital playback has come in absoute bounds with the shigaclone and the full buffalo DAC-counterpoint setup. I use the Peter Daniels NOS DAC too and it is excellent as well. First time I can say I have a digital chain as good as the vinyl one.
How big a file is it? Could you email it to me?
Fran
Only thing is that not everyone will have a decent enough playback chain to really tell. I would be very interested, although what I would need to do is burn the file to a CD and then listen that way.
My digital playback has come in absoute bounds with the shigaclone and the full buffalo DAC-counterpoint setup. I use the Peter Daniels NOS DAC too and it is excellent as well. First time I can say I have a digital chain as good as the vinyl one.
How big a file is it? Could you email it to me?
Fran
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.