Dvv . I think your pre-amp is rather wonderful . It is stuff I have been banging on about for years ( and is all talk as I haven't yet tried the ideas ) . I copied the bit that I think needs looking at .
The volume pot buffer seems a bit OTT . People argue that this primitive version is asymmetrical ( 60/40 = second harmonic predominance ) . I rather like it . It is outrageous that they did this when the rest is strictly modern , well done . Strictly 1965 ( Sugden A21 and similar ) . It is possible to work correctly without it , it might be better ( more detailed ) . I will build this circuit as I have been searching for an idea like this . I should have thought of it myself . So people know I only like circuits that look too simple to work . The jet engine comes to mind or better still the rocket engine . So respect for this circuit . The bias might stand a bit of class A , 39R emitter resistors should prevent silly things happening . A simple resistor should be enough and should be thermally stable if not too adventurous .
One thing you could try instantly is SE class A output , I choose to favour the NPN device . Find out with fingers how far you might go , hold on hard to be sure it isn't at 90 C . Fit a pot with stop resistor and listen . Being that it drives headphones it should allow plenty of current . . The servo action of the op amp will adjust for it . You might increase C901 56 pF as at 284 kHz if my maths right it could drop to 80 kHz ( + 142 pF = 150 pF ) and win more than it looses ( input bandwidth , passive filter ) . I bet anything changing the input cap ( 470 nF ) is at best 10% of any improvement . Do it last .
I suspect this was made by Marantz Japan . Philips bought Marantz as a way to get components at the right price . As much as I am not Ken Ishiwata's greatest admirer he did speak up when Philips came in . His vision was to restore the brand and in the style of the old company . Very sad he didn't contact Sid Smith . Sid would have helped . And guess what , he would have made this pre-amp work . Sid told me just before he died he wished people would buy and upgrade his transistor amps . He thought the components alone accounted for valve better than transistor . I had the great pleasure of being taught load-line graphs by him . I knew it but it was 30 years since shown any . I still am not very good at doing distortion that way . Sid said just look at the intervals between curves . It says exactly how it will sound .
The other sections of the preamp are generic . Nice to see headphones driven via the output .
It occurs to me that late in the day someone thought that the preamp sounded a bit lame and shrill ( perhaps try 2 x 4p7 NPO VAS Caps to R912/13 and basses , bet that works ? ) . This would have broken the heart of an engineer to think his adventure in discreet op amps was worse than the house NE5534 ( bet it is ) . So the learned boss says . Why not fit a buffer with a bit of colour ? I bet I am right ? My eyes tell me that the gain is too low for this design to be completely stable . I would say like an FI car this needs setting up as it's only problem . Typical Japanese design I have to say , always in a rush .
Try also a 10R + 100 nF output Zobel . It should be OK and might help . Like a buffet meal you might not want all .
You know all the tweaking stuff better than me so I won't say about PSU upgrades .
Footnote . It is possible even with small output stages you could use more driving current . It seems a bit ambitious the way it works . It might even be class B or near enough ? Perhaps R916/7 510 R could be tweaked ? Also C904/5 15 pF suggests they had problems . I never in my life needed to do this when this simple current amplifier used ( not even with CfbP ) . My brother would have said " a bit of a dogs breakfast " . Possibly it was done as standard practice so forgive me if wrong .
Last edited:
Many thanks for taking the time and trouble, Nige. Sonehow, I thought you would.
Where do you see a servo? I would have liked to have seen one, but none were to be found.
Regarding Marantz - Philips acquired Marantz through a complex deal with Sony related to the CD format, but if memory serves, than was in '82 or '83, and this baby came out in 1979. Since most of the transistors used were made by Hitachi, I assume it was made by Hitachi, since it's standard Japanese practice to use one's own semiconductors whenever possible, as much as possible. Furthermore, big companies like to talk to other big companies, and Hitachi Industries is a hell of a lot bigger than the then ailing Marantz, which was very much in the dark after being owned by Sony for years. Sony had no idea what to do with Marantz and just let it linger on, so passing over to Philips was a godsend for Marantz.
