Conference hips::uk_audioo

    
    You may find that it's just the material. A more modern CD player might
    help a bit, but personally, I didn't find that a better CD player made
    bright CDs sound better.
    
    Are you sure it isn't the speakers that are more responsible ? Linn
    speakers do have a reputation for being balanced on the bright side of
    neutral. My own Epos ES11s have a slight lift in the upper mid-range,
    and this certainly doesn't do badly-recorded CDs any favours.
    
    Trevor

    Though I have the KELVIN I think they have gone belly-up over the last
    6 months. I may be wrong but I think the QED is the one to watch at
    that price point. 
    
    I've "improved" the KELVIN with a better external power supply and some
    local PP shunt caps. from your address your based at SBP I guess.
    
    You'd be quite welcome to borrow the DAC if you'd like. What ever you
    do make sure you try YOUR transport with the DAC of coice. I borrowed a
    friends MERIDIAN 203 to try with my Marantz CD65IISE. The DAC never got
    lock-on and hence never emitted more that a wierd burr noise.
    
    Contrary to the note header I'm actually fangio::ivesj
    
    Cheers
    
    jives

    Thanks for the offer Jonathan, 
    
    It's pity about Kelvin not surviving.  Interesting about some
    convertors not agreeing with certain transports.
    
    Were the modifications you made only to the power source?  I should 
    think the inside on one of these is untouchable.  What did you perceive
    as "improved"?
    
    I am going to Reading Hi-Fi tomorrow morning. I'll take my Kenwood 
    CD/transport with me.  They now stock my Musical Fidelity Amp, but will 
    have to find something similar to the Index II's, as they have given 
    up the Linn Dealership (!!!) .  
    
    I REALLY hope I like the difference the DEQ Digit makes for the price. I 
    hope it doesn't fall down against the more expensive examples, such 
    as the Deltec Little Bit.  After all It's only a number cruncher!
    
    Must remember to clean my ears out tonight, and take some objectiveness 
    pills in the morning!  If I don't have an easy decision I should think 
    they'll let me borrow one over the weekend.  
    
    Now, what am I really going to loose by switching to a bitstream DAC?
    
    regards,

    Maybe you can add the Audio Alchemy's Digital Decoding Engine V1.0 to
    your list. In North America it costs about $450. It was rated a "Best
    Buy" by Hi-Fi Choice (No. 101). The Deltec Little Bit was rewieved in
    the same issue and received a somewhat lower rating. 
    
    I have listened to the DDE in several high-end systems, and it offers 
    quite a smooth and natural sound. It has much more flexibility and options
    for future upgrade than the QED DAC.
    
    Emil   

    I got the technical data sheets from Philips on the 7320 DAC chip.
    There sample application is more or less the same. The chip tself is a
    44 leg surface mount device and off limits to all but a micro-surgery
    expert (who does'nt get the shakes as bad as I do !).
    
    what I did was remove the internal transformer and bridge rectifier and
    small resevoir cap. I connected a cable to the points on the tracks
    which corresponded with +5v, 0v and Ground. This cable was dressed out
    of the Kelvin box through a grmet and has a 3 pin din plug on the end.
    I built s seperate power supply capablle of 4 times the current
    delivery, using a bridge of schottky diodes and lots of low ESR caps as
    a resevoir (in paralell). I also used an LM317AT variable voltage reg
    rather than a 7805 (Variabl;e voltage regulatrs have superior
    performance, ask Mr Meakins !). the KELVIN like the QED uses a single
    power supply but with some local decoupling, so my philosophy was to
    make this supply as high-current/low impedence as I could. This is all
    in another back box with a 3 pin din socket in the side.
    
    the result is more detail, basically. I notice more things going on .
    
    I then shunted all the tantalum bead caps with 0.01uF tiny wima pp
    caps. Not sure how much differtence this makes, but one of the 44 pins
    on the 7320/21/23 is to an external smoothing cap for the internal
    voltage reference. Extra smoothing and a lowering of the ESR may help
    (Mr DELTEC reccomended this in a magazine I read a while back).
    
    Future plans include a valve based analogue section (the Kelvin uses
    the chips own cooking op-amps which I would disable) using 2 twin
    triodes ECC83's. I got the schematic from Glass Audio and was able to
    canibalise my old Radford SC22 pre-amp (QUAD 22 vintage) for most of
    the parts . This should raise the output voltage to something cloaser
    to ideal passive operation i.e 2v+
    
    I read that NAIM think DACS are susceptible to vibrations so the board
    sits on sorbothame washers. I've thought about damping the chip itself
    but was worried about it overheating.
    
    it's all been quite good fun since I recognise that the QED has nobled
    any re-sale value it had.
    
    most of these ideas you could apply to the QED. The power supply is
    easy because it already uses an external unit. The QED is also POOGE4
    'able using IC op-amps as it does.
    
