First attempt at FM alignment

Discussion in 'Tuners' started by MattiaM, Jun 10, 2018.

  1. MattiaM

    MattiaM Member

    I have a Technics SA5400X receiver that I got for parts and just completed its restoration.
    The tuner seems to be working, although reception is quite weak, it only picks very few stations and it is pretty noisy.
    I therefore decided to use it to try my first ever alignment.
    I currently do not have a distortion meter, but I figured I would use an oscilloscope and reduce distortion by eye/ear for now, my main goal at this stage is to try and grasp a bit better the workflow/procedures and, mostly, to learn something new.

    I got a service manual with the alignment procedures, I have a scope and I have access to an HP 8644A RF signal generator.

    - First, the FM alignment procedure appears to want me to set the RF SG for different levels of FM percentage modulation. This is quite confusing to me: am I correct to assume that if the modulation frequency bandwith is +/- 75kHz, then 100% modulation is the entire 150kHz bandwith around the carrier frequency and a 30% modulation is 30% of that, i.e. the carrier frequency +/-22.5kHz?

    - Second, the alignment procedure defines output levels in (positive) dB, e.g. SG 98MHz (100% modulation with 400Hz), Output 72dB. Am I correct to assume that I should set the SG for an output of -72dBm? Am I also correct to assume that whenever no output is specifiefd, then I should set the SG at 0dBm?

    I tried setting the SG with a 90 MHz and 106MHz with a 400Hz modulation signal, I connected the SG to the 300 Ohm antenna inputs with a dummy antenna as specified in the service manual, and I can easily pick the 400 signal both on the oscilloscope and through my speakers. There seems to be quite a lot of interference though, and I need to crank up the volume quite a bit to see a clean sinewave.

    I checked the specs of the SG, and it seems to have a quite high FM distortion (I have the Option 04, so I get a 1% distortion), double or example the amount of distortion of an HP8656B. Given that I don't have a distortion meter and the spects of the tuner are around 0.7%THD in stereo mode, I don't think that is too much of an issue for now. However, if I want to play more with tuner alignment, should I get a lower distortion SG?

    As always, thanks for the help.
    Last edited: Jun 10, 2018
    musichal likes this.


    Please register to disable this ad.

  2. buffdriver

    buffdriver AK Subscriber Subscriber

    Seguin, Texas
    Like you, I am new to tuner alignment. I'm sure that some more knowledgeable folks will chime in soon, here's my two cents worth.

    I think that you have the right idea about the % modulation. We usually see AM expressed in terms of % and FM in terms of kHz deviation, but I think that you have the right idea.

    As for the output level, they are most likely referring to dBf, which is decibels relative to 1 femtowatt. From my limited experience using an HP8656B (which allows you to key in the output in dBf, dBm, mV, and uV), around 25-30 dBf is a very weak signal and 55-60 dBf is a very strong one.

    In addition to the HP8656B, I have an HP8903B Audio Analyzer which contains a very low distortion signal generator as well as a distortion analyzer. It is also a very nice AC voltmeter.

    For looking at the audio signal produced by your tuner, have you tried sending a tape monitor output directly to your oscilloscope? If you are getting interference, you should try to find a frequency that does not have a commercial signal being broadcast. Interference from commercial stations will make it more difficult to align the tuner.
  3. 39cross

    39cross AK Subscriber Subscriber

    Beverly, MA
    Good luck with your alignment, I'm sure you will get some advice about your HP sig gen. However, I would caution you about adjusting for distortion without a distortion meter, as the adjustment is very sensitive. There may be a method that can get you in the ballpark without a meter, which I would like to learn about if that's true; however it wouldn't take much to mess it up. There is a lot of tune-up before you get to that stage, which if you do, should help fix the reception problem (assuming it's not an antenna issue).
  4. rcs16

    rcs16 AK Subscriber Subscriber

    Baldwin, Ontario, Canada
    I just did a Pioneer SX-1250 and have tuned radios on the Motorola Comm test floor. On the Pioneer, it was basically one cap that was way off for what ever reason?
    I use a HP 8656B for RF front end adjustments. I was thought to adjust for best quieting sensitivity or a null in the noise, lowest noise indication on the ACV, not max ACV with a modulated carrier as they state in so many procedures.
    You only need the low THD for the detector linearity adjustment, sometimes IF filtering fine tuning. In this case you can use the low THD source as ext mod for the 8656B. There are a high and low LED to indicate the correct ext level.
    The 8656B THD I measure around 0.2%, with a very low ext mod source, so it not the best to use for ultras low THD adjustments. It will work because you just want to see the null in THD as well, so you can get pretty close to optimum.

    To adjust a RF front end like these types, you want to use an AC voltmeter (339A,400x etc) with a analog scale since it is easier to see a null than on a digital.
    Connect the ACV to the record out jack or tuner o/p. Adjust levels if possible, FM muting is always off, forced mono if possible.
    I also listen to the noise on some speakers, "Y" off the tuner o/p or use the 2nd set of jacks as that is another guide to a noise null. Adjust level accordingly.
    Begin the procedure as they state sort of :) but no modulation is required since you are going to adjust for best quieting of the RF carrier freq, adj tuner at high (adjust C's) and low freq of the dial range, 88,108MHz(adj L's) as they say in the manual
    You want to set RF sig level so that as you quiet using the var C/L adjustements, you decrease the RF sig level to get the noise level up out of full quieting,
    8656B has an amplitude increment function, set for say 5uV. I am sure the 8644A does as well. read the manual.
    Since the tuner was already done at the factory, it will be close, unless it had faults or a comp was faulty/replaced, so you probably are in the ball park, hopefully just fine tuning. Usually it is the C's that go off, not the L's.
    Usually have to go up/down in freq a few times to get best compromise.
    Give it a try, see what you come up with.

    Good luck

Share This Page