Speaking of which, and mentuioning Sid Smith, I can always fall back to my Marantz 3250 B preamp, the natural companion to their 170 DC power amp.. It is in perfect working order and has been erefreshed in full, new caps all over. You the one, it had an MC pre-preamp consisting of exactly one (1) FET per channel, rather old fashoned, but nevertheless capable of very satisfying sound.
Or use the Citation 21 preamp, also refreshed and in full working order, but the fate of my system still lies with the Luxman C-03 preamp, which is the best of the lot, but which is on hold currently, awaiting its own refreshing, which will be the most detailed and highest quality refresh I have ever done. I think it's that good, it deserves the best I can do. No wonder it was made for 9 years, being Luxman's greatest hit ever, and probably being the longest life preamp on the world market ever. Trust me, this didn't happen by sheer luck, they really nailed that one down, I'll send you the schematics, you will love it.
So, it's not like I'm pressed to make the Philips sound right, it's just that I have sentimetal ties to Philips, and like you I think it can do MUCH better.
Where do you see a servo? I would have liked to have seen one, but none were to be found.
Regarding Marantz - Philips acquired Marantz through a complex deal with Sony related to the CD format, but if memory serves, than was in '82 or '83, and this baby came out in 1979. Since most of the transistors used were made by Hitachi, I assume it was made by Hitachi, since it's standard Japanese practice to use one's own semiconductors whenever possible, as much as possible. Furthermore, big companies like to talk to other big companies, and Hitachi Industries is a hell of a lot bigger than the then ailing Marantz, which was very much in the dark after being owned by Sony for years. Sony had no idea what to do with Marantz and just let it linger on, so passing over to Philips was a godsend for Marantz.
Speaking of which, and mentuioning Sid Smith, I can always fall back to my Marantz 3250 B preamp, the natural companion to their 170 DC power amp.. It is in perfect working order and has been erefreshed in full, new caps all over. You the one, it had an MC pre-preamp consisting of exactly one (1) FET per channel, rather old fashoned, but nevertheless capable of very satisfying sound.
Or use the Citation 21 preamp, also refreshed and in full working order, but the fate of my system still lies with the Luxman C-03 preamp, which is the best of the lot, but which is on hold currently, awaiting its own refreshing, which will be the most detailed and highest quality refresh I have ever done. I think it's that good, it deserves the best I can do. No wonder it was made for 9 years, being Luxman's greatest hit ever, and probably being the longest life preamp on the world market ever. Trust me, this didn't happen by sheer luck, they really nailed that one down, I'll send you the schematics, you will love it.
So, it's not like I'm pressed to make the Philips sound right, it's just that I have sentimetal ties to Philips, and like you I think it can do MUCH better.
All LTP op amps use servo action . The servo you are thinking of is to enhance it . Servo from French means brain . So we all misuse the word. What I mean is the op amp action will fight very hard to make the input reference voltage on No 1 input exactly as input No2 . This is why I like feedback loops with a feedback capacitor . This gives the input of No 2 100 % DC feedback which is ideal . No bad thing if it does that , you win more than you loose . Try non polar cap and be honest , does it sound good ? Some come up with complex ideas like it is better because we like distortion . No it isn't . Used at low RMS non polar caps have about - 150 dB distortion . They cost peanuts . Even losses are not really an issues , a 100 nF 100 V bypass if you like ( EPCOS 7.5 mm spacing ) . 63V polyester has marginally more distortion than non polar's when used here . Mostly the 250 V polyester is the device of choice over most ( inc polypropylene due to it's memory effect where relevant , i.e. 200 V blocking / coupling ) .
Last edited:
I think its fantastic , perfect for my simulator ....
Ahh c'mon , your an old x, ruskan , tuff as nails ....
and i never heard anything phillips that wasnt gritty ....I think its fantastic , perfect for my simulator ....
Ahh c'mon , your an old x, ruskan , tuff as nails ....
That's just your bad luck. Much of their stuff sounds good to very good, but somehow they never make excellent.
The said tuner, AH180, is a good example. Clean and clear, very lively, very dynamic, it makes the very good list hands down, but misses the excellent mark by a nick. It cannot match my reVox B760 for absoute fidelity and neutrality, but then the reVox did cost approximately twice as much at the time.
