Cambridge Audio P55 Schematic

[madtecchy]

Hi.

I have a Scematic for the Cambridge Audio P55. It is a scan of a very poor quality photocopy. I now have it drawn in Eagle . However drawing the scematic was not too hard i wish i could say the same for the board layout. If anyone would like a copy drop me a line and i will post on forum or mail it too you. Also would anyone be kind enough to do a layout. Or even a few tips on using eagle layout editor. As im a newbie to eagle would anyone like to take a look at my attemp to see if there are any errors.

Best Wishes Ian

 

[davidsrsb]

Something is very wrong with the Q18 Q19 -ve side of the output stage. Those transistors should be in parallel and the diodes are also wrong

 

[jaycee]

That looks suspiciously like the Cyrus 2 power amp.

 

[madtecchy]

Thanks david ill sort that now . Well i hope so as the original scematic is barley readable. and my mistake is so blatent now you have pointed it out

Hi jaycee thanks i will have a look at the cyrus 2 As i may be able to get a clearer schem for that . I was also told it was similer to a AR model but that one uses complimentary output stage.

Regards Ian

 

[madtecchy]

Updated

schematic Is now fixed hopefully.

 

[davidsrsb]

Looking better, but 1N4002s on the output? - a bit under rated
0R22 in series with the output gives a low damping factor with 4R woofers

I guess this was a Stan Curtis design, which would explain the Cyrus similarity

 

[madtecchy]

Hi. Thankyou David.

Im guessing from what you say it would be a good idea to uprate the 1n4002's with somthing a little larger for reliability issues maybe. also i have just read the circuit description for the p55 it says
" Wire wound resistor R49 ensures a negative feedback signal under output short situations . Thus it is an aid to stability with reactive speaker loads."

Do you think replacing R49 with an inductor and series resistor would improve the damping factor and at the same time aid stability too.

Would increased damping factor improve sound quality. My guess is it would maybe improve the bass on higly reactive and also lower impedance speakers.
Ps Im still learning. And yes you are correct it is a stan curtis design.

Kind Regards
Ian

 

[davidsrsb]

I would leave it alone, stability problems could be very expensive.
This is a design that will work best with "valve friendly" speakers, avoid those with very low imedances as the extremes like Maggies and ribbon tweeters

 

[jaycee]

What rail voltages does the P55 use?

The circuit does work well, however if you're after building a clone the original Sanyo parts are probably difficult to get. I don't think i'd bother with the quasi-comp output stage either.

 

[madtecchy]

I would leave it alone, stability problems could be very expensive.
This is a design that will work best with "valve friendly" speakers, avoid those with very low imedances as the extremes like Maggies and ribbon tweeters

Hi David.
Well i dont have any maggies or ribbon tweeters. I do use as my main speakers a pair of DefTech BP8's. However for testing i have a set of old celestion DL4 into wich i have installed some 4700 uF in series with the input connectors for testing purposes as i dont realy like the sound of them. The amplifier as is drives the DefTechs very well they are bipoles and are quite a friendly 4 Ohm load also at 91 db W m they are quite sensitive so i dont need to use much power to get a decent listening level. Also i like Risk and i dont have a lot to lose as the amplifier cost very little im using it as a learning exercise to create a clone and learn eagle at the same time . Thankyou very much for the warning but hey im a big boy lol responsible for my own actions and wont comes shouting at you if it lets out the blue smoke. The first thing im going to do is use a seperate psu for each channel and totaly bypass The pre and tone controls as that will remove over 2 feet of track and unscreened wire each input has to travel before entering the power amp section as i only use A Dac as source i dont even need a volume control as the dac as one built in.

Kind Regards Ian

 

[madtecchy]

What rail voltages does the P55 use?

The circuit does work well, however if you're after building a clone the original Sanyo parts are probably difficult to get. I don't think i'd bother with the quasi-comp output stage either.

Hi Jaycee.
The amplifier is running 41 volt rails with the incomming mains at 238 vac.
are quasi comp output stages bad. Qm. as for the sanyo parts i guess you may be right. im guessing this may be a common problem when cloning any older amplifier.
Kind Regards
Ian

 

[jaycee]

I'd personally say 41 volts is too high for the BC546/BC556 in the second stage long term. I'd stick to 35v supplies - thats enough for 50W into 8 ohms, plenty for a domestic system.

Quasi comp output stage is a bit of a relic really - it was used when PNP transistors where nowhere near as good as NPN. No longer a big issue really.

I played with this design in LTSpice and got it stable and simulating well - 0.0017% THD simulated at 10KHz, 50W into 8 ohms resistive. Haven't built it though. Will attach a copy if you like.

 

[davidsrsb]

Quasi comp output stage is a bit of a relic really - it was used when PNP transistors where nowhere near as good as NPN. No longer a big issue really.

As Sanyo made the 2SB776 PNP complement, on the same datasheet, this was not the reason. I suspect the PNP was significantly more expensive.

Keep the rails to 35V, the SOA curves go off a cliff above 70V

 

[madtecchy]

Hi jaycee and David.
Thankyou both for your Replys. I have an original p55 It is a later version i think. The one with the aluminium sides opposed to wooden. This amplifier is the one i measured to give you the rail voltage. From what you both say about the SOA Im guessing the amplifier should have let out the blue smoke by now . it is still working fine. I personally never punish an hifi amplifier and im guessing the previous owner didnt either as i too would of expected it to have blown . The thing is i dont know if the transformer is original. I am going to heed your advices and i have two 24-0-24 200va torroids i used while playing with Lm3886 I.C's . I will fit these to be on the safe side. Jaycee yes i would like a copy of your design please. I would like to Thank both of you for the time and experience you have shared with me .

Kind Regards
Ian

 

[jaycee]

Those toroids should be perfect.

OK I've attached an image showing the schematic, and the LTSpice files + models used. The transistors are all from Fairchild as they have an excellent range of parts suitable for building power amps.

The gain is set so that it reaches full output with 2V RMS input. If you're using a DAC that should output 2V RMS max with your volume up full If you want to make it more sensitive decrease R23 - 1K6 should be about right for full power at 1V RMS for example.

Please do bear in mind I have not got further than simulation with this design, though I do like it so maybe I'll give it a go over Xmas.

 

[davidsrsb]

The output device SOA is not a problem with (near) resistive loads. A speaker with one of the transformer coupled tweeters would be

 

[grabread]

Sorry to bring such an old thread back to life, I need a schematic for this model. If the original poster can perhaps post some info about how to set the bias and perhaps a schematic of the preamp and power supply sections it would be much appreciated.
Any general info on this amp much appreciated thank you.