Bigman80
Grandmaster
AA Founding Member & Bigbottle Audio Creator
Posts: 16,358
|
Post by Bigman80 on Aug 9, 2018 17:09:59 GMT
The PSX was a good bit cheaper than a hi cap.It had a massive transformer in it too. I thought it was better value than the Naim offering. It worked beautifully with the Cyrus Two as well. I have had mixed results with hi caps, of which I’ve had many. Sometimes a benefit, yet other times just a bit too “heavy” sounding. Teddycaps piss all over them though.
|
|
|
|
Post by macca on Aug 9, 2018 17:18:06 GMT
Have you seen inside the hi-cap? Nothing in there hardly, any EE could knock you one up for a ton. The Cyrus stuff was always compromised by being stuck in those tiny little cases - yes the WAF was high but putting form over function has no place in hi-fi for me.
|
|
Bigman80
Grandmaster
AA Founding Member & Bigbottle Audio Creator
Posts: 16,358
|
Post by Bigman80 on Aug 9, 2018 17:45:54 GMT
Have you seen inside the hi-cap? Nothing in there hardly, any EE could knock you one up for a ton. The Cyrus stuff was always compromised by being stuck in those tiny little cases - yes the WAF was high but putting form over function has no place in hi-fi for me. IMO the Hi-cap is a very outdated way of doing a PSU. The first Teddycap I heard rendered them obsolete. |
|
|
|
Post by macca on Aug 9, 2018 18:27:15 GMT
I'm still of the view that a good amp will have more than sufficient power supply on board already. I've got an XTZ amp here, was £800 new, traffo the size of a grapefruit in it. I've tried to clip it but the speakers start to give up first. So it doesn't have to cost the earth.
Naim was always the holy grail when I was young, the chrome bumper stuff all racked up really does look proper and the sound was legendary - allegedly. I didn't even know anyone with a Naim, let alone heard one.
Ironically by the time I had got to the point where I could afford it I had already heard it (full 6 pack system etc) and decided it wasn't what I was after although it did look very cool, and the SQ was good. Of course they say the sound has changed again since then so I don't know about the later stuff.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 10, 2018 14:54:49 GMT
Hello all. Been away from forums for a while as I had a bit of an overload / meltdown. I get all consumed by it to the detriment of other parts of my life. Not healthy for me.
Anyway starting to feel the love a bit again.
I have dicked about a fair bit with power supplies and especially with the various parts of a Naim CD3.5. My findings with that player maybe wont transfer to others I don't know. Another thing was I was listening and appraising kit differently back then before I discovered NVA. The 3.5 started off stock with one 120VA toroid with single 19-0-19 secondary supplying everything (transport, decoder, RAM chip, front display, inverter, clock, all 3 dac supply pins, two LM317 regged supplies for the anologue stage opamps (6 of) plus output relay and anything else I have forgotten about.
Naim put in 15 or 16 LM317 regulators to supply the various bits and bobs. I think the laser mech is even doubly regulated from memory. Output stage and related opamps get two regs, dac gets three (2 anologue, 1 digital), clock gets one, inverter gets one, decoder gets 3 or 4 (cant remember) and the rest for the transport display and other purely digital gubbins.
I bought a hicap for it way back which is connected to the anologue output stage only by removing a link on the back panel and plumbing the supply in to the provided din sockets. It was a good upgrade from what I remember although pricey. I wouldn't buy a manufacturers or off the shelf psu ever again now I know how to build one. The dicking about was just about to start however once I had plucked up the courage to point the soldering iron at it.
I ended up adding more independent power supplies and that meant more transformers, rectifiers, smoothing caps. In stock form the single secondary transformer is taken to two separate rectifiers and caps.The dac has three power supply pins, one digital and two anologue which were each supplied from independent LM317 regs. I ended up adding a very small EI PCB transformer with three secondary windings for the dac. Salvaged it from a gas boiler pcb. Obviously I had tracks to cut and then feed this new psu into the regs. Lived with that for a while then dicked about with the regulators.