The literally only irritating thing about the Philips tuner is the fact that if you want it to remember the memorized stations even if there's no power from the grid, you had to buy an external add-on box (?!?), which is about as stupid an idea as I can think of.
Fortuately, brownouts do no harm, and I have prescious few blackouts.
Last edited:
Dvv . I think your pre-amp is rather wonderful . It is stuff I have been banging on about for years ( and is all talk as I haven't yet tried the ideas ) . I copied the bit that I think needs looking at .
The circuit is a bipolar variant of John Curl's full complementary circuit. I think others have used it as well. I don't know who was first. When working right it should have virtually no higher order harmonics. The buffer on the volume control is odd and will compromise everything else. Bypassing it or converting to Jfets will make a big difference there. The bipolars in that form will generate distortion at the inputs which will be highest at the -6 dB point on the volume. There is also the issue of so many caps to make it work.
@Demian,
I have found on various places that James Bongiorno is credited with the first fully complementary circuit, and later on, with the first DC servo circuit. I don't know if that is true.
I have no doubt John Curl has used it as well, as have many, many others, in various forms and for various purposes. For example, an old amp I used to have, made by the German company Koerting, from 1981, also used it, as did Dual in their top of the line CV 1700 integrated amps, W. Marshall Leach, Dan d'Agostino, and so forth through a long list.
On the other hand, while quite popular for power amps, I haven't seen that many preamps using the topology, and even fewer of them using as elaborate topology as this preamp.
Just for the record, my own headphone amps/preamps, the BJT and the FET/MOSFET versions, also use this topology with DC servos. They were introduced in late 2009/early 2010.
Shockingly, the tube version does not.
My point is, I am no newbe for that topology.
Specifically, Nigel said it, you said it and I tend to agree, that buffer seems suspect to me too. I back this opinion by the fact that the phono RIAA section uses exactly the same topology (some components change values, but otherwise identical), and when the resulting LP sound is taken out of the preamp so it doesn't pass through the line amp shown, the sound is well above the average. This leads me to believe it's not the basic topology that's at fault, leaving the buffer as the prime suspect.
So, I either:
1. Bypass it,
2, Perhaps change the trasistors zesd and their biasing scheme, or
3. As you say, change them to FETs.
I have found on various places that James Bongiorno is credited with the first fully complementary circuit, and later on, with the first DC servo circuit. I don't know if that is true.
I have no doubt John Curl has used it as well, as have many, many others, in various forms and for various purposes. For example, an old amp I used to have, made by the German company Koerting, from 1981, also used it, as did Dual in their top of the line CV 1700 integrated amps, W. Marshall Leach, Dan d'Agostino, and so forth through a long list.
On the other hand, while quite popular for power amps, I haven't seen that many preamps using the topology, and even fewer of them using as elaborate topology as this preamp.
Just for the record, my own headphone amps/preamps, the BJT and the FET/MOSFET versions, also use this topology with DC servos. They were introduced in late 2009/early 2010.
Shockingly, the tube version does not.
My point is, I am no newbe for that topology.
Specifically, Nigel said it, you said it and I tend to agree, that buffer seems suspect to me too. I back this opinion by the fact that the phono RIAA section uses exactly the same topology (some components change values, but otherwise identical), and when the resulting LP sound is taken out of the preamp so it doesn't pass through the line amp shown, the sound is well above the average. This leads me to believe it's not the basic topology that's at fault, leaving the buffer as the prime suspect.
So, I either:
1. Bypass it,
2, Perhaps change the trasistors zesd and their biasing scheme, or
3. As you say, change them to FETs.
I think the buffer is not running hot enough , it is a wonderful concept ( stress concept ) . Try output class A as it is a 2 minute job . Try 6K8 I think .
Will be interested where this goes . I understand keeping it original .
Next tackle the Sony STVF5L . Bloody awful . STVF3/4 were excellent . Poor souls who spent extra . Suggestion was the boss did the 5 and the kids the 3/4 . Same era . Japanese hi fi hey .
I have J B phone number in my phone book . Sid's mate . Never phoned him as was too scared . Idiot me .
