+ Reply to Thread
Page 2 of 2 FirstFirst 12
Results 11 to 18 of 18

Thread: Escoed Troika arrives. SUT advice

  1. #11
    Join Date: Jan 2009

    Location: Essex

    Posts: 32,049
    I'm openingabottleofwine.

    Default

    Hello Chris,

    Thanks for the link. My suspicions that you were misunderstanding the German data sheet were based on another Haufe data sheet that I had; I was unable to find the data sheet specific to the T-7883. So here goes:

    Übersetzungsverh. : 1:10 (transformer ratio 1:10)

    Quellwiderstand : 40 Ω (source impedance 40Ω)
    Lastwiderstand : 47 KΩ (load impedance 47KΩ)
    Primarpegel : dBu 0.05V
    (Primary level 50mV maximum, this is a misprint: ignore the dBu)

    Frequenzberich : 20 Hz 20kHz +/- 1dB ref. 1kHz (frequency response)
    Klirrfaktor < : 1% (distortion factor)

    Impedanz Z : 6,8 kOhm 1 kHz +/-30% an rt-gn Farbe /Pin
    (Impedance 68kΩ at 1kHz +/-30% between red –green coloured leads. Must confess I don’t know what Haufe mean by this.)

    Vormagnetisierung DC : 0mA
    (DC magnetising current 0mA, i.e. don’t allow any DC to flow through the windings.)

    Wicklungswiderstand : 11.5 Ohm : 1.3 kOhm +/-10%
    (coil/winding impedance (resistance))

    Thus the transformer has a turns-ratio of 1:10 and will, when terminated with the 47KΩ input impedance of your preamplifier, present an impedance of 470Ω to your cartridge.

    I cannot say what the correct loading impedance for your Linn Trioka should be, since I don’t use one. All I can do is point you to what Linn themselves recommend and to what other Troika users use. The recommended value seems to be 470Ω, so this is where you ought to start. If you find the sound too bright at the top end, then you may need to reduce the value from 470Ω. To do this you will need to connect a shunting resistor across the secondary winding of each transformer. The value of this resistor (in Ohms) can be calculated using the formula:

    Shunt resistance = (4700000 x required load impedance)/(47000 – (100 x required load impedance)).

    Thus if you want to reduce the load impedance to, say, 200Ω, then you will need to shunt the transformer secondary windings with (4700000 x 200)/(47000 – 20000) = 34815Ω, or 33KΩ. Strictly speaking the impedance of the transformer windings need to be taken into account, but this formula will get you into the ballpark.

    So now it’s down to you and your ears to proceed.

    Trust this has been of some help.

    Regards
    Barry
    Last edited by Barry; 23-05-2009 at 19:29.

  2. #12
    Join Date: Feb 2008

    Location: Canaries

    Posts: 193

    Default

    Thanks a lot Barry.
    By no means do I find the sound too bright, quite the opposite I might say, it doesn´t surprise me at either end of the frequency range - a bit over controlled, especially in the bass but maybe my expectations were too high. Just as the loading impedance can be lowered if too bright, can it be raised without touching my P10 ? I believe the Naim K cards were 560ohms but if I´m not mistaken that will give a negative figure with the formula.

    If it´s of any help, the 20 ohms setting on my AT-650 sounds better especially in the highs with the Troika whereas the Haufe sounded much better than the AT with the Denon DL103. Does this make any sense to you ?

    Maybe an active solution would be easier, as others have already suggested.
    Thanks again,
    Chris

  3. #13
    Join Date: Jan 2009

    Location: Essex

    Posts: 32,049
    I'm openingabottleofwine.

    Default

    Quote Originally Posted by Chris View Post

    If it´s of any help, the 20 ohms setting on my AT-650 sounds better especially in the highs with the Troika whereas the Haufe sounded much better than the AT with the Denon DL103. Does this make any sense to you ?

    Chris
    No, not at the moment - let me think about it.

    You can only reduce the impedance through using shunting resistors. If you really want to increase the impedance to 560 Ohm, then I suppose you could insert a series resistance of 56 kOhm - 47 kOhm = 9 kOhm into the secondary side of each transformer. This would, however, reduce the effective gain of the transformers from x10 to x8.4 (or a loss of 1.52dB). I have never heard of anyone doing this, so you could be the first. I wouldn't have thought the difference between using 470 Ohm or 560 Ohm would be that great.

    Let us know how you get on.

    Regards
    Barry

  4. #14
    Join Date: Jan 2009

    Location: Essex

    Posts: 32,049
    I'm openingabottleofwine.

    Default

    Hi Chris,

    I’ve been thinking about your experience regarding the use of the Haufe transformers with the Denon and Troika cartridges. Despite their simplicity the performance of transformers does depend to some extent on both the source and load impedance. As the data sheet shows, the Haufe T-7883 transformer is designed for a source impedance of 40Ω and a load impedance of 47kΩ.

    The Denon DL103 has a coil resistance of 40Ω and the Troika has a coil resistance of 4Ω [1] (similarly, the Asak has a coil resistance of 3.5Ω). It would thus seem that the 7883 is operating under ideal conditions with the Denon but not with Troika. This may go some way to explaining why you prefer the 7883 with the Denon. Loading any cartridge will dampen it; the lower the load impedance, the greater the damping. For the Denon the Haufe transformer presents an impedance ratio of 470/40 ≈ 12, whereas for the Troika it is 470/4 ≈ 120, so under these conditions the Troika may be under-damped. Devices with a lower than specified source resistance can sound over bright with a curtailed bass.

