"I didn't have a 7805, so I had to just make do with one of those audiocom super regs I have lying around." LOL
Glad its made such a big difference.
If you've upgraded to a TDA1541A, you might want to check you've got a small capacitor (eg 470pF) across pins 16 and 17 (its called Cosc in the datasheet). Those pins are NC on the non A DAC, and I've heard some Philips PCBs have the cap in place to allow for easy upgrade, but its worth checking. If you have that missing, you may well have distortion when the the signal crosses over 0V.
Glad its made such a big difference.
If you've upgraded to a TDA1541A, you might want to check you've got a small capacitor (eg 470pF) across pins 16 and 17 (its called Cosc in the datasheet). Those pins are NC on the non A DAC, and I've heard some Philips PCBs have the cap in place to allow for easy upgrade, but its worth checking. If you have that missing, you may well have distortion when the the signal crosses over 0V.
I had to re-read my post as, for a moment, I thought you misquoted me 
I will check for that cap and get one ordered up if it's missing. Thanks Phil.
I must agree about the 1541A being a considerable improvement over the non-A version. It would be interesting to try a crown later if the player proves to be a keeper.
Simon
I will check for that cap and get one ordered up if it's missing. Thanks Phil.
I must agree about the 1541A being a considerable improvement over the non-A version. It would be interesting to try a crown later if the player proves to be a keeper.
Simon
Martin's Naim is a (I think) a single crown, and it sounded awesome compared to mine, but I'm sure a lot of it was also far better regulation.
One day, I will probably invest in a crown. Thing is, there a bit pricey at best.
One day, I will probably invest in a crown. Thing is, there a bit pricey at best.
I can't see it offering as much difference as a nice op-amp or regulator, but when everything else is sorted it should be worthwhile. Brent tried a non-crown in the CD85 and said it wasn't nearly as good... also Lee luckily got a double crown for only £50 and is enjoying that in his '94. His 94 is the best old player I've heard I think, but the rest of his gear is nicely sorted too. Sadly, he doesn't post here much because of all the measurement / proof / objective bickering - he just likes to mod and then enjoy. Personally, I can just ignore the people I don't like and chat to the rest 
This forum has some wonderful members. 😎
So what model does Mr Clark have?
edit: looks like he has a 3.5. I think I've heard one of these, not bad as standard but like most players needs work! It has a Philips TDA1305 according to the Acoustica page.
Simon
This forum has some wonderful members. 😎
So what model does Mr Clark have?
edit: looks like he has a 3.5. I think I've heard one of these, not bad as standard but like most players needs work! It has a Philips TDA1305 according to the Acoustica page.
Simon
"So what model does Mr Clark have?"
I always get this wrong, Martin, if you're out there, correct me.
Its either a Naim CD2 or Naim CD3.5 or something.
And very nice it is too.
I always get this wrong, Martin, if you're out there, correct me.
Its either a Naim CD2 or Naim CD3.5 or something.
And very nice it is too.
Only just spotted this -if you are using the SAA7220, then paying great attention to the PSU for it is very important - and yes, reaps great gains. I posted more on my experience here:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=99652
Contrary to what you might expect, low frequency regulation/ and low reg output impedance really, really helps here, because it directly impacts on the bitclock phase noise. Do all you can to keep this supply quiet - a touch of RC filtering before the reg can help a lot (I have a whacking great gyrator here, with a long time constant). The complication of course is that the 7220 draws about 180mA, so you can end up wasting quite a lot of heat quite easily.
Then again, a good reg isn't enough on its own - you must pay attention to the decoupling, if only to ensure it doesn't upset your regulator; you will hear it! (I think I use 100nF stacked-film, plus 47uF OsconSP with 0.3R in series with one of its lead - works very well)
Phil, flattery will get you everywhere 😉 Yes mine's a CD2; TDA1541AS1 +SAA7220. Think I've squeezed rather more out of it since you were last here, fancy another beerdrinking evening soon if you're around ? 🙂 I really want ot hear your PMD100 hack!
http://www.diyaudio.com/forums/showthread.php?s=&threadid=99652
Contrary to what you might expect, low frequency regulation/ and low reg output impedance really, really helps here, because it directly impacts on the bitclock phase noise. Do all you can to keep this supply quiet - a touch of RC filtering before the reg can help a lot (I have a whacking great gyrator here, with a long time constant). The complication of course is that the 7220 draws about 180mA, so you can end up wasting quite a lot of heat quite easily.
Then again, a good reg isn't enough on its own - you must pay attention to the decoupling, if only to ensure it doesn't upset your regulator; you will hear it! (I think I use 100nF stacked-film, plus 47uF OsconSP with 0.3R in series with one of its lead - works very well)
Phil, flattery will get you everywhere 😉 Yes mine's a CD2; TDA1541AS1 +SAA7220. Think I've squeezed rather more out of it since you were last here, fancy another beerdrinking evening soon if you're around ? 🙂 I really want ot hear your PMD100 hack!
