Pioneer SA-7800 with burnt out resistors

I picked up another SA-7800 the other day. Some people seem to collect stray cats and dogs, others collect weird British or European cars (I do this as well), others collect this old equipment.

Cosmetically, it's in good shape even before I started cleaning it. I opened up the case to blast the cobwebs away with my air blow-off tool, and found three resistors burnt out.

I need another project like a fish needs a bicycle...

Lee.

Fish do need bikes... you didn't know that Lee!? Wow, that faceplate looks absolutely pristine in the photos.

150 ohms each, 1/2 watt, emitter resistors for the driver transistors, as well as NSA transistors by the cooked 560 ohm...
hope the stv-2h survived.

Shame it's missing a switch cover. Otherwise, it at least looks hot... and ran hot, hah.

When I get some time (there's only a few hours a week where the soldering iron isn't in reach of prying little fingers), I'm going to remove Q11, 13, 23, 25 and the outputs Q1 and Q3 and test them. And D7, the STV2H as well.

Or should I just remove all the transistors on this channel and test them?

Lee.

All of them, and the diodes too. And at least any resistor under 5000 ohms as well - one leg lifted to eliminate any circuit effects. 5000 ohms with the voltages in this circuit - we're looking at the wattage calculation for worst case current that could fry the part - which could die before blackening the surface.

Good old NSA amps - even I do 100 percent component testing. :sigh:

markthefixer said:

All of them, and the diodes too. And at least any resistor under 5000 ohms as well - one leg lifted to eliminate any circuit effects. 5000 ohms with the voltages in this circuit - we're looking at the wattage calculation for worst case current that could fry the part - which could die before blackening the surface.

Good old NSA amps - even I do 100 percent component testing. :sigh:

Click to expand...

So, the burning question is, if I'm pulling all the transistors anyway to test them, should I just replace them with all new ones? Thirty year old transistors might age better than thirty year old electrolytic capacitors, but we all know about the "black leg" problem with these small signal transistors.

Do you have a list of replacements handy? If not, I'm not frightened of using the search feature.

Thanks mate,

Lee.

That looks like one of my Kenwoods after two receivers were connected to each other though the speaker connections. Yep, it was my "Duh" :scratch2:"Stupidest thing you've ever done in Hi-Fi" thread answers.

q1,q2 2sa979 dual 100v 50ma .2w 250-800hfe
q3 q4 2sc2291 dual 100v 100ma .2w 100mhz 250-1200hfe
leave these dual transistors alone.

q31 2sa733 60v .1a .25w 180mhz 200hfe
q13 q14 q15 q16 2sa750 50v 50ma .25w 130mhz 450hfe
q19 q20 2sa872a 120v 50ma .3w 120mhz 250-800hfe
q27 q28 2sa1914 90v 50ma .2w 150mhz 250-1200hfe
512-KSA992FBU (ln)to-92 ecb 120v .05a .5w 100mhz 150-800hfe $0.05 ea

q5 q6 2sa912 150v 100ma .75w 200mhz
512-KSA1220AYS to-126 ecb 120/a160 1.2a 20w 155mhz 35-320hfe $0.39 (R=60-120, O=100/200, Y=160/320)

q29 q30 q32 2sc945 60v .1a .25w 150mhz 90-600hfe
q11 q12 q17 q18 2sc1400 100v 50ma .25w 100mhz 500hfe
q7 q8 q21 q22 2sc1775 90v 50ma .3w 200m 400-1200hfe
512-KSC1845EBU (ln)to-92 ecb 120v .05a .5w 100mhz 150-800hfe $0.05 ea

q9 q10 2sc1885 150v 50ma .75w 200mhz 150hfe
512-KSC2690AYS to-126 ecb 120/a160 1.2a 20w 155mhz 35-320hfe $0.40



q25 q26 2sa913 to-220 150v 1.5a 15w 65-330hfe 120mhz
512-KSA940TU_Q pnp to-220 bce 150v 1.5a 25w 4mhz 40-140hfe $0.54 unh unh not fast enough

q23 q24 2sc1913 to-220 150v 1.5a 15w 65-330hfe 120mhz
512-KSC2073TU npn to-220 bce 150v 1.5a 25w 4mhz 40-140hfe $0.54 unh unh not fast enough

drat I missed the : unh unh not fast enough
so:
q25 q26 2sa913 to-220 150v 1.5a 15w 65-330hfe 120mhz
too slow: 512-KSA940TU_Q pnp to-220 bce 150v 1.5a 25w 4mhz 40-140hfe $0.54 unh unh not fast enough

q23 q24 2sc1913 to-220 150v 1.5a 15w 65-330hfe 120mhz
too slow: 512-KSC2073TU npn to-220 bce 150v 1.5a 25w 4mhz 40-140hfe $0.54 unh unh not fast enough

