Cambridge Audio Azur 640

[audio_builder]

Metalized capacitors in the signal path do not sound good. By replacing the blue caps with styrens, a lot of the harshness and high frequency glare is gone. Music sounds much more natural. I still have quite a few mods to go, but I am making progress.

[bigalyourpal1]

Hi Audio Builder, are you absolutely sure in your heart of hearts changing the blue caps to polystyrene is an improvement???I am double checking because every time I have tried improving "any" of the metalised caps on the circuit board: I have been thoroughly disappointed by the results. I realise I am not agreeing with people before me, but I find changing any of the coloured capacitors to alternatives "kills" the "thrill" of the amp. May be a little better detail but a dull lack lustre boring sound.I end up going back to the original set up. I have had the same experience with the op amp caps....yes the ceramic are not so hot on the high frequencies but so far as the "overall listen is concerned" they are better than the silver mica or polystyrene. I am wondering if the only worth while mods are the power supply, signal resistors, and electrolyics......

[audio_builder]

Audio is like food (or a box of chocolates ). Some people like a little salt, some a little pepper, and a surprising number of people like MSG.

For most systems, the additional harshness and glare metalized caps provide may be a welcome addition. The type of music you listen to may also benefit from what a Metalized capacitor adds.

In my heart of hearts I prefer no capacitor, but if one needs to be use I like the lack of coloration polystyrenes offer even over a Teflon Cap.

When it comes to mods, signal resistors rank a lot lower in overall sound contribution compared to the actual circuit, op-amps, and caps in the signal path. Weather spending over 50% of a piece of gear’s value in mods is worth it is always questionable, but it sure can be enjoyable and educational.

Happy New Year!

[dokblue]

Need a little Help here after reading or Wading through this Entire thread I am going to try this mod. However the Blackgate caps most of which are No Longer Available! What can I use in their place and keep the stock enclosure? I have all the Panasonics.

[kcc123]

I think Panasonics are quite good sounding and so are the Elna Silmic and Nichicon Muse caps.

[dokblue]

Thanks kcc123!! Can I ask what exactly you used at C60,C62, and the other side of the mm and mc sections?

[kcc123]

They were replaced with Blackgate NX HIQ 0.1uF/50V capacitors.

[kcc123]

Since more than a year ago I have started modifying the 640P with many boutique parts and I have been very pleased with the results, though at a very high cost and the expense of my precious time and great efforts. I have a list of components for the modifications that I have already carried out:-

5 x Rubycon 25ZL 1000 caps (C42, C43, C47, C48, C55 -1000uF/25V)
4 x 25ZL 470 (C57, C65, C83, C84 -470uF/25V)
3 x 25ZA100 (C9, C10, C94 - 100uF/25V)
2 x bi-polar Blackgate NXHiQ 6.3V/47uF (C31, C64, original -16V/47uF)
2 x 35ZL1000 (C2, C4 -1000uF/35V)
4 x 35ZLH330 (C1, C3, C5, C6 - 50V/330uF)
1 x 25ZA10 (C11 – 10uF/25V)
2 x bi-polar Blackgate NXHiQ 6.3V/0.47uF capacitors configured as 1 Super E-Cap (C95 - 50V/1uF)
7x Schottky Barrier Rectifiers 11DQ10 (D1, D2, D3, D4, D5, D6, D7, D8, D13 IN4004 diodes)
4 x Blackgate NX 6.3V/220uF (C44, C56, C49, C61 - 220uF/6.3V)
4 x Blackgate NX HIQ 0.1uF/50V (C60, C68, C30, C62 - 100nF/63V)
2 x Blackgate N 6.3V/470uF (C88, C89 - 470uF/6.3V)
2 x Blackgate NX HIQ 22uF/6.3V (C90, C91 - 470uF/6.3V)
2 x Jensen paper in oil 0.012uF/400V (C76, C77 – 10nF/63V polyester caps)
2 x Polystyrene capacitors (C75, C35 -33pF/50V ceramic)
2 x silver mica (C92, C93 47pF/50V ceramic)
4 x 0.5W Audio Note Tantalum resistors (R50, R75, R25, R11 -100ohms)
2 x 0.5W Audio Note 2K2 (R26, R51)
2 x 0.5W Audio Note 47R (R5, R21)
2 x 0.5W Audio Note 3K9 (R14, R22 - 3.92K)
2 x 0.5W Shinkoh Tantalum resistors (R10, R13 - 75K)
2 x 0.5W Shinkoh 1K8 (R7, R24 - 1.87K)
2 x 0.5W Audio Note (R76, R78 - 47K)
2 x 0.5W Shinkoh (R15, R16 - 130 R)
2 x 0.5W Shinkoh (R17, R18 – 51 K: MM loading)
2 x 0.5W Shinkoh (R68, R85 – 560 K: MM loading)
2 x 0.5W Shinkoh (R15, R16- 130 ohms MC loading)
2 x 0.5W Audio Note (R60, R77- 470 ohms: MC loading)
2 x 0.5W Audio Note (R76, R78 - 47Kohms)
2 x 0.5W Shinkoh 9.1 Ohms - R82, R49 (leave R54, R71 empty - 18 Ohms) = +6dB output
2 x 0.5W Audio Note 47 R (R62, R79 - 91 Ohms) = +6dB output
4 x 0.5W Audio Note (R58, R60, R66, R77 - 470 R)
2 x 0.5W Audio Note (R59, R67 - 1.2k)
1 x 12V toroidal transformer
2 x LM4562, 1 x OPA 2111, 2 x OPA 637, (U3, U4, U5, U6 - 4 x NE5532 op-amps)
Blackgate NX HIQ 0.1uF/50V cap soldered across the pin 4 and pin 8 of each of the four op-amps.
Removed four brown caps (C14, C15, C28, C28) and by-passed the subsonic filter selector with 2 short solid silver wires connected between the legs.

