Modding Azur 640A

Guyz,

I've just opened the hood of my Cambridge Audio Azur 640A and find out that it only has 4400 uF per power supply rail (consist of 2 x 2200 uF/50v in parallel, means for all channel it has 8 pieces of 2200 uF). From what I've read, it is not at all sufficient for 640A's 65 watt power. No wonder if the bass is a bit weak. If I remember correctly the "rule of thumb", my 640a need at least around 6500 uF.

Question :

What will be the best way to upgrade this : for each rail :replace with 2 x 3300 uF / 4700 uF, or 1 x 6700 uF, or even jump to 10000 uF.

What will be the sonic difference if I arrange 2 caps in parallel than using 1 big cap with approximately same total value ?

Is it good to use BC components (type 037/056/057/154/114), caps for this upgrade ? This is the only brand I can get easily in France for almost all range of capacity & voltage with reasonable price. I lived in Indonesia before, and it was easier and CHEAPER to buy audio grade caps.....(and everything else off course, nothing is cheap around here)

Does somebody knows if Farnell France receiving oder in small quantity for poor people like us ? I just found some interesting Panasonic and Rifa on their catalog.

I'm new to this DIY audio stuff, so help me please....
 
I just have that amp on modification. When I take a good look on topology of whole amp , I realize that there is much to be modified. PS of output stage is quite miniature. That is not dual mono design , so all caps(4per rail) are in paralel so you have about 8800uF per rail , for both channels. The mayor problem is poor quality of these caps (and all others on PCB) so I will replace them with Panasonis 3300uF ultra low ESR , high current type. Next I will replace all diodes i PS with Schottky types. Than i will clear the signal path , and i will use higher quality OP-s for gain stage and tone controls , bypass all poor coupling caps and input buffer stage. Orginal OP-s are NE5532 in all positions. Decoupling caps are some poor bipolar electrolitics. It is the miracle that this amp have a little listenable sound. :eek:
 
Gasho:
You are right.... A poor component can do no good for the signal!
However..... How big improvement are guaranteed using million $ caps and opamps in a bad circuit (not saying that the circuit in the Cambridge is bad, - just a standard Taiwanese circuit):D


i2k92:
Cambridge (like other similar brands) does have a big lack in the quality of components used, however you will come a long way by increasing the PSU to 2x10.000uF and by shortening (if possible) or replace the DC-blocking caps in the signal path with bigger ones.....
I have notised that in many amps like Cambridge there are used caps in the nF range as bypass caps.... Replace these with 1-4.2 uF film caps (Evox, Rifa or Wima etc.) and you wont believe your ears ;)

My old modified QUAD 34 preamp had a frq. range 10-25.000 Hz or so before the mod.... Now it is dead flat from 2-300.000 Hz just by shortening/replacing caps in the signal path etc. ;)
 
Gasho said:
That is not dual mono design , so all caps(4per rail) are in paralel so you have about 8800uF per rail , for both channels. Orginal OP-s are NE5532 in all positions. Decoupling caps are some poor bipolar electrolitics.


Thanks for the correction and mod ideas.

I looking forward to replace the caps with 1 x 12000 uF per rail, as I'm not sure if I will have enough space for 8 x 3300 uF. I'm thinking about Panasonic GBA or BC component 154. What type of Panasonic you will use ?

About the opamps, are you sure the it uses 5535 on the gain stage ? I think it uses 5534.

And what do you think about the potentiometers ? I think they're bit nasty. I'm going to replace the vol and balance control. I will leave the tone control alone, I never use them anyway.

Sorry for asking too much....:clown:
 
i2k92 said:



Thanks for the correction and mod ideas.

I looking forward to replace the caps with 1 x 12000 uF per rail, as I'm not sure if I will have enough space for 8 x 3300 uF. I'm thinking about Panasonic GBA or BC component 154. What type of Panasonic you will use ?

About the opamps, are you sure the it uses 5535 on the gain stage ? I think it uses 5534.

And what do you think about the potentiometers ? I think they're bit nasty. I'm going to replace the vol and balance control. I will leave the tone control alone, I never use them anyway.

Sorry for asking too much....:clown:


I Was talking about NE 5532 which is double OP , and I am pretty shure that all OP-s are the same:cool:

Panasonic FC series.

I think it is too much problems around changes of pots. There is a space problem and holes on PCB are done for only this pot.
If you want to change this one with better like Alps "Blue velvet" motorised , you probably need to drill or cut something:dead:
 
Originally posted by ACD
Cambridge (like other similar brands) does have a big lack in the quality of components used, however you will come a long way by increasing the PSU to 2x10.000uF and by shortening (if possible) or replace the DC-blocking caps in the signal path with bigger ones.....
I have notised that in many amps like Cambridge there are used caps in the nF range as bypass caps.... Replace these with 1-4.2 uF film caps (Evox, Rifa or Wima etc.) and you wont believe your ears ;)
[/B]

Thanks,

Can you confirm this statement that I got over the net today :

The capacitors in the power supply has its own inductance. An engineer might decide to build a power supply with a single 50,000 micro farad capacitor. However, it is more likely that using five 10,000 micro farad capacitors will have less inductance and give faster current response.

