Hi Ray
Looks like brent has another winner !
Yesterday I had a long talk with Ken Ishiwata and after explaining the mods done in my CD53 he told me I should try discrete regulators.... I told him about brent´s Spower but he did not know these.
In the end I asked if he liked files and he answered yes if these are WAV.
He does not like PCM and devised a special kind of dac that does not use fixed 24bit words.... floating number of bits is better he claims and that by itself would take care of file sizes without the need for compression.
I then asked him about caps... He considers ELNA Silmic II to be even better than BG and uses Silmic or cerafine depending on the location... In the end told me to be very carefull about nonpolar caps...
It was a very informative conversation indeed.
Ricardo
Looks like brent has another winner !
Yesterday I had a long talk with Ken Ishiwata and after explaining the mods done in my CD53 he told me I should try discrete regulators.... I told him about brent´s Spower but he did not know these.
In the end I asked if he liked files and he answered yes if these are WAV.
He does not like PCM and devised a special kind of dac that does not use fixed 24bit words.... floating number of bits is better he claims and that by itself would take care of file sizes without the need for compression.
I then asked him about caps... He considers ELNA Silmic II to be even better than BG and uses Silmic or cerafine depending on the location... In the end told me to be very carefull about nonpolar caps...
It was a very informative conversation indeed.
Ricardo
Cool
Funny, I happen to have my eye on the Elna's as a replacement for the obsolete BG's. Seems that's the way to go then
Ray
Compared to BG, Silmics lack detail IMO. They have nice tonal info, but seem shut in in the upper mids and treble to me.
The best alternative I've found are Rubycon ZA, they have similar detail levels to BG FK, although they can sound a little "dry" and aren't to everyone's taste. It isn't really an issue when using good regs though.
A nice combo, is a large Silmic after the reg, and then ZA close to the load.
Cheers, Lee.
Some days ago i replaced the 7x15's with 7x12's (i somehow got that wrong), which was hard enough (note: i should get my self a desoldering iron) for some reason.
Today i converted one channel to CFP (moving one resistor from top to bottom, cutting one track and adding the addtional parts). Used two 3k resistors.
Then compared the modified channel with the original one on an acient spectrum analyser (7L5) with a 1kHz testtone on a test cd. And the distortion went...... up.
So i might have f....-up again. Or i'm measuring the wrong stuff anyway. I've said before that i was supprised at the level of distortion. I now have some doubts on the SA or better on how i'm using it. It looks to me that i'm seeing overload at the input side of the SA instead of the distortion of the source.
I do notice (with a scope, but still) that the modified channel has a bit more gain. Guess this is normal (?). So maybe i'm overloading the SA more.
I'll look at this again later, got other stuff to do. And i was not focussing on the output stage in the first place. Just needed one so i can work on other mods for this player (which are not in this thread).
Today i converted one channel to CFP (moving one resistor from top to bottom, cutting one track and adding the addtional parts). Used two 3k resistors.
Then compared the modified channel with the original one on an acient spectrum analyser (7L5) with a 1kHz testtone on a test cd. And the distortion went...... up.
So i might have f....-up again. Or i'm measuring the wrong stuff anyway. I've said before that i was supprised at the level of distortion. I now have some doubts on the SA or better on how i'm using it. It looks to me that i'm seeing overload at the input side of the SA instead of the distortion of the source.
I do notice (with a scope, but still) that the modified channel has a bit more gain. Guess this is normal (?). So maybe i'm overloading the SA more.
I'll look at this again later, got other stuff to do. And i was not focussing on the output stage in the first place. Just needed one so i can work on other mods for this player (which are not in this thread
).
Last edited:
Hi Guido,
3k is a bit low, 5k6 or 6k8 works better. But in all cases you should not see high distortion anyway. A big problem with these measurements is finding good source material. If the sinewave on the test CD is not pure (often they are of the synthesized type, with staircase-type waveform), you are measuring that instead of the DOS distortion. This is the main reason I have not yet done any THD measurements on my ezDAC and DOS.
