Linn "LINTO" MC Preamp Schematic without SMPS - Schematic for SMPS wanted

Looks like that at the moment. I will have to change a bit the housing - no problem. Replaced 14 electrolytic caps with 22 uf tantalum and opamps.
Cheers,
 

Attachments

  • CAIRN PS.JPG
    CAIRN PS.JPG
    533.6 KB · Views: 1,255
So far I never understood why a low power line amp (or phono amp) would benefit from a SMPS.
there is an advantage in all cases by using of SMPS, where an indoor power supply and transformer is in use near by the line/RIAA unit (in opposite to a classical 50 Hz transformer).
But the use of a power supply in the same housing (i. e. near by the voltage amplifier stages) is generally considered by me to be very unfavorable, even by power amplifiers.
By an outdoor power supply I don't prefer SMPS by the currently available clock rates.
Nevertheless most brands don't prefer seperate outdoor power supplies (unfortunately).
Looks like that at the moment. I will have to change a bit the housing - no problem. Replaced 14 electrolytic caps with 22 uf tantalum and opamps.
Cheers,
Great work. Unfortunately the showed device is the PCB of the line/phono preamplifier model "KAIRN" and not that one from LINTO. The still needed Schematic is that one of the SMPS device from LINTO.
The first picture shows the embedded SMPS, from which is the created schematic.
The second picture shows an other (older?) "Brilliant"- version from Linn. This schematic I don't have !!
Here some additional pictures of the first picture from there, found about Linn Brilliant SMPS - naked! - pink fish media
(except the last pic - from this version I need also a schematic diagram).
 

Attachments

  • Linn Brilliant open.jpg
    Linn Brilliant open.jpg
    76.2 KB · Views: 1,232
  • Linn Brilliant component side.jpg
    Linn Brilliant component side.jpg
    152.9 KB · Views: 1,184
  • Linn Brilliant solder side.jpg
    Linn Brilliant solder side.jpg
    124.2 KB · Views: 1,189
  • Linn Brilliant first edition.jpg
    Linn Brilliant first edition.jpg
    38.7 KB · Views: 1,088
Last edited:
there is an advantage in all cases by using of SMPS, where an indoor power supply and transformer is in use near by the line/RIAA unit (in opposite to a classical 50 Hz transformer).
But the use of a power supply in the same housing (i. e. near by the voltage amplifier stages) is generally considered by me to be very unfavorable, even by power amplifiers.
By an outdoor power supply I don't prefer SMPS by the currently available clock rates.
Nevertheless most brands don't prefer seperate outdoor power supplies (unfortunately).
Thanks for the comment. In my case I do not have a RIAA stage but I am considering to take it out - I don't like the present arrangement. Will be easy to make a separate housing next to the preamp. Not an easy job with my three LK 100's but the most difficult is to take the decision - afterwords just copy-paste.
Cheers,
 
Just a matter of curiosity - and my transistor circuit theory is rather rusty:
Perhaps one of you learned folk can answer the following.
If I look at the input circuit of the Linto signal circuitry (attached to the first post of this thread), I am puzzed. The very first signal transistor - when the audio comes from the moving coil cartridge does not appear to have a built in DC bias. The base is effectively floating being isolated from the outside world by a couple of capacitors.
How can this be? I thought stable transistor circuitry required a stable bias??

Sorry for what might me a dumb question to some of you...

don
 
Is it possible by replacing the switchmode power supple with a transformer?
No. You need additional a rectifier bridge or four pieces of high speed diodes, capacitors and voltage regulator, both for positive and negative voltage rail. Instead a serial regulator I would prefer a parallel regulator. And instead one supply for both channels I would prefer two monaural power supplies to optimize the GND management.

schematic for "LINTO" SMPS still needed

no one here who has now opened this module and drawn the schematic diagram?
 
Last edited:
No. You need additional a rectifier bridge or four pieces of high speed diodes, capacitors and voltage regulator, both for positive and negative voltage rail. Instead a serial regulator I would prefer a parallel regulator. And instead one supply for both channels I would prefer two monaural power supplies to optimize the GND management.



no one here who has now opened this module and drawn the schematic diagram?

Fully agree but I have just replaced the diodes with hyper fast ones paralleled with 0.1 uF and simply delivered the required voltage by an R-core trafo - works fine so far.
 
hi Niamex,

my Kairn SMP is also broken, could you please tell me the transformer each pin output voltage? so I can find a fitted a normal toroidal transformer. thanks.

Sorry I am late, it was long time ago and I do not remember exactly the voltages but you can see the photos and judge what to do - definitely new hyper fast soft recovering diodes /Fairchild/ and an R-core trafo.

Regards,
Ignat
 

Attachments

  • KAIRN.JPG
    KAIRN.JPG
    547.1 KB · Views: 677
  • KAIRN2.JPG
    KAIRN2.JPG
    578.6 KB · Views: 663