Eiron wrote:



Upgrade? That's a bit of an exaggeration.
And it's simply not true that 25 year old electrolytics will need replacing.
They might, but if they measure OK, they are OK.


** Quad 405 & 405-2 amps used PCB electros that were faulty.

Despite being fully sealed and sold as long life by the makers "Roedestein" they all failed early with ERS values going off scale and microfarads disappearing. See the dark red caps on this PCB.

http://upload.review33.com/images/20...1147094159.jpg

Same thing happened with the pair in the output crow-bar circuit - which could prove a tad disasterous !!


..... Phil

Don Pearce wrote:



This is worth a read. It describes an upgrade that reduces gain by
10dB - designed to match the higher output levels of more modern
source.

http://www.desmith.net/NMdS/Electron..._upgrades.html

Basically it is a change to the feedback circuit around the input op
amp, originally 330k/22k for a gain of 23dB, now 100k/22k which gives
a gain of 13dB.


** Quad built sub-sonic filters into the 405 & 405-2 for very good reasons - response was flat down to 20Hz, -6dB at 10Hz and -20dB at 5Hz.

The above mod unwisely removes this filter making the response extend to 1Hz.

Quad also built crow-bar circuits into both models which will place a short across the output if significant DC or sub-sonics appear. Perfectly harmless to do so if there is a sub-sonic filter inside the amp, since the crow bar will then only operate in case of a serious fault developing.



..... Phil

Phil Allison wrote:



http://www.desmith.net/NMdS/Electron..._upgrades.html

Basically it is a change to the feedback circuit around the input op
amp, originally 330k/22k for a gain of 23dB, now 100k/22k which gives
a gain of 13dB.


** Quad built sub-sonic filters into the 405 & 405-2 for very good reasons - response was flat down to 20Hz, -6dB at 10Hz and -20dB at 5Hz.

The above mod unwisely removes this filter making the response extend to 1Hz.


** Despite the author's claim of " ... flat response down below 1Hz " this may not be true. The feedback cap value in the op-amp stage (C4) has been scaled along with the main DC servo loop value (C2) - so there ought to be no change in the roll off behaviour at low frequencies.


..... Phil

some op amps were quite noisy and had good headroom so to speak so maybe
this is part of the reason.
Brian

--
----- -
This newsgroup posting comes to you directly from...
The Sofa of Brian Gaff...

Blind user, so no pictures please!
"Don Pearce" wrote in message
...
On Mon, 16 May 2016 13:21:52 GMT, (Don Pearce) wrote:

On Mon, 16 May 2016 13:20:49 GMT, (Don Pearce) wrote:

On Mon, 16 May 2016 14:12:56 +0100, "Dave Plowman (News)"
wrote:

Got a Quad 405-2 in for repair. Late model with IEC in and out, and the
on-off switch.
Heavily modified. PS PCB added with two lots of rectifiers and caps and
the twin secondaries of the original transformer feeding each one. Dunno
just what difference this makes in practice.

Amps are modded too. Most noticeable are some caps on the reverse side
of
the PCB and Burr-Brown OPamps in sockets - Quad has soldered in the
TL071
by then. A quick glance looks like they've upped the volts to the OPamp
too.

But the odd thing is the gain of the amps is a lot down over the
original.
Quick check shows something like 10dB. Why would this have been done -
or
have I got to look for additional faults?

Both channels the same? Sounds like a deliberate change rather than a
fault. I mean, if they dropped the gain of the 405, they had to
increase the gain of the preamp accordingly. The only schematic I have
shows an OPA 604 which is a FET input op amp from Texas instruments.

d

Ignore that. My schematic is for the original 405, not the -2

d


This is worth a read. It describes an upgrade that reduces gain by
10dB - designed to match the higher output levels of more modern
source.

http://www.desmith.net/NMdS/Electron..._upgrades.html

Basically it is a change to the feedback circuit around the input op
amp, originally 330k/22k for a gain of 23dB, now 100k/22k which gives
a gain of 13dB. Does that sound about right?

d

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

In article , Don Pearce
wrote:
My initial reaction would be to examine them for any physical signs of
distress, and then measure the ripple when the amp was delivering a
reasonable amount of power. That would give me some idea if they'd need
replacing without having to unsolder them. If all seemed well and the
amp passed its specs I'd then decide if they should be replaced or not.
If I were doing this for someone else, I'd also ask them what they
preferred if there was no obvious problem.

Jim

The problem with electrolytics is that they don't just decide to die at
random, they have a definite life curve, and after twenty years, even if
they measure OK, there is not a lot left in them.

Afraid I don't agree with that as something that always applies.

The 'lifespan' of electrolytics will vary depending on how well they were
made, how appropriately the examples were chosen, and how 'hard' their life
has been in terms of applied voltages, currents, temperature, etc.

The electrolytics in some of my amps are now over 30 years old. Each time
I've checked them and the amp performace, so far, they have been fine. Of
course that may end soon. And I'm sure there are examples where the caps
have degraded and need replacing over much shorter periods.

So I can see that it would make life simple and safe for repair work to
presume they should be changed after, say, 20 years. Of course it means
more work and thus cost for the owner, but helps avoid a possible problem a
few years later. Yet it might be that the removed examples would have been
fine if they showed no signs of any real reason to replace them.