As far as the output opamps are concerned they got the regs replaced by som ALWSR super regulators. Then I changed the hicap out for a home made one using bits acquired from Les at Avondale audio. Substantial EI frame transformer with two secondary windings was part of this. Liked that very much.
I put in a new fancy tentlabs clock regulated by a PFM diy flea and gave this its own dedicated psu. This also uses another very small EI transformer salvaged from a gas boiler. This clock upgrade was the single most musical upgrade I have done to this player. Astonishing in fact.
So the only bits supplied by the original transformer are the purely digital bits and bobs. Due to the much reduced load on it I decided to remove one of its rectifier and cap assemblies and just use the one for simplicity. Then added a gyrator / capacitance multiplier which was designed on pfm 10 years ago. It uses a mosfet and can handle large loads, in fact it prefers large loads like the digital section of this player. So added that in. Most of the other LM317 regs have been dicked about with to varying degrees.
What have I learnt from all this that is relevant to Andrew's case ? Power supplies are extremely important and individual to each type of component or design. Just because a big ass transformer works better in a power amp or phono stage does not mean the same applies to dac's, transports or anologue output stages.
I changed the very small EI transformers out for big toroids just to see what happened and all in all I preferred the small EI's (so they went back in) even though they are tiny and were originally powering a gas boiler. If a chip or transitor shares a regulator with another it usually brings a smile to your face if they get their own regulator. To take that even further chips and transistors can even be given their own dedicated psu and that in my experience can bring an even bigger smile, heck a laugh even.
So Andrew, what to do? I would say there would be an improvement adding another even if it is only from the two secondary windings being separated from the other two with a primary all of their own. EMI works in mysterious ways. Less work for one supply to do must surely be a good thing as increased load brings about increased noise. I appreciate you want to keep the thing reversible but if it were me I would be adding EI transformers left right and centre !
All IMO of course.
All the best, Stu
|
|
|
|
Post by macca on Aug 10, 2018 15:16:55 GMT
Hi Stu
good to see someone who is prepared to share their thoughts on this, most EEs either say it is a waste of time or act all cagey like they might be giving away trade secrets or something. I never understood that.
I'm interested that you found the clock upgrade made the most difference? Can I ask what you did specifically? I know people think that the improvement comes from reduced jitter but that is highly unlikely to be the case so it would seem there is something else occurring.
|
|
Bigman80
Grandmaster
AA Founding Member & Bigbottle Audio Creator
Posts: 16,358
|
Post by Bigman80 on Aug 10, 2018 15:32:26 GMT
Thanks Stu. Nice to see you back in circulation and great to have advice from someone who speaks from experience. I hope to grab a listen to the DAC in about 2 weeks. I will get to know it for a little while but I think I will dabble with another transformer. It’s about the simplest piece of modding I will ever find and even tenn I will be a bit nervous  |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 10, 2018 16:19:11 GMT
Hi Stu good to see someone who is prepared to share their thoughts on this, most EEs either say it is a waste of time or act all cagey like they might be giving away trade secrets or something. I never understood that. I'm interested that you found the clock upgrade made the most difference? Can I ask what you did specifically? I know people think that the improvement comes from reduced jitter but that is highly unlikely to be the case so it would seem there is something else occurring. Hi macca, I always openly share my thoughts as that is very much the ethos of diy (or it should be). Others helped me a decade ago when I knew less than bugger all. I am not an EE - far from it in fact. No formal education in electronics just what I have picked up along the way mostly at pfm, diyaudio and recently hfs. I don't use test equipment as I don't own any and wouldn't know one end of an oscope from another. I use my ears as the one and only test measurement  Of course the EE's (designers & manufacturers) get cagey. I think you are right that they don't want to give too much away and then why