advertisement


QUAD 44 PREAMP: UPGRADES

eguth

eguth

pfm Member
MEMOIRS OF A DIY HI FI NUT - Part XVIX​

QUAD 44 PREAMP: UPGRADES​
I
AS I HEAR IT​

Which components in the hi fi chain are usually most responsible for the disappointment that comes from music reproduced in the home at not a high enough standard? We all hear things differently. Joe may be delighted with what disappoints Ann. Some of us listen more betterer than others. Moreover, most of us do not possess good enough equipment to reproduce a wide variety of music so that it sounds like it did at the performance. Even if we did, not everyone would agree the result. I disregard ‘enjoyment’ here; some get great pleasure from listening to a small transistor radio. Why bother with hi fi?

For those of you who think turntables are frightfully important I agree- but only provided they are up to a high standard. After that standard is met I am convinced that preamps are more important. They are as important- perhaps more important- than amps. We need equalisation before the amp gets the signal. Preamp or just a phono stage? Simple is beautiful. But…

A perennial puzzle for those of us (me) who are not able to design or build a top quality unit is the best point in time- with an eye to commercial developments- when we should purchase a well received design. The worst breed of parasites are the Marketing Men. They get in between you and the music; and grab your money in the process. Or should we take a chance on some (possibly better) second hand item that could turn out to be a problem?

There has been much interest in Quad 44 and other ancient Quad preamps recently. Apart from the current low prices I can’t imagine why. Some admire their looks: but I see them as butt ugly. So far as I am concerned ‘sound’ usually accounts for 99% of the true worth of a piece of hi fi; ‘looks’ 1%.

If you can possibly afford to avoid buying an old Quad preamp then please, please don’t buy one. Graham Nalty advised me, in the 1980s, to start modifying a better preamp instead of the Quad 44. I did not take his advice: fool I! How I wish I had!

The bog standard ’44 is not worth listening to.

I purchased my Quad 44 without hearing it on the strength of the Hi Fi News review (Jan 1980, p.135-7). It was said to be ’one of the most versatile preamps ever made’. I needed such versatility like a hole in the head, because I have only ever had two sources: vinyl and tuner. My tuner bypasses the preamp and goes straight into the passive preamp. My ‘44 is serial #12941 purchased, new, on 7.2.81 from Westwood and Mason, Oxford, for £244.00. I still have it and use it almost every day: 30 years on.
I remember my reaction when I got it home and hooked it up into my system. This could be summed up in one word: “mediocre”. ‘Why not chuck it into the ocean and make some deep sea diver’s day?’ I mused. After all ”Quad” spelt backwards is dauQ (almost duck) so it would like to be in water.

What I heard was:-
1) no deep bass and inadequate bass above that;
2) not much by way of extended high frequencies;
3) disappointing in terms of realism;
4) midrange lacking in dynamics and transparency.

That is only a start: I won’t bore you with the rest.
So…sell the family silver and start off with something better. Only one problem: I had no family silver to sell. I reluctantly decided to try to modify the unit.

Well… here I am 30 years later still using my (very extensively) modified Quad 44. “Why?” I hear you ask. Wait and I will tell you.

I begin by saying that I kept on and on doggedly modifying the Quad 44 partially because of lack of funds and frustration and also because there was no ocean handy and because every time I changed something the sound improved; sometimes dramatically. Thoughts of seaside began to recede.

My approach to my hi fi equipment is to preferably upgrade what I already have rather than abandon it. By this means I have avoided a roomful of hi fi junk piling up and blocking the sound waves.

Read on, and at the end- as a reward for your patience- I will give you my verdict on my final version compared with a ‘show stopper’ preamp (“best in show”)- both installed in my system for the comparison. “Haw,haw (I hear you laugh)…he is an idiot for not throwing his ’44 into the deep sea in the first place.” So be it.

I am very critical of the Quad 44 in this Memoir; but I try to be fair to it. This Quad preamp needs major mods to bring it up to acceptable audiophile standards as I hear it; however, it is well made and will give trouble free service for many years provided that you keep it out of salt water- which is more than you can say about some other commercial preamps.

Putting the ’44 into context in my system as it was when I purchased it- this was after years of using my Radford SC22 (1960’s) hybrid valve/transistor preamp. The Radford is an elegant and superbly engineered preamp; in its day it was first class.

Yet the bog standard Quad 44 preamp was an audible upgrade from it when used with the MM module and a MC transformer via the Tape Recording outlet into a passive preamp. To be fair to the Radford I had never used it in this way so perhaps the comparison is improper. Still…..

Next: the first tranche of mods.
_______________________________________________
 
The bog standard ’44 is not worth listening to.
Click to expand...

