Alignment questions for the experienced

[ConradH]

I'm learning to do alignment with my "junkyard dog", a Sansui 2000x, found for free at the dump. It's actually a very nice unit. I'm starting with the AM just to get a better feel for things, and the following questions arise:

1) Do service instructions commonly contain errors or make assumptions about "general alignment knowledge" that makes it hard to simply follow them as written? The 2000x manual shows a test point for the AM IF alignment that simply isn't where one finds a suitable IF waveform. OTOH, the factory seems to have soldered a wire loop at the tail of the IF that works great.

2) They say to couple in at the antenna terms, but don't mention the fact that unless you tune to the low end of the scale, no signal is going to make it to the IF. They also don't mention that one can probably get cleaner results by killing the oscillator temporarily.

3) There are 3 IF cans with 2 slugs each. That's 6 slugs to play with to get a good waveform shape. The shape isn't super well defined, but is obviously broader than a simple peaked alignment. Is there any deterministic procedure for tweaking the 6 slugs to get the desired shape? Though I can get there by random fiddling, that's not very satisfying- have I truely got maximum response with the right shape, or just sloppy tuning that looks sorta right?

4) I see a slight shift in IF frequency when changing the amplitude of the signal. Is this normal? The shape of the waveform also changes significantly. Is a minimum signal, say down near the noise, the best signal to align with?

Thanks!

 

[gadget73]

1) Sometimes. It depends on the specifics of how it wants you to align it. Are you supposed to be monitoring AF output or RF output on the AM side? Audio level output is pretty common, though admittedly I haven't seen service procedures for much made after about 1965 so maybe they do it different now. I can tell you that often tying the scope straight to an IF transformer can load down the transformer in such a way that its actually not properly aligned if you remove the probe. Usually using the 10x or 100x (if you have it) probe helps considerably, or you can cheat and use a 100k resistor or so to decouple things.

2) it actually wants you to couple in the RF gen at IF frequency at the antenna and use that to peak the strip? Many sets I'm familiar with (again, waaaay older than what you're working on) have the IF injected at the mixer and you peak it that way. There was also sometimes a step involving coupling to the antenna, then adjusting a wave trap for a null to prevent any signal that happened to be at the IF frequency from blasting through the front end.

3) best way to get a symmetrical waveform is to use a sweep generator. Not sure you can really get it super clean with a fixed frequency gen, though on AM this is less critical. In the old days, the usual method was to use a modulated RF generator and simply adjust for peak volume at the speaker, or peak for strongest AGC action. With FM, a proper waveform through the IF strip is critical for proper performance, and you really need a sweeper to get this.

4) I usually like to use the least signal possible. Running it up too high will start to kick in the AGC circuitry and you can end up with an improper peak. They used to include a mention to decrease sign gen output as necessary to keep the output on the meter under some level to prevent this from happening.

 

[ConradH]

Thanks! I am sweeping (linear) with a Wavetek 185, going from 430 to 480, to match the divisions across my scope at 5 kHz/div. They say to couple in via the antenna terminals, and monitor the (rectified & filtered?) IF envelope through a resistor at the end of the IF strip. I use a 10X probe (also have 100X) and signal level is no issue because I use a high sensitivity diff plug in for the scope. Thus, I think I'm seeing "truth" without affecting the circuitry much, but my confidence level about the correctness of the response shape isn't what I'd like it to be. At some point maybe I can add some scope photos. IMO, this is why experience is so essential, since their (minimal) instructions can only get me so far. :thmbsp:

 

[ConradH]

Well, the AM seems to be as happy as can be. Now I'm starting the non-stereo FM parts. This has also been no real problem, but the service manual is still confusing me. They say to do the IF with 10.7 MHz +/- 200 kHz sweep. That makes perfect sense. For the rest of the procedure they say things like 90 MHz 400 Hz 100% modulation. I don't understand. It's FM, so there is no 100% modulation, only deviation. Certainly an AM signal would make no sense, so what signal are they asking for? I did the procedure with a very low level signal and just enough 400 Hz FM to get a clean tone, and that seemed to allow peaking things with no problem. In fact, the only big problem with this tuner is that I put on the 'phones and end up wasting an hour just listening to it. It's just remarkably good!

 

[Jon_Logan]

Sure you can 100% FM modulation. It's just frequency modulation rather than Amplitude Modulation. The 100% modulation means the deviation is as deep (+/- frequency) as the spec allows.

So, what they're asking for is a modulated signal that is centered on a broadcast frequency, but is modulated about that frequency by a certain amount. They say +/-200kHz. This allows you to view a signal on a spectrum analyzer so you can check/adjust the IF for an even response in the passband.
Dual sweep allows you to use a scope to view the shape of the IF bandpass in a spectrum analyzer type of view.

 

[terra1]

It may be helpful if it is an online manual to post a link for others to follow. They usually list the equipment and signal generators in the service manual.

I am not a tech or expert, but I imagine you can check your work simply by tuning to a station to check your work and not get too caught up on the technical side. The goal is to get it to sound good and if it's "remarkably good," then what's the problem?

I don't have any of your equipment, but all I use on the AM side as a signal generator is the existing AM antenna rod (which usually has a slot for adjustment) on the back.

I use a pair of rabbit ear antennas for my FM signal.

From there I tune into a station in the middle, upper, and lower frequencies where I can get a strong signal. I follow the steps in the service manual as far as I can go. Naturally I ignore the multiplex steps that require an oscilloscope. Without the proper equipment, I am not going to be spot on without the proper equipment, but I can usually manage to get the tuner to get maximum signal strength, have the tuning meter needle center on FM, and get the holy grail Stereo indicator to light. And I can get the tuner to sound "pretty good." Disclaimer and warning: the devil is in the details.

I am not advocating not using proper equipment. Just giving you a step back and admire your work point of view.