It is only a guide line for those who are interested and any preferred components can be substituted as required.

[Fatmangolf]

This photo of the 640P PCB may be helpful. It has labels describing the caps and IC's that have been discussed in this forum.



Not shown here are the 0.1uF power supply filter caps from pins 4 to 8 under each op amp IC.

I agree with an earlier posting about using capacitors from the Michicon Muse ranges. You can see the green ES bipolars in the signal decoupling and black KZ (and a few Rubicons) in the power supply filtering.

[Fatmangolf]

PatOMalley is right, those caps block dc through the cartridge. If one were to short them out, I reckon there would be a few uA of dc in the cartridge coils. Can't advocate doing this, and I don't have a MC set up at this time to try, but that doesn't seem too bad to me. That would place about the same dc power as signal power in the coils, I reckon, FWIW.

In my 640P there is a DC voltage of a few millivolts on bases of the transistors which would mean a current through the cartridge if the decoupling capacitors were removed/shorted out. I suspect the resulting current through the coils would be bad for the cartridge and could affect the response of the transistors in the input gain stage. (The 540P didn't have an input decoupling capacitor which meant about 1mV across the cartridge coils and this voltage varied depending on the cartridge's internal resistance.)

Also if you want to experiment with the resistive loading on the cartridge, e.g trying less than 47k MM or 470R MC, the decoupling capacitor will ensure this does not affect the operation of the gain stage.

[Ldg]

PatOMalley is right, those caps block dc through the cartridge. If one were to short them out, I reckon there would be a few uA of dc in the cartridge coils. Can't advocate doing this, and I don't have a MC set up at this time to try, but that doesn't seem too bad to me. That would place about the same dc power as signal power in the coils, I reckon, FWIW.

In my 640P there is a DC voltage of a few millivolts on bases of the transistors which would mean a current through the cartridge if the decoupling capacitors were removed/shorted out. I suspect the resulting current through the coils would be bad for the cartridge and could affect the response of the transistors in the input gain stage. (The 540P didn't have an input decoupling capacitor which meant about 1mV across the cartridge coils and this voltage varied depending on the cartridge's internal resistance.)

Also if you want to experiment with the resistive loading on the cartridge, e.g trying less than 47k MM or 470R MC, the decoupling capacitor will ensure this does not affect the operation of the gain stage.

Yes, it would suit someone brave enough to try it, but a few uA between friends might just be fine. There is potentially an upside, of course ! Could simulate it with a resistive load equivalent to the cartridge, and measure the current (v across resistor) first ?

[PhP]

Since more than a year ago I have started modifying the 640P with many boutique parts and I have been very pleased with the results, though at a very high cost and the expense of my precious time and great efforts. I have a list of components for the modifications that I have already carried out:-

