Cambridge Audio Azur 340A Acting Funny...

[Williham]

Hey guys!
Been a while since I posted on this forum, nice to see it's as epically busy as ever.

So my heavily modified Cambridge Azur 340A is acting a bit funny, but before I tell you what's wrong I'll tell you the mods I've done:
-Removed the DC input blocking caps next to the inputs.
-Upgraded the DC blocking caps on the buffer input to 4.7uf Mundorf Supremes, giving a lower bass rolloff (They were spares! ).
-Replaced the DC blocking caps on the LM3886 input with 10uF Wima MkS 2 Xls
-Replaced the buffer opamp with a LM4562HA, which is beautiful sounding...
-Replaced the power amp filter capacitors with big ol' 15000uF Mundorfs
-Replaced the rectifying diodes on the poweramp and preamp supply with 5A Shottky diodes
-Replaced the Poweramp feedback (I think feedback anyway) with 100uF military grade Tantalum
-Replaced the preamp reservoir caps with nichikons
-Replaced the Poweramp decoupling caps with Nichicons
-Bypassed the tone controls, balance and volume comtrols
And
-Added a filter circuit to the voltage regulators, comprised of an RC circuit and a bypass transistor

Mike Snowdon helped me with all of these modifications so lot's of thanks to him!

But anyway, the problem is that one of the voltage rails to the preamp is 9v higher than it should be; giving -24V instead of -15V, I think It's the negative rail, I'll have to check again to verify.
Now I think that the problem has arisen from the voltage regulators, or at least that's what I'm guessing. On trying to find the problem I realised the LM4562HAs metal can was heating up quite a lot, so I swapped in a spare NE5332 and on powering up it just literally cracked in half with quite a nasty noise (and smell!) produced...

So that's what I've done and think is going on, what do you guys think? Should I just get some more voltage regulators?

Cheers,
Will

[wakibaki]

If you've got an obvious fault, an extra 9V on what is supposed to be a regulated voltage rail, then the first thing to do is fix that...

w

[jaycee]

Yes, U9 will be responsible - however they do not normally fail shorted so it is possible you have a bridge on the PCB from the rework you've done.

When you say "Added a filter circuit to the voltage regulators, comprised of an RC circuit and a bypass transistor" - what does this consist of ?

[Williham]

K, that's good to know it is the regulator; to be honest with the added circuit the regs stick very high in the case and I have shorted the tops... But it's failed about a month after that happening, which is odd?

The noise filtering circuit is as follows:

[jaycee]

I see, looks like a capacitance multiplier. TBH if you want to improve things, use an LM317/337 based regulator and bypass the ADJ pin with 10uF. Any more than that is not worth it in this application.

One other thing to check - you haven't added large capacitors on the regulator output have you? This is a no-no.

[Williham]

Darn, I just ordered the same ones!
Would the regs you suggest just swap in then? And if you don't mind me asking, what makes them superior?
Also, my knowledge of that circuit was that it acted like a low pass filter, with the transistor somehow improving the filter capabilities, I'm guessing a capacitance multiplier is something else, could you explain? I'll do a bit of wiki browsing now as well, as I'd like to understand this better.
And no I didn't put large capacitors on the reg outputs.
I've just started a circuits lab module at university this year, so am hoping that will help with all of this! I felt cool (In a delightfully nerdy way) when I realised I already had the reading for the course, The Art of Electronics! :P
Cheers for the responses jaycee.

[jaycee]

LM317/337 will not just drop in but it should be a small mod. The 317 is a better regulator over the 78/79 series for noise if used properly.

You are quite right that the capacitance multiplier does work as a low pass filter. See here: Capacitance multiplier - Wikipedia, the free encyclopedia

[Williham]

So, out of curiosity, how do you work out the crossover frequency for this type of filter?
Also, the bypass capacitor for the ADJ pin, what would I bypass this to? the Input pin? And would I then use this instead of the "common" pin on the 7*15s?
Cheers.

[Williham]

So I've replaced the negative rail voltage regulator with the LM7915 I ordered to no avail; still is -24/25V instead of -15... So I'm going to try removing the capacitance multiplier and just use the voltage regulators, see if that works. To be honest a lot of the tracks on the PCB have seen better days (The modding work was my first "big" project), so I've bought some strip board and may just put the preamp with it's power supply onto the stripboard.

[Williham]

Ok, after some consideration, I've decided I'm going to completely remove the preamp section from my Azur 340A and build a seperate external preamp; reusing the high quality components I have, adding a volume control and having a gain of 6dB.
Does anyone have any recommendations for a kit that is average difficulty, is opamp based and high quility? I'll scour the forums as soon as possible, but if anything pops into mind feel free to post it up.