Respectfully completely disagree. I use one with a 303 and ESL 57s and I think the ensemble makes a splendid noise. Now perhaps it could be bettered, but since I like what it does so much, why bother?
 
line inputs are ok but phono stage i think uses a poor opamp having that upgraded makes a major improvement from memory the ic was a 351
 
I have a quad 44, naim 32.5 and Hicap, cyrus aA7, creek passive and a RB4. The RB is by an arm and a leg better than the lot for reasonable quality line level sources.

The results are clean, detailed, but not bright.

this run through a 140. Rather than the 405s I have. Less seems to be more than more.
 
The signal switching in the later Quad PreAmps was effected by the 4066 - CMos chip, designed originally for the Plessy System 'X' telephone exchanges. Being severely bandwidth limited, the 4066 has the unique capability of strangling anything resembling music and so is inappropriate for use in anything aspiring to the description of High Fidelity Equipment.

Bottom line: You may replace components, re-engineer, upgrade, rewire, re-power, conjure, apply magic dust, cast bones etc., etc., etc., but until you dump the 4066s, the 44 will only be at best, "Sleepy Valley Audio" in performance - and good for lift (elevator to our colonial cousins), music only.
 
I had a 44 circa 1990 - an electronic engineer friend suggested swapping out the phono opamps - my memory is TL071 - with 5532s, which improved things no end.

It was inoffensive sounding, but easily bettered by a modified Hafler DH101. However the industrial design of the 44 and especially its controls is (IMO!) yet to be improved on.

eguth, it will be interesting to hear to what extent your mods were detail refinements vs wholesale re-engineering...
 
"However the industrial design of the 44 and especially its controls is (IMO!) yet to be improved on."

How true and credit given where it is richly deserved.......
 
I would like a demo of a bog standard Quad 44 in an average sized British lift with a full range speaker driven by a fully kitted out front and rear end- i.e. with a sota turntable, hi end arm and cart, a full range hi end speaker plus a subwoofer and supertweeter - playing a Grateful Dead LP with volume turned up. Please provide ear plugs and an emergency stop at the next floor.
 
Dunno about the 4066, they are used in the Quad 99 Pre, too.
Each signal passes through one MOS transistor in the 4066. Can't see any prob there if the signal is below clipping level, maybe the issue is that unclean supply comes in...
 
LesW said:
The signal switching in the later Quad PreAmps was effected by the 4066 - CMos chip, designed originally for the Plessy System 'X' telephone exchanges. Being severely bandwidth limited, the 4066 has the unique capability of strangling anything resembling music and so is inappropriate for use in anything aspiring to the description of High Fidelity Equipment.

Bottom line: You may replace components, re-engineer, upgrade, rewire, re-power, conjure, apply magic dust, cast bones etc., etc., etc., but until you dump the 4066s, the 44 will only be at best, "Sleepy Valley Audio" in performance - and good for lift (elevator to our colonial cousins), music only.
Click to expand...

I often read this but it completely ignores the working conditions inside the Quad units.

The switches have high 'on' resistance of about 200 Ohms when supplied by the rails in the 44. They also have a THD performance which varies hugely depending on the current passing through them and the impedance of the circuit in which they operate. In the Quad 44 (&34) currents are tiny and circuit impedance high (each stage feeds the input of a J-FET op amp). Distortion under these conditions remains very low and bandwidth entirely adequate for audio use.
A Quad 44 on the bench will typically deliver <0.002% THD - that includes the 4066 devices and the compliment of 1970s op amps.

You can easily fit modern 4066 ICs as i have in my 34 and these have on resistance of typically 30 Ohms and roughly a five-fold reduction in distortion compared to the old 4066s.

I've presented direct v Quad pre demos and the different is incredibly difficult to detect, and often people cant. I have the files for anyone to download if they wish and they can determine for themselves if a Quad pre is only fit for elevator music.

More generally:

I recently did a full overhaul of my 34 (essentially a 44 with less inputs) and it is very difficult to get any change to the performance of the line inputs. You can eek out a slightly lower noise floor and and improve distortion slightly but the whole exercise illustrated that the units are basically fine as supplied. Using OPA134 and OPA2134 in place of the old TL071/2 does reduce dc offset at each stage to the point where the coupling caps can be eliminated. Each stage has a 100uf coupler and all can be removed, however there is no audible change IMO, probably because the circuit filter time constants are set elsewhere in the circuit and these caps do literally nothing other than block dc. In other words, they are massively over-specified at 100uf.

The one area where a 44 and 34 can be improved is the phono section.
There are two main areas.
The first stage can usefully use modern low noise transistors operating at lower collector currents (needs a small circuit mod) and the phono stage op amp can benefit from using a modern very low noise type such as the LME49710HA. Be careful as the device must be stable at unity gain (and we have this at HF due to the RIAA filter). The AD797 would be nice - but it wont work.
The second benefit is to lower the LF response roll-off knee and you can do this by using a 3.3uf Wima 50v cap in place of the standard 2.2uf Quad part. The latter would have been sensible in the 70s with high compliance cartridges chucking out lots of LF rubbish and noisy decks, but things have improved over the years.