Second problem is that most equipment has 50 ohm inputs and that's way too low for audio equipment. This will dramatically increase distortion if connected directly. You can try adding a series resistor, like a 10x scope probe, since this is a relative measurement anyway. Just set the 1kHz spike to reference level, like 0dBm. Levels will be small though, because of the 9M ohm vs. 50 ohm divider. A simple 47k resistor may work better.
Regards,
Ray
3k is a bit low, 5k6 or 6k8 works better. But in all cases you should not see high distortion anyway. A big problem with these measurements is finding good source material. If the sinewave on the test CD is not pure (often they are of the synthesized type, with staircase-type waveform), you are measuring that instead of the DOS distortion. This is the main reason I have not yet done any THD measurements on my ezDAC and DOS.
Second problem is that most equipment has 50 ohm inputs and that's way too low for audio equipment. This will dramatically increase distortion if connected directly. You can try adding a series resistor, like a 10x scope probe, since this is a relative measurement anyway. Just set the 1kHz spike to reference level, like 0dBm. Levels will be small though, because of the 9M ohm vs. 50 ohm divider. A simple 47k resistor may work better.
Regards,
Ray
Last edited:
I'm using two 3k resistors in series. I'll check the source material with another (standard) player, see how that shows up on the screen. It's a test cd indeed with 1kHz at 0dB and at -10.
My 7L5 has a plugin with switchable 1M / 50 ohm input (and 75 too i think).
Also it can withstand some DC at the input. And yes, it certainly a relative measurement. No way this thing is still within calibration.
My 7L5 has a plugin with switchable 1M / 50 ohm input (and 75 too i think).
Also it can withstand some DC at the input. And yes, it certainly a relative measurement. No way this thing is still within calibration.
Hooked up a standard 371 (which comes with the not so standard cdm4/11 with hall btw) to the SA. Put the 1kHz tone at 0dB by working on the amplifier calibration and using some attenuators before the L3 input (which can do 600 Ohm and not 75 as i said earlier).
Anyway:
peak at 2kHz at -67dB and another one at 3kHz at -50dB.
Not shure what's going on yet, but i was indeed not measuring the DOS. Could be the test disk or the way i'm (mis)using the SA. This is for later.
Anyway:
peak at 2kHz at -67dB and another one at 3kHz at -50dB.
Not shure what's going on yet, but i was indeed not measuring the DOS. Could be the test disk or the way i'm (mis)using the SA. This is for later.
1Meg input is great, I have no analyzer at work that has it, very annoying. But -67dBc is not too bad, I have an Agilent 33250A generator here that does somewhere between -55 and -60dBc, so the player is better
What you could do is see if there's an increase in distortion with the DOS, then you know it's not adding anything to the numbers. You can also use a bandwidth filter of 20Hz highpass and 20kHz lowpass if the analyzer has it, it's a common range to specify THD in audio. This will lower the number even more.
Ray
What you could do is see if there's an increase in distortion with the DOS, then you know it's not adding anything to the numbers. You can also use a bandwidth filter of 20Hz highpass and 20kHz lowpass if the analyzer has it, it's a common range to specify THD in audio. This will lower the number even more.
Ray
Hi everybody...first time on this discussion so here's the question...
I bought this player a few years ago, and when I opened it recently I found something interesting...
Can someone tell me about this... I see it have a tube... is this normal or not?
http://img444.imageshack.us/img444/470/marantz63.jpg
I bought this player a few years ago, and when I opened it recently I found something interesting...
Can someone tell me about this... I see it have a tube... is this normal or not?
http://img444.imageshack.us/img444/470/marantz63.jpg
Hi everybody...first time on this discussion so here's the question...
I bought this player a few years ago, and when I opened it recently I found something interesting...
Can someone tell me about this... I see it have a tube... is this normal or not?
http://img444.imageshack.us/img444/470/marantz63.jpg
Depends on your point of view
It's not a standard 63, that's for shure. But if it says tjoeb or something on the front, then yes, it's normal Edit: seems tjoeb started with the cd38 and did not modify the 63. Maybe some other company or even home made.
Last edited:
As i was seeing some weird stuff (change the attenuator 1dB and the signal was going up 5) i took out the manuals. And my suspicion was right, i was f*cking up the measurement by overloading the input (with the 1kHz signal).