should they. The diy enthusiast community is completely different I would add. There are many hobbyists out there who have dicked around with this and that - some of them are very knowledgeable but some are too objective. The clock upgrade involved disabling the original one (just two or three parts to remove) and the new clock feed wired in to some pcb pads vacated by these removed components. The clock was called a pfm flea and was designed by some very clever chaps in the pfm diy room (where I cut my teeth). It uses a very high quality oscillator made by tentlabs and the clever part is the regulator (which is on the same pcb) that was built around a AD797 ? opamp. Can only deliver small amounts of current at low voltage which is ideal for a clock as it only consumes about 10ma at 5V. The opamp restricts the current it can supply due to its limited heat dissipation. This could have been powered from a raw DC rail in the player but ultimate performance came from giving the flea its own dedicated psu. A small transformer is all that is needed - mine is 3VA. The earthing (0V) connection from the psu and reg to the cdp motherboard is where the application can be ruined and this was the cause of much chin stroking. I said the clock upgrade brought the biggest musical upgrade. Not to be confused with hifi improvement. Sure there was bags more detail but the players performance became much more emotional, instruments sounding much more like the real thing, the interplay between them being enhanced also. Looking back it was like the musicians were playing at the other side of a waterfall before the mod and after they were not blurred and wavy like before. That's the best description I can come up with i'm afraid. Whether the improved musical enjoyment was due to reduced jitter I have no idea or way of knowing. The clock was put in at the same time as the new regulator which replaced the original LM317. I have a lot of experience of messing about with or replacing LM317's and the kind of improvement I heard was not due to regulator being improved. Sure there must have been some improvement due to the reg but the oscillator from tentlabs was far superior from what I remember. I tried if first from a raw dc rail in the cdp first and it was a really good mod. I then gave it its own psu and the result became astonishing. So, a bit of everything thrown in I suppose - new oscillator, improved regulator, added dedicated psu. In my case (CD3.5) I was able to bypass the inverter completely. As it had a reg of its own I also removed that and the now unused LM317 that was powering the original clock. Hope that kinda answered your question. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 10, 2018 16:23:42 GMT
Thanks Stu. Nice to see you back in circulation and great to have advice from someone who speaks from experience. I hope to grab a listen to the DAC in about 2 weeks. I will get to know it for a little while but I think I will dabble with another transformer. It’s about the simplest piece of modding I will ever find and even tenn I will be a bit nervous Hey Andrew, Experience is the only true way to know for yourself with these things. You have to judge who is giving the advice, their motivations and if they think about music in a similar manner to you. You only need to wire in the extra trannie to the mains switch/fuse holder on the primary side and the secondaries to the motherboard. You can double up the secondary windings (in parallel) and make full use of the transformer. Voltage output will stay the same but current handling increases. |
|
|
|
Post by macca on Aug 10, 2018 16:56:46 GMT
Thanks Stu. As far as I am concerned if you can do work like that you are a double E, qualifications mean nothing. You can either do summat or you can't.
Do you know what jitter actually sounds like? It sounds exactly like cassette tape when the tape is a bit slack or the tape deck is just a bit crap and you get that wowing sound especially noticeable on piano. You hear it with a vinyl LP also if the hole is not dead centre. So peoples issues with digital ('it sounds harsh, flat etc) are nothing to do with jitter since jitter doesn't have that effect.
I was hoping you could have shed a bit of light on why improving power supplies and adding power supplies improves digital. I've a lot of cd players and the ones that are clearly better all have multiple power supplies - otherwise they doesn't seem to be any major differences to account for it. There's no question in my mind that you get results exactly like what you described, less fuzziness and more weight and depth to the sound.