    

    
    It's interesting to hear that Reading Hi-Fi are stocking Musical
    Fidelity again. About 18 months ago, when I was last thinking of buying
    something from them, they were trying to dispose of all their MF stock
    on the cheap! Still that was before they lost the Linn dealership.
    
    Trevor

    re: .5
    
    Ah! at last someone doing something interesting with power supplies.
    
> using a bridge of schottky diodes
    
    I've often thought about using these, have you found they let more
    detail through?  I've read all the Russ Andrews comments, I know he's
    sold on them.... I mean sells them....  Must admit I have them in my
    plans for a psu.
    
> and lots of low ESR caps as a resevoir
    
    I'd be interested to know which type and how many.
    
> I also used an LM317AT variable voltage reg
  
    Is this the big power transistor looking T05 type, like the one Naim
    use in the Hi-Cap?  If so, have you ever seen a 337 version, I've not
    and would like to build a dual rail supply with 317 and 337.
    
    Did you use the large elect cap on the 317 as our friend Walt suggests
    and did you use the protection diodes or just build the reg as
    standard?
    

    I use components that I get from local distributors, thewre is a shop
    in Farnbrough which I think orders from Farnell.
    
    I tried the diodes because of what RA said but ordered commercial ones
    at a cheaper price. for values under 1AMP they are not much more
    expensive. You can even get schottky bridges now.
    
    Always remember to watch the PIV value though as this should be greater
    than 3 times the secondary winding voltage I believe.
    
    I use MAPLIN low ESR caps which are roughly equivalent to Rubycons,
    I've never found anyone who does Elnas and I can buy these over the
    counter in Reading (Oxford Rd).
    
    the LM317 I use (I thought it was AT, I now it's higher spec than a
    normal 317) is the 3 pin variety NOT then CAN/transistor style. I find
    these easier as they are smaller. I tried the WALT CLARK suggested
    config with the 1000uf across the adjust , but the sound seemed
    excessively bright. Maybe it was psychoacoustical or what , but I went
    back to the text book config (without safety diodes) and this is what
    I'm using. I need to figure out a better fuse arrangement, so that I
    can leave it on all the time. At present the only fuse is the 3amp one
    in the mains plug ! I use WIMA PP caps which I get from electromail,
    these are the tiny ones with a gap between the tags about the same as
    the Tantalum beads that I'm shunting.
    
    Finally I have "burnt" in the new power supply/caps by playing a CD on
    repeat for around 50 hours.
    
    I'm fairly unscientific in these things (life is too short) I rely
    mainly on a multimeter and my ears, though I do believe that initial
    reactions can be misleading.
    
    re. different previous note.
    
    My experiences of Reading HiFi are that they are a really good shop,
    they don't hoist there opinions on you like Audio T here in
    Basingstoke. I've always been happy when I've gone there, and have got
    a couple of things from them.

    Oh I forgot. I use around 3,000uF of smothing cap based around 6*500uf
    values (the largest number I could get in the box), with a 1Uf PP shunt
    and a 0.0001uf PP shunt. I've never seen the MAPLIN stuff compared with
    the exotic stuff but I know Rubycon are rated by Musical Fidelity and
    on paper the MAPLIN caps are equivalent interms of Impedence at
    Frequency extremes. Interestingly the MAPLIN Metal Oxide resistors are
    superior (on paper) to just about everything including the holcos.
    There temperature coef is better as is the tolerance. Hayden Boardman
    who runs a company rebuild Leak Amps apparently swares by them.
    
    I'd be interested in your PSU experiences. Is this for you LINDSAY HOOD
    amp ? 
    
    Around this time I also built a DC 6v heater supply for the
    driver/phase splitter on my valve amp. I found that the high current
    regulators (like I guess you would use for a power amp) are VERY
    unstable if not adequately cooled. in fact bolting it to the case work
    was the only way to stabilise the supply. Once that was done it was
    steady as a rock, I found this out when I blew some dust off and the
    supply stabilised for a few seconds!

>    I'd be interested in your PSU experiences. Is this for you LINLSEY HOOD
>    amp ? 
    
    Sort of, but not the one you're thinking of.  It's for the RIAA preamp I'm
    about to submit a review of to HFN.  Hope they're interested in it. 
    BTW, it sounds very good indeed - in spite of using 78- and 79 regs.  I
    was thinking about 317's and 337's very close to the circuit board with
    schottky diodes and multiple caps.  I also want to use a large
    transformer (100 to 150VA).  It's just getting the time to try this.  
    
    The power amps already have a stab psu.  I would like to try them with
    unstab for high current supply and totally separate stab supply for the
    low current section.  Again time's the problem.
    
    Like you, I just use a multimeter and ears.  The problem is that some
    changes cannot be A-B compared due to their nature and almost become an
    act of faith.  I'm never happy with this.

    RE .9: Do you know any shop accept mail order of the MAPLIN resistor
           and the list of it values available ? Any good quality 36
           stepper for passive preamp volume control ?
           Three postion dual gang swich for input selection ?
    
