sx-780 Tuner drift

[larryderouin]

Have a 780 here, I recapped.(including the tuner). It still has some tuner drift (always to a higher freq) in FM only, after about 10 minutes. I can hear it drift off frequency, by loosing Stereo and then the station totally. It takes about 10 minutes each time. This problem was there before the recap and continues after the recap without change.

I was reading this thread and was wondering if the 2sc461 in the tuner front end would be causing this problem and if the KSC945 is a suitable replacement for THIS problem. Or should I go with a BC549?

 

[Punker X]

Have you tried cleaning the front end yet? Read the sticky at the top of the tuner forum for instructions. It will reduce the most if not all your drift. I would not under any circumstances replace components in the front end, unless you have exhausted all else.

 

[larryderouin]

I've cleaned the front end already with CRC QD Electronics non residue cleaner. 1st thing after vacuuming unit and last thing I did during cleanup of the unit and after the recap. Also added a 1/2 drop of Faderlube to the shaft clips with a 24ga 1" needle to minimize grounding problems (I do this with all of my tuners.) Other than the upward drift, it works fine. Voltages on the board agree with those on the schematic +/-1volt max. All variances are consistant.

Larry

 

[rcs16]

tuner drift

Hi Larry,

Most ecap's in a tuner usually have nothing to do with drift. They are used as supply de-coupling and signal coupling. The drift is most probably in the local oscillator, which can be caused by the associated components. This is a fundamental problem with this design. Components change in value with temperature and their is no means of compensation. It is possible that the LO supply is drifting around, so could measure both DCV and ACV supply to see if rock stable.
2sc461 is a VHF bjt, so ksc945/bc549 are not very good choices. I have attached the DS for the Hitachi bjt that is used.
I have to download the SM to look at this ckt, to comment more, but it is the same as all the others. Not sure if this one has a jfet buffer like my 950 does.
Will get back to you.
Rick

 

[rcs16]

Larry,

Now that I think about it a bit more & looked at the ckt.
Hybrid, opamps,STK050, tuner looks standard low cost Pioneer, using PAxxxx parts.
Has this tuner been checked for alignment, which usually means fiddling with the coil slugs on one end of the band(Lo), then adj the vcap trim C's(hi end), the ones that were painted set,
set IF coupling xformer for max sig at rec o/p. Then adj the quad coil T2.
Especially if you have did anything to these sensitive fins, even spraying air, anything.
Look at the procedure on sec 8,p14
This gets better, no tuning slugs, adjust fins, you have be kidding me, special spatula too. = sharp toothpick
Mark have you did this before?

This why I hate these things. consequently designed my own tuner.:thmbsp:

It should drift more as it warms up, then if it has been sitting for some time, to temp stabilize? Kore or less drift?
Recommend to follow the FM adj procedure and report back, before you touch any more parts. Only 3 fins to fiddle with
Have fun
Rick

 

[Watthour]

I've tweaked the VC fins on several of these just to regain linearity through the tuning range. It's a PITA to get even close to correct. I've been able to get them to within 0.5 MHz indicated on the scale from 88.7 to 105.7. This is, of course, after a thorough cleaning and drying and re-lube of the bearings and gears. It's most difficult to open or close the fins at one end of the scale without affecting the rest of the range.

I'd agree that the design is pretty spartan but it can be made to work, albeit not easily.

Regardless of that, the linearity adjustment doesn't necessarily have any effect upon drift. That's almost surely a thermal change in components, be it a cap or a choke.

 

[rcs16]

There ya go Larry, you found a fellow fiddler in watthour.

pretty spartan

I like the analogy, designed by Pythagoras

Larry,
Do you own a FM SG?
Are L1-5 in shielded cans or just air wound coils. Like the old days, with sponge and wax? Hard to see from the SM.

 

[Watthour]

Waxed? They're not quite that frugal. Like this:

You may notice that the FM RF sections have the fins splayed a little while the IF section is straight. "Someone" has been spudging in there to restore a little linearity.

 

[rcs16]

Thx for the pic, helps a lot.
while the IF section is straight. :scratch2:
LO is under the RF shield(dark grey) , top/left corner of var cap.
I can see the FM one in the middle, (the 3 fin ones) 2nd tuned ckt, VC2 right side is splayed, noticably)
You can tune the coils from the looks of it. L1,2,3,4, L5, I can not see as it is under the dark grey shield for the LO. Notice the LO trimmer C?
So lots to muck with.

 

[Watthour]

Actually, the LO trimmer is just next to the shield. The parallax of the photo might make it look like it's under there, but it's a straight shot in with the trim tool.