2sa985/a pnp to-220 bce 120v/150v 1.5a 25w 180mhz 35-320hfe orig 180mhz nsa driver
2sc2275/a npn to-220 bce 120v/150v 1.5a 25w 200mhz 35-320hfe orig 200mhz nsa driver
b&d 2sa985,2sc2275, 2sa985a,2sc2275a

***************!!!!!!!!!!!!!!!!!!!!
757-2SA1930(Q,M) pnp to-220 bce 180v 2a 20w 200mhz 50-320hfe 200mhz nsa driver
757-2SC5171(Q,M) npn to-220 bce 180v 2a 20w 200mhz 50-320hfe 200mhz nsa driver
***************!!!!!!!!!!!!!!!!!!!!

b&d
2sa913/a to-220 bce 150/180v 1a 15w 65-330hfe 120mhz sa-7800 nsa driver
2sa1011 pnp to-220 bce 180v 1.5a 25w 100mhz 50-200hfe 100mhz nsa driver
2sc2344 npn to-220 bce 180v 1.5a 25w 100mhz 50-200hfe 100mhz nsa driver



q33 2sc1384 50v 1.5a 1w 200mhz



d1 - 6, 9-16 1s2076 replace: replace 1n4148 100v .2a
d7 d8 stv2h
d17-20 d23 1s2471 replace: replace 1n4148 100v .2a
d21 d22 d24 (prot) 1s2472 55v replace 1n4148 100v .2a
d25 mz-130 512-1N5243B Fairchild 13 V, 0.5W Zener


sa-7800 Power amp GWH118

leesonic, Are you going to recap it? Please post a list if you can!

markthefixer said:

q25 q26 2sa913 to-220 150v 1.5a 15w 65-330hfe 120mhz
512-KSA940TU_Q pnp to-220 bce 150v 1.5a 25w 4mhz 40-140hfe $0.54 unh unh not fast enough

q23 q24 2sc1913 to-220 150v 1.5a 15w 65-330hfe 120mhz
512-KSC2073TU npn to-220 bce 150v 1.5a 25w 4mhz 40-140hfe $0.54 unh unh not fast enough

Click to expand...

Any suggested replacements for these? I read in this thread that you were investigating some TO-123 transistors that would be fast enough (see? I DO do my research). Did you ever find any? I'll by testing the ones I have, but it would be nice to have a replacement lined up just in case.

Lee.

leesonic said:

Any suggested replacements for these? I read in this thread that you were investigating some TO-123 transistors that would be fast enough (see? I DO do my research). Did you ever find any? I'll by testing the ones I have, but it would be nice to have a replacement lined up just in case.

Lee.

Click to expand...

http://www.audiokarma.org/forums/showpost.php?p=6359787&postcount=3

drat, I missed this warning to myself in my notes before i posted them...

markthefixer said:

q25 q26 2sa913 to-220 150v 1.5a 15w 65-330hfe 120mhz
512-KSA940TU_Q pnp to-220 bce 150v 1.5a 25w 4mhz 40-140hfe $0.54 unh unh not fast enough

q23 q24 2sc1913 to-220 150v 1.5a 15w 65-330hfe 120mhz
512-KSC2073TU npn to-220 bce 150v 1.5a 25w 4mhz 40-140hfe $0.54 unh unh not fast enough

Click to expand...

757-2SA1930(Q,M) pnp to-220 bce 180v 2a 20w 200mhz 50-320hfe 200mhz nsa driver
757-2SC5171(Q,M) npn to-220 bce 180v 2a 20w 200mhz 50-320hfe 200mhz nsa driver

$0.57 ea, but the 757-2SC5171(Q,M) is a 1/21/2013 availability date... :tears:

You know, I don't think those things go up past an amp in use, you could also try these at 1.2 amps:
512-KSA1220AYS to-126 ecb 120/a160 1.2a 20w 155mhz 35-320hfe $0.39 (R=60-120, O=100/200, Y=160/320)
512-KSC2690AYS to-126 ecb 120/a160 1.2a 20w 155mhz 35-320hfe $0.40

they have been in the back of my mind to try for a year or two now...
watch out: to-220 is BCE, to-126 is ECB

Marty said:

leesonic, Are you going to recap it? Please post a list if you can!

Click to expand...

Here it is! All will become clear if you copy the text below, paste it into Notepad, Wordpad, or any other text editor, but then save it with the extension CSV and then open it in Excel. This listing is for the other SA-7800 I reworked, but still should be good.

Pioneer SA-7800 re-cap,,,,,Nichicon UKL = low leakage
,,,,,Nichicon UES then UVP = bipolar
,,,,,Nichicon UKW = audio grade
Phono Stage,,,,,
Part Number,uF,V,Notes,Values if different,Mouser Part Number
Capacitors,,,,,
1,3.3,25,low leakage,50v,647-UKL1H3R3KDDANA
2,3.3,25,low leakage,=,647-UKL1H3R3KDDANA
23,4.7,25,bipolar,35v,647-UES1V4R7MDM
24,4.7,25,bipolar,35v,647-UES1V4R7MDM
13,47,25,,=,647-UKW1E470MDD
14,47,25,,=,647-UKW1E470MDD
7,47,50,,=,647-UKW1H470MED
8,47,50,,=,647-UKW1H470MED
9,47,50,,=,647-UKW1H470MED
10,47,50,,=,647-UKW1H470MED
11,470,6,,=,647-UKW0J471MED
12,470,6,,=,647-UKW0J471MED

Transistors,,,,,
"1,2",2SA1114,KSA992,,,
"3,4",2SC1117A,KSC1845PBU,,,
"5,6",2SA912,KSA1142 + heatsinks,,,

Power Supply,,,,,
28,2.2,50,,=,647-UKW1H2R2MDD
34,47,50,low leakage,=,647-UKL1H470KPDANA
35,47,50,low leakage,=,647-UKL1H470KPDANA
36,47,63,,=,647-UKW1J470MED
37,47,63,,=,647-UKW1J470MED
33,100,10,,=,647-UKW1A101MDD
29,220,80,,100v,647-UKW2A221MHD
30,220,80,,100v,647-UKW2A221MHD
39,8000,50,KG,8200uF 50v,647-LKG1H822MESCBK
40,8000,50,KG,8200uF 50v,647-LKG1H822MESCBK
41,8000,50,KG,8200uF 50v,647-LKG1H822MESCBK
42,8000,50,KG,8200uF 50v,647-LKG1H822MESCBK

Display,,,,,
3,1,50,,=,647-UKW1H010MDD
4,1,50,,=,647-UKW1H010MDD
13,2.2,50,,=,647-UKW1H2R2MDD
5,2.2,50,,=,647-UKW1H2R2MDD
6,2.2,50,,=,647-UKW1H2R2MDD
9,47,25,,=,647-UKW1E470MDD
10,47,25,,=,647-UKW1E470MDD
12,47,25,,=,647-UKW1E470MDD
11,47,50,,=,647-UKW1H470MED

Tone control,,,,,
39,4.7,25,bipolar,35v,647-UES1V4R7MDM
40,4.7,25,bipolar,35v,647-UES1V4R7MDM
41,22,16,bipolar,=,647-UES1C220MEM
42,22,16,bipolar,=,647-UES1C220MEM
51,22,16,bipolar,=,647-UES1C220MEM
52,22,16,bipolar,=,647-UES1C220MEM

Power amp,,,,,
21,0.22,50,low leakage,=,647-UKL1HR22KDDANA
22,0.22,50,low leakage,=,647-UKL1HR22KDDANA
26,100,50,,=,647-UKW1H101MPD
25,330,6.3,,=,647-UKW0J331MED

Trimmers ,,,,,
100R,,,,,652-3296Y-1-101LF
68K,,,,50k,652-3296Y-1-503LF

I tried keeping the same values as the original, but some capacitors could only be obtained in slightly higher voltages, and the main smoothing caps are slightly higher capacitance. The column marked "values if different" should help here, if the value shown is "=" then the same capacitance and voltage as the originals are used.

I used Nichicon ES for bipolar, KL for low-leakage (essential to use these), and KW "audio grade" for everything else. I know Mark uses PWs for all the others, so you might want to substitute his recommendation for mine. I know on the other SA-7800 I re-worked, the thing sounded really nice, especially on the phono input.

I haven't managed to pull any transistors or other components from this amplifier yet, as I already have a Kenwood, a Sansui and an Optonica amp in pieces at the moment. Once they're done though, I'll get to work on this one.

Lee.

Cool! :tresbon: I'll make a link to this post in the recap list I started!