           Request from oversae "Singapore"
    
    Thanks
    Parry

    re .10
    
    How importatnt do you think is where the regulator sits in reference to
    the "user" of that supply. I regulate in the PSU box and have 12" of
    QED incon (screened) carrying the low impedence stab supply into the
    DAC where it's just soldered onto the tracks. I guess this raises the
    impedence somewhat. I guess I could move the regulator down into the
    DAC and feed in unstab voltage. Do you reckon this is significant.
    
    Also I'm embarrassed to say that I "wimp" out setting up the 317 by
    using a pot for the adjust pin. I don't have the algorithm by which the
    ratio of the two resistors decide the oout put supply (+5v) for this
    chip. I guess you use two metal oxide resistors. I should do as I guess
    the pot is a) awfull in performance and b) likely to drift over time
    .

    re .12 and regulator positioning
    
    The reg should certainly be as close to it's "user" as possible.  One
    technique is to regulate down to target voltage + 5ish volts in the
    remote psu and then down to the required voltage by the most important
    chip.
    
    Audio Synthesis and Walt Clark have separate regs for most of the
    chips, this can be done by building a daughter board over the mother
    board.  
    
    re the 317 algorithm, it's in the Maplin catalogue, I seem to remember
    1k5 and 100R for 20V, I check it tonight.

    re .12
    
    The 317 regulator algorithm is: 
    
                  1.25(1+R2/R1)
    
    where the circuit is:
    
             3|_____|2---|
                 1      R1
                 |_______|
                R2
                 |
    -----------------------------

    Keep in mind that the current flow through the resistors should be
    5..10 mA, so the second equation for R1 and R2 will be
    
           R1+R2=Vout/0.01 or R1+R2=Vout*100
    


                                                 - Rik -
    

We'll I've now had time to get used to the new addition to my system. My 
decision was to buy the Deltec Little Bit DAC, reduced to �200 at 
Reading Hi-Fi.

I compared it to a Meridian 203 DAC, an older spec model, being priced 
at �400.  There was also a �300+ New Sony CD Player.  It was already 
setup, in case the salesman didn't think my Kenwood performed well as a 
dedicated transport.  He considered it quite good, so the Sony remained 
unused.

He managed to duplicate my Amp and speakers exactly, in conversation he 
did mention that the Linn Index speakers can sometimes be a bit bright.  
Involving, fast, exciting, but a little bright.  Hmm, but I do so enjoy 
them for a lot of material...

Listening to some tracks I selected, some because they annoy me with 
harshness when I want to enjoy the music.  Others because I enjoy them.
 
The Deltec did not make my naked player sound cheep, it just had a 
little more refinement in the treble, especially female vocals.  Things 
still arn't perfect though.  Some "SSS" words still sound far from 
natural.

The Meridian sounded very close to the Deltec.  (I picked on the old 
spec at �400, it was the same sound as the �570 current model).  The guy 
was obviously playing the Deltec first, the the Meridian.  I think 
psychologically, one is bound to prefer the second time round, if things 
are this close.

This was a difficult decision.  I asked not to be told which convertor 
was being swapped in.  But I KNEW he would continue to play the Meridian 
last.  The Meridian may have had a very slight grip on the bass notes, 
but it was very, very close.  On a Chic single (which I immediately 
liked, but did not know) he may as well have played it twice through the 
same DAC!

So, I decided on the cheaper Deltec.  Since listening at home, I enjoy 
the difference more and more.  Kick drums are far more natural 
and pleasing, who said Bitstream lacks bass?  I also find the music 
contains metalic cymbols, not poorly ending random collections of hiss 
and zings.

Detail, yeah Clive and Jonathon; I know know just what you mean, just 
that extra feeling of what's being played and how.  I don't know if I 
could manage to convince someone else this yet!

I opted not to purchase a digital signal connecting cable, at �29-�70 
for about half a meter.  I remembered a note in this conference and 
asked a kind Network specialist if I could take a foot of Thin-Wire 
ethernet from his junk pile.  I terminated this with two 50p gold plated 
phono plugs.  

Since talking to Jonathon, I'm unsure about the thin-wire impedence, I 
think it may be 50 ohms, when 75 may be required.  Anyone know for 
sure????  I'm borrowing a TV coaxial lead of Johnathon's, I'll see if 
this improves things.

A further suggested tweak is ferrous rings, best placed around the power 
cord for the DAC and Transport to help isolate them further from each 
other.  I don't think I'll whip off the covers and build in a 
"regulated-spike protected-unleaded-caffein free-smoothed power supply" 
just yet!

Bitumen deadsheet can be applied around the transport to avoid any 
resonance around the player.  Similarly isolating the DAC from any 
vibrations may also be a good move.

Any further ideas that won't invalidate the guarantee??

Cheers,
Rob.