|
|
Bigman80
Grandmaster
AA Founding Member & Bigbottle Audio Creator
Posts: 16,358
|
Post by Bigman80 on Aug 10, 2018 17:16:10 GMT
Thanks Stu. Nice to see you back in circulation and great to have advice from someone who speaks from experience. I hope to grab a listen to the DAC in about 2 weeks. I will get to know it for a little while but I think I will dabble with another transformer. It’s about the simplest piece of modding I will ever find and even tenn I will be a bit nervous Hey Andrew, Experience is the only true way to know for yourself with these things. You have to judge who is giving the advice, their motivations and if they think about music in a similar manner to you. You only need to wire in the extra trannie to the mains switch/fuse holder on the primary side and the secondaries to the motherboard. You can double up the secondary windings (in parallel) and make full use of the transformer. Voltage output will stay the same but current handling increases. Thanks Stu. I didn’t know I could double up the secondaries like that. It actually makes it neater too. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 10, 2018 19:03:01 GMT
Thanks Stu. As far as I am concerned if you can do work like that you are a double E, qualifications mean nothing. You can either do summat or you can't. Do you know what jitter actually sounds like? It sounds exactly like cassette tape when the tape is a bit slack or the tape deck is just a bit crap and you get that wowing sound especially noticeable on piano. You hear it with a vinyl LP also if the hole is not dead centre. So peoples issues with digital ('it sounds harsh, flat etc) are nothing to do with jitter since jitter doesn't have that effect. I was hoping you could have shed a bit of light on why improving power supplies and adding power supplies improves digital. I've a lot of cd players and the ones that are clearly better all have multiple power supplies - otherwise they doesn't seem to be any major differences to account for it. There's no question in my mind that you get results exactly like what you described, less fuzziness and more weight and depth to the sound. Cheers macca. My knowledge is patchy and i had alot of help over at pfm. Doing is the best way to learn however. Your description of jitter is not that far from my waterfall description so it must have been something to do with reduced jitter, amongst other things. I would suggest that most of what contributes to the sound of a cdp is the dac, anologue stage and associated psu (s). I would also suggest that excessive complexity will also have a part to play and dont forget the grounding scheme and layout can make or break a design. Trying to chose my words carefully here. One psu powering everything is definitely hampering things. The noisiest parts of a cdp are the purely digital parts i.e transport, decoder etc (everything before the dac). This noise makes its way back to the psu where ultimately it gets into the dac and anologue stage . Like modulation i suppose. I have monitored the current draw of the digital section of my cd3.5 and it was in the order of 350mA and further to that was fluctuating all over the place. A transformer of higher VA rating will sound better in this application as it can deal with the variation in load much better due to its lower impedance or winding resistance (otherwise known as stiffness). The dac (whilst a digital device) i prefer to view it as anologue as it is taking the digital signal and outputting anologue. Naim gave each supply pin its own reg so the ultimate was to give each reg its own transformer winding. It really was a major improvement to my ears. It was only a very small transformer that i used but i think the fact it is an EI transformer that made the diference. Less inter winding capacitance means less mains crap being transferred to the secondary winding and thus the rest of the psu. The dac has a constant load (does not vary widely like the digital section before it) and so my thoughts were that the size of the transformer was not critical. When i swapped back to toroids for comparison my suspicion was confirmed. So the EI trannie was quickly put back in. The ano,ogue output stage was a different kettle of fish altogether. Whilst mainly made of 6 opamps i found here that an EI was my favourite but bigger was better unlike withe dac. Opamps usually benefit from a nice clean supply depending on design and application and the fast opa627 i use are certainly no exception. Currently i have the 6 opamps split across two psu's but i am planning on trying a psu for each of the 6. Yes this cdp is spawning psu's. Others have found with the digital side that improving the psu before the regulators had little effect. I agree with this to some extent but they have not tried everything and neither have i. Bigger transformer does make a difference though especially as the current drawn is up and down constantly. The final regulators before the various chips are probably what hampers any other potential improvements that may be available. I use the term improvements and this does not always involve adding extra complexity. Sometimes it can be from simplifying things. A word on transformers. They are not all made equally. Some are better than others and it is audible imo. An EI trannie lets through less mains crap than a toroid but usually has higher winding resistance and a bigger stray field. The bigger the transformer the bigger the stray field and higher the chance of problems from it. So by all means put in a bigger transformer or an EI but you have to be more careful with where you put it. I have done a fair bit of work on this. Increasing distance between a transformer and sensitive circuitry does reap large musical gains. It even improved when there were no audible hum problems to begin with or other problems. All IMO of course. I wait for some body to come along and tell me i have got it all wrong. The internet is full of bollocks and wrong information. Judge my ramblings for yourself before taking it as gospel. Stu |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 10, 2018 19:10:53 GMT