Those are all pretty standard, but once set correctly (or close) the stations should not drift. I'm guessing something is changing due to thermals. It might even be slight movement in the air core turns. Maybe a wax slug wouldn't hurt after all, but I'd suspect a cap or ceramic filter first.

 

[rcs16]

but I'd suspect a cap or ceramic filter first.

= method of substitution!!:scratch2: = PITA
have fun Larry, no definitive answer

 

[larryderouin]

There ya go Larry, you found a fellow fiddler in watthour.

I like the analogy, designed by Pythagoras

Larry,
Do you own a FM SG?
Are L1-5 in shielded cans or just air wound coils. Like the old days, with sponge and wax? Hard to see from the SM.

No Siggen. It's dead on across the dial as far as frequencies (linearity) are concerned. No I haven't adjusted anything on it. 88.1mHz is between 88.0 and the 1st mark past it (88.2) and 107.5 is between the 2nd and 3rd marks after the 107 long mark on the dial. Everything in between is dead on. I don't want to mess with it if I don't have to. But this drift is annoying more than anything else. It's on every station I tune to on the dial, not just one. I listen mainly to 88.1 and 100.7 in the Baltimore area and occasionally to 106.5. It will drift upwards, never downward after 10 minutes, continuously whether cold or hot. This to me is indicative of the Oscillator not working correctly. Not a linearity problem.

Larry

 

[Marty Socal]

Larry, Maybe you could try a little cold canned air on the various components after it drifts off to see which one is being effected by the temperature change?

Thinking outside the box and a little off subject:
Could a better spec Tuner Board off a higher line Pioneer Receiver or Tuner be swapped into a low/mid level Pioneer receiver like the SX-780? :scratch2: Looking at the tuner board schematics of my SX-1050 and the SX-780, show they both run off 13V. It obviously would take some creative mounting bracket fabrication and verification that it would physically fit, but most of the wires would connect to the amp sections in a similar manner. The butchers on epay list them from time to time, usually they are not that expensive. (The gears are grinding in my head )

 

[rcs16]

I agree, the LO is the first on the list. Next IF, possible CF's or quad detector.
Better models used a LO buffer, so that there is no pull from strong sigs I believe that was the goal!!
I would try to tune the quad coil, to see if that makes any diff. steps 16-17. Two cores in the double tuned transformer. Need to access both sides. Must be a hole through the pcb? use a hex nylon tuning tool. Need a DA as well, got one of those?
So do you change the bjt and ceramic caps, maybe chose NPO/COG for zero temp co?
A trail and error exercise, but with no SG, is this possible to do?
From my 950 notes,
Q3 (2SC717) VHF Oscillator (Hitachi,use 2sc1906 for new designs!)
Q4 (2SC461-B) VHF Amp
Some options for the LO bjt
http://ca.mouser.com/Semiconductors...F-Bipolar/_/N-ax1s5?P=1yxv7v7Z1z0y4diZ1z0z63x

 

[Watthour]

Agreed. LD. Tracking/linearity seems right on by your description.

If you have a 'scope in your trailer can you connect to the LO at the collector of Q3? Start it cold and tune for center channel and check the wavelength, then monitor it while sipping your favorite beverage as she heats up. It might be revealing. You might also want to mini-grab in there with your voltmeter and watch the 13V circuit supply into Pin 22 for changes at the same time.

Just guessing.

 

[Watthour]

... Two cores in the double tuned transformer. Need to access both sides. Must be a hole through the pcb? use a hex nylon tuning tool. Need a DA as well, got one of those?

Yes, the board is punched to access the lower core.

 

[rcs16]

I have SX-950,SX-1050 & a Sansui G-7500 tuner boards!!
If I knew an easy way, it would be great to through my Si4735 tuner in there.
Take the tune meter out, put in a small display for freq. replace the tuner fly wheel with a rotary encoder. Anyone game?

 

[rcs16]

Hey I have a brain fart, we need a AFC ckt, doesn't the 1980 have one we can use

 

[larryderouin]

Q3 is under the shield. I've got a single probe 10mHz scope. Haven't ever used it tho. That, a VTVM and DMM are all the testing gear I have.

Let me see what I can do with the scope, and a DMM, and note any changes.

Larry

 

[Watthour]

The 10 meg o'scope should read the 10.7MHz LO, and will display several waves in the raster. Just set the trigger and timing so the waveform display is stable and note the wavelength on the reticle. If it begins to grow or shrink (more likely) it should be evident.

Look at the schematic and you'll see that you should be able to read the LO at the bottom end of the coupling transformer, C16, R14, R15, and maybe some other places. Bear in mind that the scope probe may affect the oscillator tank and require you to re-tune. The point is to look for change as the FM section thermally changes.