There is an easy trick on the analyser by using the input buffer. As this is attenuating with 8dB and then amplifying again with 8dB, the signal should not change (a lot) when pressing this button. Only the noisefloor changes.
If there is a change (as in my case), the input is overloaded. Duh..
So paying attention to this, i made some new measurements. Hardly any scientific value as i compared one modified channel with another unmodified channel. So i could well have been measuring differences between left and right circuits instead of the CFP.
One note upfront, the CFP version has a 1dB higher output value then the standard DOS. The standard DOS btw was equal to a standard philips player.
The measurement with 1kHz testtone, set at 0dB at the analyser:
__________________2kHz__________3kHz
371 standard player_-65dB__________-65dB
DOS______________-68dB__________-55dB
CFP______________-68dB__________-60dB
So some 5dB better at 3kHz. Does this allign with the theory ???
There is an easy trick on the analyser by using the input buffer. As this is attenuating with 8dB and then amplifying again with 8dB, the signal should not change (a lot) when pressing this button. Only the noisefloor changes.
If there is a change (as in my case), the input is overloaded. Duh..
So paying attention to this, i made some new measurements. Hardly any scientific value as i compared one modified channel with another unmodified channel. So i could well have been measuring differences between left and right circuits instead of the CFP.
One note upfront, the CFP version has a 1dB higher output value then the standard DOS. The standard DOS btw was equal to a standard philips player.
The measurement with 1kHz testtone, set at 0dB at the analyser:
__________________2kHz__________3kHz
371 standard player_-65dB__________-65dB
DOS______________-68dB__________-55dB
CFP______________-68dB__________-60dB
So some 5dB better at 3kHz. Does this allign with the theory ???
Last edited:
Hi Guys
I was down at the Heathrow enthusiasts HiFi show over the weekend Helping out Inspire HiFi with their turntable room - Inspire
They have a new 12" arm deck out that costs £5500 with SME V 12" arm. I also took down my old Marantz to see how it faired against this good turntable and other expensive equipment (like the £40K system 2 doors down). We used an ATC SIA-150 amp and ATC SCM40 speakers.
All weekend people were saying how good the sound was in the Inspire room (including HiFi World magazine) compared to much more expensive setups. Later on saturday and sunday I fired up the 63...... basically people could not believe it was a cd player (some of which hate cd with a passion, customers were saying how analogue it was, two guys said it had the best sound stage they had heard all day (that pleased me lol). It outperformed the Inspire 12" deck and Whest £3K phono stage (totaling nearly £8K) in detail, bass and attack. The other cd players we listened to at the show that ranged from £10K to £18K and on much more expensive setups sounded detailed but non of them had that analogue sound I or the turntables were producing, they left me very cold.
Basically I am of course buzzing with the player etc but it also shows where we are all heading. Some of the more modded cd players on here are probably punching much higher up in the audiophile range than they think.
Brent
I was down at the Heathrow enthusiasts HiFi show over the weekend Helping out Inspire HiFi with their turntable room - Inspire
They have a new 12" arm deck out that costs £5500 with SME V 12" arm. I also took down my old Marantz to see how it faired against this good turntable and other expensive equipment (like the £40K system 2 doors down). We used an ATC SIA-150 amp and ATC SCM40 speakers.
All weekend people were saying how good the sound was in the Inspire room (including HiFi World magazine) compared to much more expensive setups. Later on saturday and sunday I fired up the 63...... basically people could not believe it was a cd player (some of which hate cd with a passion, customers were saying how analogue it was, two guys said it had the best sound stage they had heard all day (that pleased me lol). It outperformed the Inspire 12" deck and Whest £3K phono stage (totaling nearly £8K) in detail, bass and attack. The other cd players we listened to at the show that ranged from £10K to £18K and on much more expensive setups sounded detailed but non of them had that analogue sound I or the turntables were producing, they left me very cold.
Basically I am of course buzzing with the player etc but it also shows where we are all heading. Some of the more modded cd players on here are probably punching much higher up in the audiophile range than they think.
Brent
Last edited:
