1. Things you need to know about the new ‘Conversations’ PM system:

    a) DO NOT REPLY TO THE NOTIFICATION EMAIL! I get them, not the intended recipient. I get a lot of them and I do not want them! It is just a notification, log into the site and reply from there.

    b) To delete old conversations use the ‘Leave conversation’ option. This is just delete by another name.
    Dismiss Notice

Quad 34 MM load

Discussion in 'd.i.y.' started by chartz, Sep 4, 2019.

  1. chartz

    chartz pfm Member

    I use a 303 or a 405. I also use ESL.
  2. Tony L

    Tony L Administrator

    I really can’t explain it. The 34 has a dreadful reputation with ‘internet experts’, I suspect due to the anti-Quad sentiment of many ‘80s ‘flat-earth’ dealers, but has really surprised me, albeit after a little work. You do need to get rid of that HPF on the mainboard though as it just sits on dynamics to my ears. I don’t know which components it is or what to replace them with, but they definitely reside on the mainboard, not the plug-in phono boards.

    I’ll just ping @Robert in the hope he sees this thread and can advise. Removing that and setting the phono card loading for my Ortofon cart turned it from something I really couldn’t live with at all on vinyl into a preamp that has surprised everyone who has heard it in my system. It sounds very good with my current MP-500 too, though these days I do prefer my JC Verdier valve pre (which one would sincerely hope given the huge price difference!). The Verdier hated the Ortofons so I used the Quad for several years. It is a good preamp, in fact great for the money.
  3. chartz

    chartz pfm Member

    Okay, so let's wait for Robert's perspective. Cheers Tony!

    No flat Earth here by the way! ;)
  4. davidsrsb

    davidsrsb pfm Member

    On the main board C27 and C29 are limiting, change from 470n to 1u or even 2u2
    On the plug in board C12 and C13 could be increased from 1u to 4u7
  5. chartz

    chartz pfm Member

    Thanks. So that means an electrolytic for C27 then. Nothing else will fit. Are C12 and C13 part of the active EQ feedback loop?

  6. davidsrsb

    davidsrsb pfm Member

    Changing the 39k resistor loads on C27&29 to about 100k should also work
    C12 is outside the feedback loop.
  7. Jim Audiomisc

    Jim Audiomisc pfm Member

    Maybe I'm misunderstanding something, but C12/13 look to me to be inside a loop. The diagram also, to me, underscores PJWs tendency to have a high order LF rolloff. FWIW I use 34s and like them. But as general preamps, modified a bit. I've avoided using their 'disc' (LP) input stage.

    BTW One of the reasons for tweaking the LF rolloff is that such a filter tends to affect higher frequencies as well. So may sound like an oddity at other frequencies. Similarly, the presence of low level LF affects how we hear higher components.
  8. davidsrsb

    davidsrsb pfm Member

    Apologies, correct Jim. I was working from my phone earlier and did not notice this.
    Unusually for old Quad there is a bit of poor design here. Simulation shows those two 1.5uF capacitors turn out to be quite critical and one thing you never did was depend on an electrolytics value.
    The ideal design goes off a cliff at ~16Hz, degrade those capacitors to 1uF and it shifts to 25Hz, but boosts 100Hz
    Try 2.2uF and the cut drops to about 10Hz, but 100Hz is now down

    The 2.2uF C18 also turns out to be sensitive.
    Even when new this circuit must have been a bit variable, add a lot of age and who knows...
    The mid and top RIAA tracking is fairly good, just a tiny bit of treble boost, which is a common tweak
  9. chartz

    chartz pfm Member

    So, what do I do?
  10. Jim Audiomisc

    Jim Audiomisc pfm Member

    Underlines my earlier feeling, I think. For 'line' inputs the 34 can work very nicely. But I'd tend to either use something else for an RIAA stage or check what the 34 has carefully and consider modding it or ensuring optimum values for my purpose.

    In my case this was 'no contest' because I can use the Armstrong 320 preamp design which I'm biassed in favour of anyway. :) I can't help feeling others didn't use that design because of the 'not invented here' factor. (Uses a Hitachi IC and circuit. An IC that impressed me so much I also used it for line stages!)

    And the variability of electrolytics is, for me, another reason to make the LF roll-off low down so any variations don't intrude into the audible range. The required value becomes just "big". :)
  11. davidsrsb

    davidsrsb pfm Member

    The "correct" way to do this in the analog world if you want a sharp warp filter, is to put the warp filter after the RIAA equaliser and make use of the gain to use higher impedances to allow sane precision capacitor values
  12. Jim Audiomisc

    Jim Audiomisc pfm Member

    I'd agree. I used that approach FWIW. The snag is you need to ensure the RIAA can cope with any unwanted large LF wobbles, etc. A bonus is putting any warp/DC filter *after* the RIAA means you can easily make it user-switchable as well.
  13. davidsrsb

    davidsrsb pfm Member

    The Quad 34 was launched when circuit simulation was becoming available, but checking for tolerance effects not so thought about. The 44 schematic is harder to find. Does anybody know what that is like
  14. chartz

    chartz pfm Member


    Very interesting answers.
    Still waiting for Robert’s comments.

    Right now I can’t live with the 34. It’s just too lacklustre on LP.

    The 44 diagram is on HFE. Its original diagram is very different from that of the 34. Then it was revised and modified with the addition of the two transistors.
    I had always thought the TDA1034 was a better op-amp...
  15. John_73

    John_73 pfm Member

    Interesting comments! Dada recommended changing C18 to 3.3 or even 4.7uF. Reading the comments here, I’m assuming it’s not that simple to ‘fix’ this phono stage then with just a couple of cap changes? Using 3.3uF Panasonic ECWF film caps for those positions, and I definitely think they were an improvement, FWIW.
  16. davidsrsb

    davidsrsb pfm Member

    Running 3.3 and 4.7u on the simulator, this actually cuts the bass below 100Hz, substantially at 30Hz.
    BUT it also cuts the filter Q and therefore reduces group delay
    The default Quad filter likely has group delay in the audible range
  17. chartz

    chartz pfm Member

    So, would 10 µF be advisable then?

    Below 100 Hz, really?
  18. davidsrsb

    davidsrsb pfm Member

    Found it.
    So Quad went to a more conventional opamp circuit (with the extra IEC sub bass pole) and the reverted to a circuit similar to the 34.
    The 66 is even closer to the 34 design.
    he only thing that comes to mind is that ESLs might be extra sensitive to warps, so Quad have used a very aggressive warp filter again and again.
  19. chartz

    chartz pfm Member

    Yes I know, ESLs need that cut, but how does this explain the poor quality of the unmodified circuit?
    Still waiting for what to do by the way :p
  20. Arkless Electronics

    Arkless Electronics Trade: Amp design and repairs.

    OK I've worked out a series of mods to hopefully improve the 34 disc input.... flat down to a couple of Hz and about 1dB down at 1Hz.

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice