Setup for IF Alignment

Discussion in 'Tuners' started by two.dogs, Mar 11, 2018.

  1. two.dogs

    two.dogs Active Member

    Messages:
    352
    Location:
    Denver, CO
    I just did my first successful alignment of a Sansui 1000X. Now I'm trying an Onkyo 4500 but I'm having trouble on the first step. Can someone give me a hint? I aligned the Sansui on my ST-1000A and never tried to display the S-curve. The ST-1000A produces a display that looks more like a loaf of bread. I haven't been able to get anything approaching the classic S-curve I've heard so much about.

    IFalign.png
    I'd love some more detail here. I've got a nice sweep generator but unsure how to set it up. 10.7 MHz +/- 200 Hz at .1 mV is my best guess.

    tuner.png
    I have the signal generator connected to a 50 ohm terminator, then to an alligator clip on the IF pin of the tuner board (upper left on the schematic). The oscilloscope is clipped onto R111 near Q124 (bottom rightish on the schematic)

    signalin.JPG
    Here is the input on the board.

    quadout.JPG
    Here is the output on the board just past the quadrature detector.

    signalout.JPG
    I'm guessing that this arrangement of capacitor and resistor is basically a time constant to give me a Smooth S-curve output rather than a sinusoidal output of varying amplitude? One margarita down, probably several more to go.

    Z.JPG
    This is the Z output from the back of the signal generator that synchronizes to the sweep. I'm thinking that I have to trigger off this signal, then display the quadrature output in the vertical axis. I've played with XY mode and other combinations of scope settings with no luck.
     
    Last edited: Mar 11, 2018

     

    Please register to disable this ad.

  2. txturbo

    txturbo AK Subscriber Subscriber

    Messages:
    914
    Location:
    Austin, TX
    What type of probe are you using between the detector output and the scope?
     
  3. two.dogs

    two.dogs Active Member

    Messages:
    352
    Location:
    Denver, CO
    I soldered a clip to a red wire, then connected that to a 10K resistor and .47uf capacitor to ground as directed (first pic). Then I connected a 1X/10X oscilloscope probe to the junction of the resistor and cap. You can see it in the second to last photo leaning against the tuner pulley.
     
  4. rcs16

    rcs16 Super Member

    Messages:
    3,800
    Location:
    Baldwin, Ontario, Canada
    I do not think I'd worry too much about the "S" curve. It is kind of an old method, for folks without a thd analyzer. But since you are playing with it, lower the modulation rate to around 22KHz, mono, see what happens.
    I suggest to use a different method for a quad detector adjustment, first adj T101, usually bottom core for AFC zero (DMM across TP1 with a noisy, no input signal, adj for closest to 0VDC) and then adjust for linearity (lowest 1KHz thd), top core, using mono modulation at 75KHz deviation.
    Do not use the 3325 as your audio source, it has too much thd, use the distortion analyzer(1700B) as your source for modulation. Once adjusted you can then compare different modulation sources and see what the thd is for them.
     
    I LIKE MUSIC likes this.
  5. two.dogs

    two.dogs Active Member

    Messages:
    352
    Location:
    Denver, CO
    Now I'm forced to admit that I've never used the ST1700. I'll have to get into the manual. I also didn't know to avoid the 3325 - I thought it was supposed to produce a pretty clean sine wave.
     
  6. rcs16

    rcs16 Super Member

    Messages:
    3,800
    Location:
    Baldwin, Ontario, Canada
    "I LIKE MUSIC" is following along, he has experience with this stuff too.

    I also have a HP 3325A, it is a relatively clean source but it is not in the same league as the ST1700B as far as thd goes. 3325 is perfect for freq response measurements, as a general source, etc.
    I do not own any ST test gear but most of these instruments op the same way.
    There are a number of old ST documents on the net, for the operation and use of these instruments. Stan Curtis web site comes to mind.
    http://www.stancurtis.com/soundtech.htm
    Suggest to loop the 1700 back to itself and measure its inherent thd,
    then use the 3325 as the source to compare against. make sure you terminate the 3325 with a 50 ohm load at the load end usually.
    then connect the 1700 o/p to the 1000A ext modulation, select ext modulation, adjust level(s) to get 75KHz peak deviation, mono mode, then RF o/p to tuner input. Usually set RF level high 50-60dBuV (~1mV) so the signal is way above noise floor.
    Connect audio o/p, tuner or tape rec o/p usually to the 1700 i/p, set sig level to a normalized level in ACV mode, then switch to thd mode and null out the fundamental, measure residuals. See what you get before any adjustments. then adjust linearity for lowest thd.
    Once you get the hang of it, you should be off to the races.

    Good Luck
    Rick
     
    Last edited: Mar 14, 2018
    cpu_ and KentTeffeteller like this.

     

    Please register to disable this ad.

  7. AudioWizard

    AudioWizard AK Subscriber Subscriber

    Messages:
    182
    Location:
    Reno, NV
    Look for the "How to align Stereo Receivers" brochure from Sound Tech. It is very detailed on using the 1000a. Basically you align for the squarest "loaf of bread" for the discriminator. ST's way is really aligning for the lowest distortion.
     
    KentTeffeteller likes this.
  8. two.dogs

    two.dogs Active Member

    Messages:
    352
    Location:
    Denver, CO
    I found the Stan Curtis site before I bought the ST equipment. I read the "how to align" article and put what I think is a decent alignment on my Sansui 1000X. This Onkyo has some ICs instead of discreet components and has some fancy features like quartz locked tuning and a circuit that detects when your hand touches the tuning dial. So it seems trickier than the Sansui or at least different.
    The first alignment step in the Onkyo manual is the quartz locked circuit alignment that I pictured above. In later steps the manual covers the front end alignment. I think the reason the manual calls for injecting the 10.7MHz is that the front end is assumed to be out of alignment. In any case, that's why I reached for the HP 3325 - it can produce a swept 10.7MHz signal where my ST1000 doesn't go that low (just the FM band from about 88-108MHz). I didn't even think about using the ST1700.
    So, if I'm hearing you guys right, we won't inject the IF directly. Instead, we'll go in through the front end and let the Onkyo generate the IF. In that case, I think I should align the oscillator first, along with the rest of the front end filters. Then I can get back to the quartz locked circuit. Or am I confusing myself?
     
  9. AudioWizard

    AudioWizard AK Subscriber Subscriber

    Messages:
    182
    Location:
    Reno, NV
    You might see if there is a way to disable the quartz lock while you align. Some times the auto stuff fights you. Usually the lock is enabled when the discriminator is balanced, that is the zero volts on your shown TP1. In those modern tuners there is little or no alignments for the IF, just the discriminator basically. They are having you align theIF/discriminator for balance so that it then when you align the front end you achieve lock when balanced, hope that makes since?
     
  10. two.dogs

    two.dogs Active Member

    Messages:
    352
    Location:
    Denver, CO
    That's what I was wondering.
    I think I made some progress tonight. I worked on the front end instead of the quartz lock. I had the coils and trimmer caps pretty out of whack from tinkering. First lesson learned is that the manual incorrectly says that T5 is the oscillator trimmer cap - it's actually T6. Second lesson is that oscillator coil and trimmer cap affect both the low and high end of the dial. I thought coil adjusted low end and trimmer cap did the high end.
    http://audiokarma.org/forums/index....ver-signal-reception-fix.584682/#post-7698928
     
    Last edited: Mar 15, 2018
  11. txturbo

    txturbo AK Subscriber Subscriber

    Messages:
    914
    Location:
    Austin, TX
    If I am reading the schematic right...

    I think you can effectively disable the quartz lock by finding the wire that connects to the tuning knob and grounding it.

    The s-curve adj looks to be for the quartz lock capture range. The output of the circuit is not part of the audio path so using the standard null and minimum distortion method may not provide a useful result.
     

     

    Please register to disable this ad.

Share This Page