Bottom line is that these units are fine but can use a little light work to freshen them up - use a few better components but don't redesign them.

One last thing for those using these with 306/606/405 power amplifiers.
These power amplifiers have high input sensitivity (375-500mv) and the output of the Quad pre amplifiers is padded-down to suit.
You can usefully lower the overall noise floor by reducing the input gain on the power amps to 1.5v and raising output on the pre amps to match.
Small but useful benefit with the 306/606 but very useful indeed with the noisier 405.
 
I think it is fair to say even a freshly serviced 44 sounds "On the polite side of neutral"
Giving generally a good sound with the edges "Rounded off" always pleasant to listen to especially on bright recordings,but never showing good recordings to sound as they should,however with a few changes it can be brought up to modern standards.
 
tones said:
Respectfully completely disagree. I use one with a 303 and ESL 57s and I think the ensemble makes a splendid noise. Now perhaps it could be bettered, but since I like what it does so much, why bother?
Click to expand...

34-306-57 here and agree entirely.
 
Robert

The point Les W made (as I read his Post #5)) is primarily criticising the 44’s ‘severely limited bandwith’.

If you want to argue with this it is an argument that can be settled via an oscilloscope. I have seen one in use on the (partly modified) ’44 and what Les said is quite correct. I arrived at the same view as Les’s on Day 1 of my purchase using only my ears which, so far as I have been able to see, are not fitted with an oscilloscope.

Your discussion of Les W’s point- instead of dealing with bandwith- talks primarily about distortion and noise. You do mention the possibility of lowering the LF rollof in the phono section, but if you are using speakers that are bass- light (e.g. Quad ESL 57s or 63s or Impulse H2s)) the difference would be incredibly difficult to detect (read ‘probably impossible by ear’) so why bother?

I think your point about fitting modern 4066 Ic’s is a useful one. But, for reasons that shall emerge in due course, I have not done this and will not do it.
 
At this point I would submit that the 4066, although seemingly representing the veracity of Mr Walker's legendary 'straight wire with gain' claim, in fact does no such thing. Years ago, we daisy chained some 15 or so 4066 switches and introduced the chains into the line level signal paths between pre and power amp. What emerged was a shadow of the actual programme along with a fair bit of noise.

In times past, I designed a custom reed relay selector PCB to replace the 4066s. It worked a treat and the difference in clarity, dynamics and low level detail, was truly marked. Last week even, I had a 44 in for service which I'd converted this way twenty years ago. It was still able to draw sighs of admiration from the listeners who heard it in my system.

Question what I've posted by all means & continue to use the 4066 if you wish - your choice as my work here is now done.......
 
martin clark said:
H2s bass light, Rob..?!
Click to expand...

I know Martin, the mind boggles sometimes at what appears on the internet!

Some other points from todays contributions:

A Quad 44 is not a Plessy telephone exchange.

15 4066 ICs in series is also not a Quad 44 - and there is no mention of operating conditions. Anything can be made to look bad if you try hard enough and have a point to prove.

A Quad 44 measures out at +/-0.25dB 20Hz-20KHz on line with the LF -3dB point at 10Hz.*
Eminently sensible and not hampered by 4066 switching as used in the Quad circuit.


* yes I have measured it myself on my own 34.
 
“…A Quad 44 measures out at +/-0.25dB 20Hz-20KHz on line with the LF -3dB point at 10Hz.*..”

Following your asterisk you say you have measured NOT a ’44, but a ’34. I have carefully avoided any mention of ‘34s, because they are not ‘44s and there are differences besides flexibility. This thread is not concerned with Quad '34s.

If what you say is correct there would have been thousands of damaged Quad electrostatics when used with ‘44s. The low frequency rollof engineered into Quad preamps by Quad was deliberate in order to avoid arcing in Quad electrostatic speakers.

Years ago John Atkinson measured an ESL 63 in a large American living room with the panels near the side walls to maximise bass response. He found (from memory) that it started rolling off at circa 70Hz and was very significantly down in dB by 40Hz. After Quad finally recommended using a subwoofer with these panels (the bass-light Gradient subwoofer) there was an immediate improvement, as I hear it. This with a hugely inadequate subwoofer. Before finally accepting that an H2 is not bass light why not try a good subwoofer with it?

I can’t remember what John atkinson found re: high frequency extension but it cannot have been impressive. That bit is for those who are most impressed by measurements. Me? I prefer to consider at measurements but then, like so many engineers have said, to let the ears be the final arbiter.
 
You must log in or register to reply here.


advertisement


Back
Top