I really want to redo my regulators.
Yes, Martin. A beer sounds like a plan. Mine now sports a PMD100, so its probably a good time to meet up (except for maybe Christmas getting in the way). Anyway, that's maybe off topic. Perhaps.
Yes, Martin. A beer sounds like a plan. Mine now sports a PMD100, so its probably a good time to meet up (except for maybe Christmas getting in the way). Anyway, that's maybe off topic. Perhaps.
Just been to a Pendulum gig - full of bloody kids shoving around - so I'm too sozzled and ear-beaten to say anything interesting right now. However, thanks Martin for the link, which I have seen before. I am still reading all of your regulator stuff on Acoustica, quite interesting it is too, even though I can't think of anywhere to use a 317.... it's easiest to save up for a part that someone else designed well to begin with.
Anyway, about the digital filter, is a multiple cap approach with series resistance really better than just a 220uF Rubycon ZA?
Anyway, about the digital filter, is a multiple cap approach with series resistance really better than just a 220uF Rubycon ZA?
An externally hosted image should be here but it was not working when we last tested it.
philpoole said:
This is MY thread... and I love off-topic
Talking of meets - anyone going near Sheffield is welcome to come round and listen to a set of cheap open-baffles fed from a £10 Philips CD player hehe
"is a multiple cap approach with series resistance really better than just a 220uF Rubycon ZA?"
Well, if you have (and I'm not sure this is exactly what Martin means) the two capacitors in parallel, with the resistor in series between the two, you have a decoupling capacitor in series with a first order low pass filter, or essentially a 2nd order filter (with one of the resistors missing).
I tend to use an inductor (where the safety resistor would live - which incidentally doubles as part of a filter) in series before the circuit in questions decoupling capacitor, a few mH. So in this case, if you think of the main PSU cap, you essentially have a filter network, 2nd or 3rd order (depending on if you use the inductor or safety resistor). However, its not strictly an ideal filter as the first capacitor is shared with all other circuits.
Anyway adding further filtering to you rubycon 220uF would probably be better than having it work alone.
Well, if you have (and I'm not sure this is exactly what Martin means) the two capacitors in parallel, with the resistor in series between the two, you have a decoupling capacitor in series with a first order low pass filter, or essentially a 2nd order filter (with one of the resistors missing).
I tend to use an inductor (where the safety resistor would live - which incidentally doubles as part of a filter) in series before the circuit in questions decoupling capacitor, a few mH. So in this case, if you think of the main PSU cap, you essentially have a filter network, 2nd or 3rd order (depending on if you use the inductor or safety resistor). However, its not strictly an ideal filter as the first capacitor is shared with all other circuits.
Anyway adding further filtering to you rubycon 220uF would probably be better than having it work alone.
Yes, call it what you want - it's basically an RC lowpass with an emitter -follower tacked on.
About the caps - I meant using two caps in parallel, but the larger has some artificial ESR added. It stops any uncontrolled ringing from paralleling low-esr caps, and actually can outperform both. I can't speak for using the 220uF rubycon, but not all regs would be entirely happy with such a thing on the output.
About the caps - I meant using two caps in parallel, but the larger has some artificial ESR added. It stops any uncontrolled ringing from paralleling low-esr caps, and actually can outperform both. I can't speak for using the 220uF rubycon, but not all regs would be entirely happy with such a thing on the output.
Any "newbie-friendly" way to find out if it's happy or not? It works and sounds superb, but is there more to it than that?
BTW, it's worth noting this Philips is loaded with SMT parts on the PCB underside - things like the DAC and, possibly, digital filter have surface-mount bypass caps under there already. It's a nicely designed item in some ways I think.
Simon
BTW, it's worth noting this Philips is loaded with SMT parts on the PCB underside - things like the DAC and, possibly, digital filter have surface-mount bypass caps under there already. It's a nicely designed item in some ways I think.
Simon
That's good enough 🙂
Harder (but doable) is to knock-up a rig with your reg and proposed cap, and switch a dummy load (resistor ) in and out with transistor, and use a scope to watch what happens - whether you get a nice smooth settling at each transition, or evidence of ringing. Just a soundcard plus software will show you a lot in the audioband.
There's a really neat demonstration of this in the excellent application note AN47: High speed amplifier techniques by Jim Williams:
http://www.linear.com/pc/downloadDocument.do?navId=H0,C1,C1154,C1009,C1028,P1219,D4138
-take a look at the 2 or 3 pages on decoupling.
Well I'm still reading that article, it's really a good recommendation, nice one. I do have an old 'scope but I don't really know how to use it. I'll get it out of the attic and have a play one day.
Simon
Simon
TOP NOTCH HEALTH & SAFETY TIP!!