This Philips circuit is a variation of symmetrical complimentary . It manages to get more out of it by the arrangement of LTP's . It is special . Like some engines special isn't always good . By that I mean the wrinkles are not ironed out . Just imagine making an op amp . No room for error .
Will be interested where this goes . I understand keeping it original .
Next tackle the Sony STVF5L . Bloody awful . STVF3/4 were excellent . Poor souls who spent extra . Suggestion was the boss did the 5 and the kids the 3/4 . Same era . Japanese hi fi hey .
I have J B phone number in my phone book . Sid's mate . Never phoned him as was too scared . Idiot me .
This Philips circuit is a variation of symmetrical complimentary . It manages to get more out of it by the arrangement of LTP's . It is special . Like some engines special isn't always good . By that I mean the wrinkles are not ironed out . Just imagine making an op amp . No room for error .
Last edited:
Ok, agree, weird group, them all ..........
Watching this FMA video , reminded me of seeing Manny in the early's 90's , most if not all of us were doing nothing but Digital and we had just received the top of the line 2 box Wadia and LP's were the furthest thing from my mind. But there was Manny, fighting the trend and demoing with LP's, i still remember thinking how odd it was he was still doing so ....
Ahhh, he wasn't fooled like the rest of us and is still staying the course..
HiFi Deluxe Vids 1
Watching this FMA video , reminded me of seeing Manny in the early's 90's , most if not all of us were doing nothing but Digital and we had just received the top of the line 2 box Wadia and LP's were the furthest thing from my mind. But there was Manny, fighting the trend and demoing with LP's, i still remember thinking how odd it was he was still doing so ....
Ahhh, he wasn't fooled like the rest of us and is still staying the course..
HiFi Deluxe Vids 1
D,
Why the centurion need Cojones ....
Heavy Load: How Loudspeakers Torture Amplifiers Page 2 | Stereophile.com
Why the centurion need Cojones ....
Heavy Load: How Loudspeakers Torture Amplifiers Page 2 | Stereophile.com
D,
Why the centurion need Cojones ....
Heavy Load: How Loudspeakers Torture Amplifiers Page 2 | Stereophile.com
It needs a weighting curve to be useful .
Folk
Jazz
Techno
Gradient coils
The latter is for body scanners and they kill amps .
D,
Why the centurion need Cojones ....
Heavy Load: How Loudspeakers Torture Amplifiers Page 2 | Stereophile.com
FYI, my Citation 24 amp has two power trannie pairs per channel, each rated at 150W. That doesn't stop it from pumping out just short of 600W into 2 Ohms in peaks (t=20 mS).
By comparison, I use 3 pairs, each rated at 200W. 24 has a theoretical dissipation capability of (4x150W) 600W, mine has (6x200) 1,200W.
Much depends on subsidiary requirements. I am not interested in building lab instruments, so long term power output into say 2 Ohms is of no interest to me because I assume whoever is using it will not listen to full sine wave power output into low impedances.
I am also not interested in tremendous power capabilities on a continuous basis. My view is that we need to always have around 12 dB of dynamic power reserve for those transients above our average dissipated power. High load tolerabnce is also a must.
So, this is where I get ambitious. Nominally, it must deliver 28.3 Vrms into both 8 and 4 Ohms, and must lose little into 3 and 2 Ohms, but in peaks. Furthermore, it must be able to deliver around 170W into 8 Ohms in peaks, which is why it has relatively high power supply lines. And why the input stage, VAS and predriver are ran off regulated power supplies at +/- 61.5V, while the driver and output stage are ran off separate supply lines and +/- 56.5V.
On basis of that, I expect it to comply in full to the IHF standards, and then wlak another mile or so. Experience teaches me that many users will never need that power, but it must be there neverteless, just on the off chance.
I do NOT see you as one of the users because your requirements are far too specific, so you don't count. For like 99% of users, that will be quite enough, assuming I get it right and that I get it as I want it to be. There is quite a list of requirements it must satisfy, but then, it's the development stage, from a fully functional electrical circuit to a good sounding amp, that's always the most interesting, that is the REAL challenge.
- Status
- Not open for further replies.