5 x Rubycon 25ZL 1000 caps (C42, C43, C47, C48, C55 -1000uF/25V)
4 x 25ZL 470 (C57, C65, C83, C84 -470uF/25V)
3 x 25ZA100 (C9, C10, C94 - 100uF/25V)
2 x bi-polar Blackgate NXHiQ 6.3V/47uF (C31, C64, original -16V/47uF)
2 x 35ZL1000 (C2, C4 -1000uF/35V)
4 x 35ZLH330 (C1, C3, C5, C6 - 50V/330uF)
1 x 25ZA10 (C11 – 10uF/25V)
2 x bi-polar Blackgate NXHiQ 6.3V/0.47uF capacitors configured as 1 Super E-Cap (C95 - 50V/1uF)
7x Schottky Barrier Rectifiers 11DQ10 (D1, D2, D3, D4, D5, D6, D7, D8, D13 IN4004 diodes)
4 x Blackgate NX 6.3V/220uF (C44, C56, C49, C61 - 220uF/6.3V)
4 x Blackgate NX HIQ 0.1uF/50V (C60, C68, C30, C62 - 100nF/63V)
2 x Blackgate N 6.3V/470uF (C88, C89 - 470uF/6.3V)
2 x Blackgate NX HIQ 22uF/6.3V (C90, C91 - 470uF/6.3V)
2 x Jensen paper in oil 0.012uF/400V (C76, C77 – 10nF/63V polyester caps)
2 x Polystyrene capacitors (C75, C35 -33pF/50V ceramic)
2 x silver mica (C92, C93 47pF/50V ceramic)
4 x 0.5W Audio Note Tantalum resistors (R50, R75, R25, R11 -100ohms)
2 x 0.5W Audio Note 2K2 (R26, R51)
2 x 0.5W Audio Note 47R (R5, R21)
2 x 0.5W Audio Note 3K9 (R14, R22 - 3.92K)
2 x 0.5W Shinkoh Tantalum resistors (R10, R13 - 75K)
2 x 0.5W Shinkoh 1K8 (R7, R24 - 1.87K)
2 x 0.5W Audio Note (R76, R78 - 47K)
2 x 0.5W Shinkoh (R15, R16 - 130 R)
2 x 0.5W Shinkoh (R17, R18 – 51 K: MM loading)
2 x 0.5W Shinkoh (R68, R85 – 560 K: MM loading)
2 x 0.5W Shinkoh (R15, R16- 130 ohms MC loading)
2 x 0.5W Audio Note (R60, R77- 470 ohms: MC loading)
2 x 0.5W Audio Note (R76, R78 - 47Kohms)
2 x 0.5W Shinkoh 9.1 Ohms - R82, R49 (leave R54, R71 empty - 18 Ohms) = +6dB output
2 x 0.5W Audio Note 47 R (R62, R79 - 91 Ohms) = +6dB output
4 x 0.5W Audio Note (R58, R60, R66, R77 - 470 R)
2 x 0.5W Audio Note (R59, R67 - 1.2k)
1 x 12V toroidal transformer
2 x LM4562, 1 x OPA 2111, 2 x OPA 637, (U3, U4, U5, U6 - 4 x NE5532 op-amps)
Blackgate NX HIQ 0.1uF/50V cap soldered across the pin 4 and pin 8 of each of the four op-amps.
Removed four brown caps (C14, C15, C28, C28) and by-passed the subsonic filter selector with 2 short solid silver wires connected between the legs.

It is only a guide line for those who are interested and any preferred components can be substituted as required.

Thank you very much KCC123 for the list I did asked you. Certainly very appreciated from all CA 640P owners! So far I have replaced all (except one small electrolytic) in the PS with Nichicon FC,FM and For Audio, and with one small 1uf 50v Elna Silmic II caps. Did also put Elna Silmic II in place of the caps before the 4 brown polypro (?) capacitors and in C94, C9 and C10 (same value but these Elna are way bigger caps...). Did put Nichicon For Audio in the MC (C42,43) and Nichicon Muse KZ (2.5 times bigger than the Nichicon For Audio) in the MM sections (C47,48). That is it for now until I receive my Blackgate nx and some resistors. Work in progress... Advise to all: take your time to use a lime to reduce the diameter of those big Nichicon and Elna caps, worth it if you don't want to get mad...the holes are small...
Can't comment on sound now since I have to put a HOMC cartridge to compare with what I had before on my tonearm. For now a little Sumiko Oyster MM cart sound good. Need more burn in time...and will see with the HOMC and a LOMC to compare with what I do remember with stock 640P.
Phil

[stumpy57]

hello,
I took a look to the shematics of the 640P and I saw that there are two
gain stages, one with dif cascode transitors and the other with an AOP.
If I want to increase de gain by 10db on the MM input what do you mean is the better way ?
Increase the gain of the input stage or the AOP stage or both ?
What is the advantage to increase the gain of the cascode stage ?

Thanks

Alex

[dokblue]

Thanks SO much kc123 for the List and all the Great pi`s and help. I am just starting this upgrade now that I Filayy have most of the parts and can`t wait to hear the difference!!