    Doing a search, it would seem that the recommended loading for the Troika is 560Ω (in parallel with 5.6nF) [1], so I cannot be too dogmatic about the Troika being under damped with 470Ω loading. Similarly the Linn LINNK preamp, made for Linn by Naim and based on Naim’s 323 circuit boards, was designed to partner the Asak and presents a load of 390Ω (in parallel with 7.5nF).

    Just as the transformer reduces the input impedance of the preamp by a factor of 100 from 47kΩ to 470Ω, conversely it increases the capacitance of the preamp by the same factor. A typical phono-preamp will have an input capacitance of say 180pF. The effect of the transformer then is to effectively make this 18nF, which again may be too much for the Troika, as the recommended capacitative loading is 5.6nF. Excessive preamp capacitance can modify the treble response by depressing it. This may help counteract the effects of under-damping, but if overdone the overall effect could be a dull sounding cartridge. I think you complained of the cartridge sounding ‘shut in’.

    There is another problem with using transformers. Just as they modify the load impedance, they also modify the source impedance, by the same amount. A typical phono input stage is designed to achieve a good noise performance (and bandwidth) for source impedances < 1kΩ. The effective source impedance of the Denon is thus 4kΩ, whereas that for the Troika is better at only 400Ω.

    Turning to your question about active devices, these are unaffected by either the source or load impedance, so their performance is more consistent. They come with their own set of problems. For example, being active they are an additional source of noise, but overall the total noise (from the cartridge + mc amp) is probably less than that caused by the effective increase in cartridge source impedance.

    Finally, the impedance settings on your AT 650 puzzle me. It would seem that they refer to the source impedance of the cartridge. That is the only way I can reconcile the gain with the impedance. For example the ‘20Ω’ setting provides a gain of x15. With a load impedance of 47kΩ, the impedance presented to the cartridge will be 47000/15² = 209Ω. As a rule of thumb voltage sources should be loaded with approximately 10x their source impedance, thus the assumed load impedance to source impedance ratio of 10:1 is reasonable. The effective capacitance now seen by the Troika is 40.5nF, considerably greater than the recommended 5.6nF.

    Maybe your Troika sounds better to your ears when loaded by 200Ω, rather than 470Ω; so try shunting the Haufe transformer secondary windings with 28kΩ (i.e. 27kΩ + 1kΩ) and see if that works. For the latter calculation, I have taken the resistance of the transformer coils into account.

    [1] http://www.pinkfishmedia.net/forum/a...p/t-24159.html

    Regards
    Barry

  5. #15
    Join Date: Feb 2008

    Location: Canaries

    Posts: 193

    Default

    Wow, thanks Barry.
    I have just read your reply and I think I´ll need to read it another ten times to get my head round that but I´ll go straight to the bottom line and get hold of some shunt resistors. When you say to place them across the secondary, that is the output terminals right ?
    A friend has just lent me his Black Cube SE to test the active angle, too - this is going to be good. The search for my missing bass slam is on.
    Thanks again and will be in touch.
    Chris

  6. #16
    Join Date: Jan 2009

    Location: Essex

    Posts: 32,049
    I'm openingabottleofwine.

    Default

    Hi Chris

    Yes, connect the resistors across the output of each transformer. I assume you are using phono sockets, these then would be a convenient pace to fit the resistors.

    I may have mislead you regarding the capacitance presented to the cartridge by the transformer. 50pF is a more realistic input capacitance for the phono input on amplifiers. Through a 1:10 transformer this will effectively become 5nF; virtually the same as recommended by Linn. Your AT-650 transformer on the '20Ω' setting (x15) will change the capacitance to 11nF, a bit high.

    Sorry about the 'information overload' - I'm always doing that. I’ll summarise the main points as follows:

    ● Transformers are designed to perform at their best between specific source and load impedances.

    ● The Haufe 7883 will perform optimally with the Denon, but not with the Troika.

    ● According to other Troika users and Linn themselves, the recommended load for the Troika is 560Ω (in parallel with 5.6nF).

    ● The AT-650 on the ‘3Ω’ setting ought to be a good match for the Troika. You prefer the ‘20Ω setting, so try 28kΩ shunting resistors across the output of the Haufe 7883s.

    ● The performance of active devices (so called pre-preamplifiers) is independent of the source impedance of the cartridge feeding it as well as the load impedance of the phono input of the main amplifier.

    Regards
    Barry

  7. #17
    Join Date: Feb 2008

    Location: Canaries

    Posts: 193

    Default

    Hi Barry,
    Just to let you know the outcome. I shunted the secondary on my Haufes with 27kOhms and yes, it improved, cleaning and opening up a fair bit. I still prefer my AT on the 20 ohm setting though, finding the 3 ohm setting too overpowering, not in gain, but just not inducive to relaxation. I have also tried a Black Cube SE active phono stage but it didn´t do anything for me, I suspect my PureSound P10 is much better all round.
    So I will carry on with my AT-650 until such time as I find an active pre-pre stage. A friend has just bought the Hagerman Piccolo and I´ll wait and see what the verdict is - he uses a Karma which is not a million miles from a Troika.
    Once more, thanks for all your help.

  8. #18
    Join Date: Jan 2009

    Location: Essex

    Posts: 32,049
    I'm openingabottleofwine.

    Default

    My pleasure Chris,

    Glad to hear that you at least had partial success with loading the Haufe transformers. Again let usknow how you get on with the Hagerman amp.

    Regards
    Barry

+ Reply to Thread
Page 2 of 2 FirstFirst 12

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •