Technics SA-828 Receiver from 70s- Is it worth trying to save this aging patient?

[dorke]

The photo isn't clear enough.
If possible, a better pic please.
Most TRs have the logo of Mitsubishi (the triangle)
Q631/Q633 suffix could be Q or R,very hard to tell.
Q625 could have a "T" suffix,verify with the other channel Tr Q626.

In general,another better way to get the suffix is to look at the corresponding transistor of the good channel.
Simply add 1 to the part number like so:
Q625 ---> Q626 on the other channel etc.
No need to remove the part,a flash light magnifying glass,and a small "dentist mirror" could be handy for the hard to access parts.


The list of replacements should have suffixes as well.
Finally that should be verified with Digikey before purchasing.
If they don't have the correct suffix,another solution should be considered.

 

[Iceman from McMurdo]

Sorry, trying again with better photos (closeup).

Q625 sure looks like T (definitely not an R or S). Unfortunately, can't check correlating part on other channel (even with mirrors & magnifying glass). It's under a heat sink and mounted in a parallel row with two others. Have a bad feeling about trying to unsolder everything including bat-wings on that (WORKING!) channel just to check a suffix. I'm going with a T...

 

[Iceman from McMurdo]

Q631 & Q633 photos should be readable now (definitely NOT an alphabetical suffix). See if those numerical markings help any...

 

[dorke]

About Q625,
Don't un-solder anything.
I see you have Q623 with an "S" suffix,
what suffix is there on Q624?

Q631/Q633 look like a "modified" Q , more then anything else.
Same on the other channel Q632/Q634 ?

 

[73's de Edd]

I forgot . . . . it's been so far back, but isn't the problem with the left channel now, and with the right channel still operating properly.
If so is there a high probability that the right channel has ALL of its Original Equipment Manufacturer semi conductor devices present, and it if so, just use the markings of a companion transistor function of the right channel and get its suffix markings.
Your driver transistor last shown is from the Matsushita family and that '913 is only offered with P,Q,R and S suffixes and you have a " Q " there with a Beta span of 90-155.

Now your very special puppy the '745 only comes with RST suffixes and your T falls in the 360-700 Beta span.







73's de Edd

 

[Iceman from McMurdo]

Yep, left channel is kaput, and right channel is still operating perfectly. I'd definitely like to keep it that way, so unsoldering components just to confirm suffixes is not worth it. Unfortunately, the TRs most in question are small and mounted super close together in parallel under a little L-shaped heatsink, so it's impossible to read anything off the ones we'd like to verify (namely Q622/Q624/Q626).

That said, I double-checked Q623 again and it most definitely has an S suffix, so there's no need to check its twin Q624 anyway.

As for Q631/Q633, thanks for narrowing the choices down (P/Q/R/S) - it's definitely a Q suffix (though an oddly shaped one, ha!) Oh those creative Japanese electronics wizards...

And if you say Q625 only comes with R, S, or T - then it's most definitely a T suffix.

Think that completes all the suffix checking. THANKS!

 

[dorke]

For Q625,Q623 we can use suffix "S" (or "T") whichever is available.

Q621 2SD661-S or 2SC1815-GR

Remember the matching pairs are:
Q635/Q637 Sanken 2SA1215 -O / 2SC2921-O (suffix is the letter "O")
Q631/Q633 couldn't find Sanken 2SA1667/2SC4381 suffix data.
Q627/Q629 On-semi/Fairchild KSC2383-O /KSA1013 -O (suffix is the letter "O")


You can now complete the list and contact Digikey for appropriate replacements.

 

[Iceman from McMurdo]

Ok, got some good news... and some bad news on the resistors. Here are the test results (in ohms):

R633 = 34.5
R643 = 82.3
R647 = 67.8
R649 = OL (BAD! - should be 1/4 100 per schematic)
R671 = 6.5k+ (lower than 8.2k shown on schematic, but it was still climbing when I ended the test)
R673 = 1k
R677 = OL (BAD! - should be 1/4 220 per schematic)
R679 = 6.1
R681 = 6.1
R683 = 2.7
R685 = 2.5

Checked test results against resistor values shown in schematic and all were close enough to be judged good EXCEPT the two obvious shorted ones (R649 & R677). Even though the test result on R671 appears lower than specified 8.2k, I think it's probably good as the values were still increasing (albeit slowly) after a few minutes & I got tired of holding the test probes in place. Checked the rise rate on the corresponding resistor on the good channel and it read the same way, so I see no need to replace R671 do you?

Man, my tired old eyes ache after trying to read that darned schematic with a magnifying glass! I think that would get real tiring on you professionals in very short order... But it was definitely a good call to check them, as that's two more parts I'll need to add to the order list. Thanks!

 

[Iceman from McMurdo]

BTW, never ordered a resistor before... How do I search for these two I need? Schematic identifies them as:

R649 = 1/4 100
R677 = 1/4 220

Does the 1/4 represent a power rating or something (like maybe 1/4 watt)? Can anyone point me to possible part numbers or manufacturers that might help hasten the search process?

 

[dorke]

Yes 1/4 is indeed 1/4 watt.
We need to know the accuracy of the resistors,could you please post a pic of the corresponding resistors of the good channel.
Resistors can be bought from any manufacture (unless they are very special type) from Ebay as well.

Please check the following as well:
R625,R629,
R663,R665,R667,R669,
Turn VR603 fully CW (clockwise) and measure it's resistance ,then fully CCW(counter clockwise) and measure again.

BTW,
O.L means an open resistor,not a shorted one and they usually look burned.

 

[dorke]

Just to be on the safe side.
I would re-check all "surviving Trs" :
Q605,Q607,Q609,Q611,Q615
and diodes:
D603,D605,D607,D611,D613,D615,D617.

 

[Iceman from McMurdo]

OK, last band of bad resistors is gold, so if I did my homework right that means their rated tolerance is 5%, correct? Upon close examination with magnifying glass, both R649 & R677 did appear kinda toasty & discolored in appearance, so I confirmed the gold color on right channel equivalents (see pics of R650 & R678 below).

Also checked remaining resistors you mentioned and looks like we lucked out - all appeared very close to spec values on schematic:

R625 = 99.6
R629 = 99.7
R663 = 6.5k+ (and rising, so expect it to top out around spec value of 8.2 if I waited long enough)
R665 = .775k
R667 = 1.4K
R669 = .775k

As for VR603, full range resistance values are:
Full CW = 4.3
Full CCW = .845k
Also checked several midpoints & got proportional values, so it appears to retain its variable range.

Will try and finish testing remaining TRs and diodes tomorrow...

 

[dorke]

Yes,gold is indeed 5%.
Please check R659 and R661 as well.

 

[Iceman from McMurdo]

Just checked those last 2 resistors you mentioned - found another dead soldier in there:
R659 = .464k
R661 = OL (should be 1/4 470 per schematic; confirmed 5% by checking gold band on counterpart R662)

Moving on to check remaining TRs & diodes now (got my fingers crossed)...

 

[dorke]

O.K
Please check R651 and R653 as well.

 

[Iceman from McMurdo]

OK, here are the TR/diode results (got a bad feeling about Q609, but will await your confirmation). As usual, first 3 readings are red probe to base, second 3 readings are black probe to base.

Q605
B-C = OL
B-E = OL
C-E = OL
B-C = .608
B-E = .608
C-E = OL

Q607
B-C = OL
B-E = OL
C-E = OL
B-C = .625
B-E = .625
C-E = OL

Q609
B-C = .000
B-E = OL
C-E = OL
B-C = .000
B-E = .581
C-E = .581

Q611
B-C = OL
B-E = OL
C-E = OL
B-C = .605
B-E = .624
C-E = OL

Q615
B-C = .587
B-E = .602
C-E = OL
B-C = OL
B-E = OL
C-E = OL

Diodes all appear to check good:

D603 = OL / 571
D605 = OL / 625
D607 = OL / 567
D611 = OL / 559
D613 = OL / 567
D615 = OL / 251 (perhaps germanium?)
D617 = OL / 247 (perhaps germanium?)

Whaddya think?

 

[Iceman from McMurdo]

Follow-up resistor checks - R651 & R653 both check good:
R651 = 4.59k
R653 = 4.61k

 

[dorke]

O.K
All Trs and Diodes look fine.
Including Q609 which has it's B and C wired together in the circuit,
and D615/D617 which are Germanium ones.
Q607 is a 5 pin device(dual Tr),so it's measurement isn't complete here,
but I believe we have checked it before.

3 more resistors need to be replaced:
R649 = 100 1/4w
R677 = 220 1/4w
R661 = 470 1/4w


Please check resistance of R675
It should read about 2.94K.

One more thing to check would be voltages on IC601(8 pins) relative to GND with a DVM in 2VDC and 20VDC range.
With unit powered up and volume at minimum.
Take care -no probe slips/shorts!

Don't remember asking before.
Is it possible that you may have an oscilloscope?

 

[Iceman from McMurdo]

Wow, guess I dodged a bullet there on Q609 - thought for sure the news was gonna be bad. Good to hear it's still alive! (And, yes, we have checked Q607 previously and it was good).

Already got the 3 resistors added to my parts list, thanks.

Been looking all over for that darned R675 and can't seem to find it on the PCB. I see it on the schematic and found all its odd-numbered brethren above and below, but it must be hiding. Probably has its cloaking device activated! Will go back and try again...

And that's a negative on the oscilloscope.

 

[Iceman from McMurdo]

Ha, found it! Correctly deduced from schematic that it should have been mounted right next to TH601 (had to look up what one of those was!) and once I found the thermistor, I saw the little R675 hiding right next to it (covering its identifier on the PCB). Think it checks out ok, as the reading was up to 2.6k and climbing before I got tired of holding the probes in place.