Oops put my reply in the quote box.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 10, 2018 19:24:08 GMT
Hey Andrew, Experience is the only true way to know for yourself with these things. You have to judge who is giving the advice, their motivations and if they think about music in a similar manner to you. You only need to wire in the extra trannie to the mains switch/fuse holder on the primary side and the secondaries to the motherboard. You can double up the secondary windings (in parallel) and make full use of the transformer. Voltage output will stay the same but current handling increases. Thanks Stu. I didn’t know I could double up the secondaries like that. It actually makes it neater too. It didnt dawn on me for years either. If you only connect one secondary you are wasting half the transformer. The primary winding stays the same obviously and so does the size of the core but maybe putting the secondary windings in parallel reduces the resistance much like when putting two identical resistors in parallel halves the resistance. Are the secondary windings centre tapped ? If not then two secondary windings of say 0-18 can be wired in such away as to make a centre tap ie 18-0-18. Naim always use centre tapped even when making a single rail supply (positive voltage only rather than split rail with positive and negative) because it sounds better due to the secondary windings connection to 0V not having to go through a rectifier diode. In essence half of the transformer is not used but in this instance the gains vastly out weigh the loss. Hope my ramblings are correct and i have not bored you to death. Stu |
|
Bigman80
Grandmaster
AA Founding Member & Bigbottle Audio Creator
Posts: 16,358
|
Post by Bigman80 on Aug 10, 2018 19:37:12 GMT
Hi Stu. I wouldn’t know that one, or even know how to find out. Is there an easy way of knowing?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 10, 2018 19:56:58 GMT
It usually says on the transformer. 0-18 (or what ever voltage it is) will mean two wires coming from the trannie to the dac board. If there are two secondary windings the there will be two pairs ie 0-18 and another 0-18.
If the widings are centre tapped then it might say 18-0-18 18-0-18 on it. If not then if there ate two sets of three wires coming out and going to the dac board then that will be two centre tapped windings.
Failing that its time to get the multi meter out
|
|
|
|
Post by dsjr on Aug 10, 2018 20:00:38 GMT
My 5V supply for the cheapo dac has paralleled secondary's outputting 12V (a bit higher unloaded) to the single rail adjustable regulator. No idea if it doubles the current capability or not though...
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 10, 2018 20:34:04 GMT
My 5V supply for the cheapo dac has paralleled secondary's outputting 12V (a bit higher unloaded) to the single rail adjustable regulator. No idea if it doubles the current capability or not though... Hey dave, I believe it does. Dont want to peddle misinformation though. I have not tested it. How ya doin' ? You seem to be hanging out alot here and not so much on hfs. Missin' ya dude. Stu |
|
Bigman80
Grandmaster
AA Founding Member & Bigbottle Audio Creator
Posts: 16,358
|
Post by Bigman80 on Aug 10, 2018 20:42:20 GMT
I took some pics but can’t see that part of the label, so I will take the lid off again and have a look. Now you’ve reminded me, I’d heard that before. If only I’d retained it!
|
|
|
|
Post by macca on Aug 10, 2018 21:38:32 GMT
Oops put my reply in the quote box. No problem thanks for the comprehensive reply. What you say ties in with my thinking and observations, especially regards the best areas for improvement being before the DAC. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 10, 2018 21:46:33 GMT
Hey Macca,
No probs. Is that a typo cos i was trying to say the best improvements came from the dac and after not before. If my ramlings have suggested otherwise i apologise.
ATB, Stu
|
|
|
|
Post by macca on Aug 10, 2018 21:52:55 GMT
Hey Macca, No probs. Is that a typo cos i was trying to say the best improvements came from the dac and after not before. If my ramlings have suggested otherwise i apologise. ATB, Stu I was referring to this that you said; ' The noisiest parts of a cdp are the purely digital parts i.e transport, decoder etc (everything before the dac). This noise makes its way back to the psu where ultimately it gets into the dac and anologue stage'As you go on to say it is intermodulation distortion and it increases the noise floor and starts putting a blanket on everything. It's why CD transports sound different. I agree the DAC and the output stage also make a difference. But I think the real key is what goes before them. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 11, 2018 8:04:15 GMT
Oh, I get ya now.