OK, I got to do some testing on this today, and my findings weren't good. I started with the outputs, desoldering them and testing, the 2SC2525 was shorted from B-C and open from C-E (B=Base, C=Collector, E=Emitter). The 2SA1075 was shorted across all three pins in every direction. I was so astounded, I had to go and find an known good (new) transistor to test, just to make sure I wasn't dreaming. I even pulled it off the heatsink to check there were no short circuits there.

I was wondering, can I use the On-Semi MJL3281 and MJL1302 outputs that I bought for my SA-9800 but didn't use? I think I can get them to fit on the heatsink with a little persuasion, of course I will post pictures here for the benefit of others. I already have all the other transistors needed for the amp board.

Just for sh*ts and giggles, I tested the outputs on the other channel, they are both fine.

Lee.

p.s. The SA-9800 isn't forgotten, it's just buried in my storage unit right now since we moved house.

the drivers (q23, q24, q25, q26)for the outputs have to b e 100mhz plus transistors.
the mjl3281 are 30mhz, so yes you have a fighting chance of good low distortion output.

fitting them is going to be quite a physical challenge.

markthefixer said:

q27 q28 2sa1914 90v 50ma .2w 150mhz 250-1200hfe
512-KSA992FBU (ln)to-92 ecb 120v .05a .5w 100mhz 150-800hfe $0.05 ea

Click to expand...

Just getting back into this. I don't want to tug on Superman's cape here, but Q27 and Q28 are 2SC1914A, not 2SA1914. I've searched other threads, and it looks like they can be replaced with :

512-KSC1845EBU (ln)to-92 ecb 120v .05a .5w 100mhz 150-800hfe $0.05 ea

Lee.

Did some more work on this yesterday, I replaced all the transistors on the bad channel. I don't know if I'm the only one who does this, but sometimes I like to make up a little picture like the one below. It shows me at a glance what should be coming out of what location, and what should be going back in. Any variation, and it's back to the schematics, the service manual, or the search feature here on AK.



Here a all the transistors replaced, plus the three burnt out resistors. My transistors don't look at neat as Pioneers, but that's partly caused because I clamp each lead with heatsink tweezers to reduce any damage caused by soldering. (note to self : must see if I can get any smaller heatsink tweezers).



And here are the new OnSemi outputs installed. It was surprisingly easy, all I had to do was solder some wires on the leads, and attach them onto the circuit board. What you can't see in the picture are the insulators that went under each transistor. I was able to use one of the original screw holes for these, I didn't have to do any drilling or anything.

After installing these, I powered it up, and was amazed to hear the protection relay click before I'd even set the bias or offset.



Lee.

OK, so I got the second pair of outputs installed, as well as the new transistors on the right channel. I had to move the STV-2H diode on this side, you can see it in it's new location below, circled in yellow. I powered it up, and it clicked out of protection as before. I set the bias using the procedure in the service manual, starting with the 53mV measurement (VR5 and VR6). I waited about 5 minutes and adjusted them back to 53mV for a total of four times, then I went on to VR7 and VR8, again adjusting them to
66mV as specifiied, waiting for the values to settle, and then re-adjusting.

The offsets were measured at 4.5mV on the right channel, dropping to 2.4mV, but the left channel seemed kinda high at -27.6mV, rising to -45.7mV. Hooking it up to my test speakers, it seemed to play fine until the power meter was between 1 and 10 watts, at which point the protection relay would click off. I found I could back the volume off slightly, and have the protection relay click off in time with the beat of the music. I tried turning the balance over to the left and then the right channel, it would still do the same thing. I also tried pulling off the source feed from the CD player channels one by one, still the same. It also didn't matter whether the speakers were connected or not.

I took some measurements on the left channel, but I don't think the protection clicking on and off like that is related to the relatively high offset compared to the other channel. Maybe I should try re-capping it?

Here are the results, the service manual figures are shown in brackets :

Q5
B 45.7 (45.7)
C 3.7 (3.8)
E 46.2 (46.2)

Q7
B -45.4 (-45.3)
C -2.6 (-2.5)
E -44.9 (-45.9)

Q9
B -44.9 (-45.9)
C -2.6 (-2.5)
E -45.4 (-46.5)

Q23
B 1.1 (1.2)
C 45.8 (46)
E 0.6 (0.6)

Q25
B -1.2 (-1.2)
C -46 (-46)
E -0.6 (-0.6)

Q1 (output)
B 575mV
C 45.6
E 11mV

Q3 (output)
B -652mV
C -45.6
E 83mV