Obviously if this was for someone else I would advise them of the
problem that could arise sooner or later, but for me, no, they would all
go.

I was talking a while ago with Mike S. when I he renovated my 626. My
initial wish was to have its thermal delay bypassed, and better bridge
diodes fitted. My reason being that the delays did tend to fail being
electromechanical parts.

However he pointed out that most of the failures were in the first few
years. The ones that survived, in his experience, went on working quite
happily. Given that he has serviced and renovated many more sets than me, I
let him leave the delay in circuit.

One aspect of electrolytics is that they can actually 'self heal'. If used
biassed correctly any leakage current can tend to reform any thinning spots
in the insulation layer. So, used well within their limits, decently made
ones can last a fair time in my experience.

However I *don't* ever repair sets for anyone else. If I did, I'd discuss
such matters with them and let them decide unless I found a clear reason to
replace.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

On Tue, 17 May 2016 09:47:31 +0100, Jim Lesurf
wrote:

In article , Don Pearce
wrote:
My initial reaction would be to examine them for any physical signs of
distress, and then measure the ripple when the amp was delivering a
reasonable amount of power. That would give me some idea if they'd need
replacing without having to unsolder them. If all seemed well and the
amp passed its specs I'd then decide if they should be replaced or not.
If I were doing this for someone else, I'd also ask them what they
preferred if there was no obvious problem.

Jim

The problem with electrolytics is that they don't just decide to die at
random, they have a definite life curve, and after twenty years, even if
they measure OK, there is not a lot left in them.

Afraid I don't agree with that as something that always applies.

The 'lifespan' of electrolytics will vary depending on how well they were
made, how appropriately the examples were chosen, and how 'hard' their life
has been in terms of applied voltages, currents, temperature, etc.

The electrolytics in some of my amps are now over 30 years old. Each time
I've checked them and the amp performace, so far, they have been fine. Of
course that may end soon. And I'm sure there are examples where the caps
have degraded and need replacing over much shorter periods.

So I can see that it would make life simple and safe for repair work to
presume they should be changed after, say, 20 years. Of course it means
more work and thus cost for the owner, but helps avoid a possible problem a
few years later. Yet it might be that the removed examples would have been
fine if they showed no signs of any real reason to replace them.

Obviously if this was for someone else I would advise them of the
problem that could arise sooner or later, but for me, no, they would all
go.

I was talking a while ago with Mike S. when I he renovated my 626. My
initial wish was to have its thermal delay bypassed, and better bridge
diodes fitted. My reason being that the delays did tend to fail being
electromechanical parts.

However he pointed out that most of the failures were in the first few
years. The ones that survived, in his experience, went on working quite
happily. Given that he has serviced and renovated many more sets than me, I
let him leave the delay in circuit.

One aspect of electrolytics is that they can actually 'self heal'. If used
biassed correctly any leakage current can tend to reform any thinning spots
in the insulation layer. So, used well within their limits, decently made
ones can last a fair time in my experience.

However I *don't* ever repair sets for anyone else. If I did, I'd discuss
such matters with them and let them decide unless I found a clear reason to
replace.

Jim

I agree that it doesn't always apply, but this is really a statistical
thing. I recently had to design a switched-mode power supply for a
demanding US customer. The spec demanded a seven year lifetime, which
was determined almost entirely by the main reservoir electrolytic. It
needed exhaustive testing and analysis of temperature, ripple current,
inrush current etc. I eventually got the seven year lifetime, but it
needed a very expensive Nichicon cap, that was apparently vastly
overrated for the job.

The thing is that as you identify, there is an early failure
mechanism, which I tackled with a tough burn-in regime, and then a
reasonably stable life followed by a long statistical tail. Caps can
have lifetimes well into that tail but the likelihood of failure is
increasing all the time. For me, as I say, although it is quite
possible that caps would last a long time, they are pretty cheap in
the greater scheme of things, so I would change them.

d

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Dave Plowman (News) wrote:


But the odd thing is the gain of the amps is a lot down over the original..
Quick check shows something like 10dB. Why would this have been done -



** The only practical reason is to reduce residual hum and noise when the amp is operating in a typical installation - which these days may include supply grounded items like TV receivers and computers.

Unbalanced hi-fi gear is prone to ground loop hum which can be very tedious to fully eliminate and the 405 having a fixed gain of 56 times makes that task so much harder.

The Quad II valve amp had a gain of only 7.1 or 10 times - depending if it were set to 8 or 16 ohms. As a result, the input level was 1.5V with noise and hum around 95dB below 15W and noise alone nearly 110dB below - remarkable for anything using valves.

With the 33/303 combo, Quad increased amplifier gain to 40 by making the input level 500mV. 405s followed suit with the same input level producing 100W output - so it would mate perfectly with the 303 pre amp. Connecting a supply grounded accessory ( like a tape deck or tuner) normally resulted in audible hum at higher gain settings.

Reducing the gain of a 405 to 17 times is an excellent idea, makes any residual noise 10dB less and really helps with hum issues.

Here's that link again:.

http://www.desmith.net/NMdS/Electron..._upgrades.html


..... Phil