How about replace the DC blocking caps with Audyn 2.2 uF MKP Plus Metallized Foil ? I think that the price is reasonable enough.

BTW, I attached the 640A schema in case somebody interested to it and perhaps will give me some more input.

thanks...
 

Attachments

  • azur640asch.jpg
    azur640asch.jpg
    88.8 KB · Views: 3,897
i2k92:
About capacitors in the PSU there are both inductance and resistance to worry about ;)
I would personally prefer the solution with 5x10.000uF and with a small film cap in parallel with each of the beer cans ;)

The schematic shows that you only got DC blocking caps in the input to the power section...

Otherwise the schematic shows the use of the good Sanken output devices with build-in diode.... I'm surpriced :confused:
 
OK, I'm a total noob as fas a electronics go, However, I would like to improve my cambridge audio 640A by replacing some caps.

Is there a picture guide on doing something like that?

From the posts above I understood , the biggest improvement would be to replace the power amp input caps with some better film caps...

By noob, I mean really noob, Couldn't tell a resistor or capacitor apart....
 
With no expertise, you are really best leaving well alone or you will end up with a dead amp. Sorry to be so negative about it but I have seen plenty of people kill their gear by attempting mods they read about on the internet, but they had no experience of implementing them.
 
I have the Azur 640A V2.0 and it seems that it clips easily :mad: :mad:
whenever I try to drive it volume pass 11 o'clock even I disable the clip protection cricuit (CAP5) during power up :confused: , and I can't play me Kitaro and Eagle's Live CD satisfy :bawling: :bawling: .

FYI, I have a pair of Mission 753 Freedom connected at the output. :)

Is there anyone who had try to modify the protection circuit to make it less sensitive or best still, to totally dissable it. :D :D

I need the protection section circuit and if anyone have, can please email to me at tayalex@tm.net.my.

Many thanks in advance.

:)
 
Quote

" Why do you think there is a protection circuit ?? "

From the WEB at Cambridge Audio and the given customer operation manual, its said there is a build-in protection circuit called "CAP5" and it also taught on how to disable it.

During driven hard, you will be able to hear the relay to speaker output clips, there no sound from the speakers, and the red LED on the front panel blinks.

I played with the source direct button ON and even that it clips.

You will have to power OFF the set and restarts it again inorder to continue playing.

;) ;)

Quote

" You might just be trying to push the amp too hard."

NO .:) :) NO .:) :)

I have a Rotel RC-980BX pre-out connected to an Acram 8P ;) and I try to play the same CD's with the same sound level but there's any problems.

BTW the Acram 8P only rated 50W output but the CA 640A V2.0 had 70W. :xeye:

I did try to measure the rail voltage during Clipping with a digital Multimeter and it looks OK !! Only a slight drops of +- 2,3 volts during hard drive.

With the CA 640A V2.0 output, I believe that it's better than my faithful Acram 8P but real test review otherwise.

As for the sound stage, it is good with tight bass and clear at high end when play at normal volume level. :) :) .

Quote

" Modding the protection circuit should only be with great care...
"

That's why I would like someone who know the circuit well or have a circuit diagram to guide me on that.

Many thanks.:D
 
just a standard Taiwanese circuit
This amp is infact british designed and only built in the far east. :)

I own A 340A SE which has 10'000uF caps per rail and its only a 45 wpc amp. I would say go with the 12'000uF per rail or possibly a tad higher if you have space.

Input buffer (preamp) mods:

Use 2.2 Film MKP types for signal path decoupling (I used Mundorf's). Upgrade the local decoupling caps (grey box types) to larger high quality caps (I used 100uF 25v Nichichon MUSE). Fit a socket and try some better opamps (I uesd LM4562). Also upgrade the local decoupling caps on the input selector chip.

l_fbb1ef7de5a6426082294e8a8a1d8e2d.jpg


Im assuming the circuits are simmilar on 340,540,640A's so the following should apply too:

There are signal decoupling caps before the power stages so change these to 2.2 film types. Also in the tone circuit there are a pair each (tone and 'tone bypass') which need replacing. Again use film types. If you use the tone control then replace the ones in that circuit and try some better opamps.

PSU: Fit some good quality low ESR caps (such as Panasonic FC, Blackgate etc) inplace of the cheapo lytics - before and after the opamp regulators. I have replaced the 100uF caps on each rail -after the regs - with 220uF Blackgates. Replace the diodes on the power amp supply with hexfreds. Use Shottkey type on the preamp stage. Lastly fit some low noise regulators such as 'Spower' inplace of the 7815,7915.

l_20d11a81ade743e3b728fbdeca5e1e91.jpg


l_89b6d317b051401f9907a84012e8ddf8.jpg
 
Well boys... have schematic for uncle Charlie?

Please....be kind with your old friend here:

panzertoo@yahoo.com

I am , JUST CRAZY, for this amplifier... also i will be very, very, very, very happy to watch the schematic and to assemble it in my home to listen.

This is Doctor Self last design...and they guy use to do good things... and this unit has some modifications into the bias system...some kind of dinamic biasing from A to AB. and the oposite..seems something different than usual automatic shift, something usually happens from class A to AB when we increase volume

regards,

Carlos