Stopping that noise polluting the rest of the player is important I think. Best way to do that is give the noisy bits their own psu.
Stopping mains noise (ripple) getting to circuitry is all well and good (throwing fancy regulators at it) but there are other things you can do. That's why I like EI transformers, they seem to let less crap through than a toroid but obviously cant stop the noise from the rectification process. Some people put snubbers in everything (caps and resistors) but I tend to share the view of RD in that can effect the music.
RD swears by toroids for all his gear. I have EI trannies on the front end of my diy NVA amps and like them a lot. Were all different at the end of the day which is why you should take what I say as just my opinion. Les at Avondale also loves a good EI transformer. I think we are all agreed though that they are not as good as a toroid for power amp output stages. Never tried an EI on the digital sections of my CDP - I may just have to give that whirl.
I had a wail of a time ffing about with that cdp and still not finished. Still got lots to try - charcroft bulk foil resistors in the anologue stage, giving each opamp its own psu, EI transformer on digital psu. Its never ending fun. A humble LM317 reg can be improved without adding to the parts count. On regs set for 5V output the voltage set resistor can be replaced by two ordinary green diodes for lower impedance and noise. I still have a couple of 5V regs to do that on including the reg that is hidden in the swing out transport tray.
Some say that all transports sound the same, others that they can hear a difference. Same for dac's. I don't know what is going but one thing is for sure and that we all judge sound differently, the gear is all different and the only way of really knowing is to do it yourself.
Everything is made down to a price (well almost everything). Corners have been cut, money has been saved, marketing bollox spouted blah blah blah. Trust no one. There is always an agenda. Complex is not always better, sometimes simplicity is key but the art is knowing when and where.
Sorry Andrew, gone off topic and its your thread really.
|
|
|
|
Post by dsjr on Aug 11, 2018 8:37:00 GMT
My 5V supply for the cheapo dac has paralleled secondary's outputting 12V (a bit higher unloaded) to the single rail adjustable regulator. No idea if it doubles the current capability or not though... Hey dave, I believe it does. Dont want to peddle misinformation though. I have not tested it. How ya doin' ? You seem to be hanging out alot here and not so much on hfs. Missin' ya dude. Stu I like to talk about the gear I once knew and some I didn't - and on and on and on and on..... I don't like dissing other people or banging on about fifty year old prog albums.... |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 11, 2018 12:44:13 GMT
|
|
Bigman80
Grandmaster
AA Founding Member & Bigbottle Audio Creator
Posts: 16,358
|
Post by Bigman80 on Sept 4, 2018 12:26:29 GMT
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 4, 2018 13:33:15 GMT
Very neat and smart.
But. I know it's good gear, but it looks a bit too 'homogenised' for my liking, the total opposite to my system, which is a very eclectic mix.
|
|
Bigman80
Grandmaster
AA Founding Member & Bigbottle Audio Creator
Posts: 16,358
|
Post by Bigman80 on Sept 4, 2018 14:29:46 GMT
I don’t think I’ve ever had matching kit in my life, so it’s a new experience. Once I get somthing nice to sit it on, I will know more clearly how I feel about the look of it all together.
If placement isn’t an issue, I would like to have DAC under the CD player and the phono stage under the Campion: Then have the pair sat side by side. Where I put a deck is anybody’s guess. The PS6750 is set up and so is one of the two Revolvers I own. None really excite me so I’m looking for something prettier and more interesting.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 4, 2018 16:13:28 GMT
Very neat and smart. But. I know it's good gear, but it looks a bit too 'homogenised' for my liking, the total opposite to my system, which is a very eclectic mix. Your system is Hapa! S. |
|
Bigman80
Grandmaster
AA Founding Member & Bigbottle Audio Creator
Posts: 16,358
|
Post by Bigman80 on Sept 4, 2018 16:15:07 GMT
|
|
