Sherwood s8900 - transistor sourcing help

[Moving Ahead]

Is this the first time hooking up speakers? Did you have no hum before?

try this.
1/ reverse polarity on power lead. ie: unplug turn 180 deg and plug back in. any change? replaced x cap?
2/ get another amp and use the in/out feature. sherwood out /other amp in. lets eliminate the preamp and power supply.
3/ check polarity of your new caps. refer to pictures of the old caps that you've taken.
4/ check your trimpots. Sherwood of this age are flaky.
5/ check your ground connections. The points directly soldered to chassis can age and crack.- there are 3 ground points per main amp cards. re flow solder
6/ can you change main amp card from left to right and right to left? to determine the fault is on that card.
thats a start.

 

[special_20]

Moving ahead,

So the hum is only on the bad channel. No hum on the good channel. Good channel sounds strong and volume does what it's supposed to.

Bad channel only hums, no hint of source coming through. Volume doesn't seem to affect the hum.

I did go back over all my connections. I did have the filter cap on the bad channel installed incorrectly. This was my first experience with common ground cases. I had the ground connection attached to the positive terminal by accident. I remedied that but still no joy. Would that be a big enough whoops to kill some small signal transistors?

I also noticed that when I went to try to adjust bias, the potentiometers on the pre-driver did nothing to the bias value. I did have both pots off the board and did test them by connecting a DMM to the connections to confirm I was getting functionality. While they are ugly, they were giving functional results (0-500 kohms of resistance; didn't see any OL readings during sweep). Likewise, I did test the diode on the board and it tested good (at least not shorted and conducting only in 1 direction).

Is it rational or wise to try comparing voltages between the two boards? Seems like it would take a steady hand but that's about all I can think of. I entertained the thought of swapping the boards, however the quantity of leads into and out of boards isn't trivial (~8 connections per board and not the easiest to get to).

 

[Moving Ahead]

With a hum I think it can be any of the following, bad cap, bad ground or a crappy surface contact. There are plenty of these in this sherwood of this vintage but nothing that cant be addressed and fixed. At this stage I try and figure out whats working and whats not. Did you use the pre in main out feature with another amp? This can tell you if the issue is in the early stages (power/pre/tone/potentiometers) or in the later stages(main amp and coupling caps (quite possibly your problem right here)) Its an easy procedure that doesn't need a solder touched. You need to find out which half of the haystack you need to focus on.

1/ reverse polarity on power lead. ie: unplug turn 180 deg and plug back in. any change? replaced x cap?

Did you try this it only takes a second. The X2 caps need to be replaced regardless. (ever thought of a new polarized power cord with ground)

Bad channel only hums, no hint of source coming through. Volume doesn't seem to affect the hum

As suggested use pre out and see if source is coming though the early stage.

I did go back over all my connections. I did have

.....................caps in backwards can cause hum. double check your work. look at solder joints too- a (old) cold solder joint could be your problem.

Is it rational or wise to try comparing voltages between the two boards

you have the schematic so you know what to look for. I would like to think you have competent electrical knowledge before performing such acts thou. Working live is precarious.

the thought of swapping the boards, however the quantity of leads into and out of boards isn't trivial (~8 connections per board and not the easiest to get to

Once you undo the mounting screws and start you'd be surprised how effortless the procedure is.

I would like to peer into a crystal ball and have the answer for you but wheres the fun in that? Take small steps and eliminate what you know is good. You'll get there. Didn't I say you'll develop a love -hate relationship with these old sherwoods? When you get her singing right- bliss.

Suggestion- replace the 4000uF/50v/35mm dia coupling caps. They cost money but they are always the first to go on my list. (you could use 4700uF if you like.)

 

[special_20]

Moving ahead,

Working through your suggestions.

• Flipped the plug around in wall socket: No change.
• Check chassis grounds: reflowed solder grounds as well as capacitor ground connections in power supply and filter cap: no change
• Connect outside preamp: afraid this unit doesn't have jumpers between pre-amp and amp. Not sure how to do this with board connections.
• Replace the 4000 ufd couplig caps: already done. The one on the bad channel had gone open. Replaced with 4700ufd caps
• Swapping boards: getting close to doing...

Mostly so I can keep track of what I've done:

Recap on repairs and parts

• Replaced all caps in the power supply
• Replaced both filter caps
• Tested power supply diodes and everything checked out
• replaced transistors on burned board
  • Small signal transistors are Fairchild 2N4401's all around (Q105,106, 107,Q 108 and Q111)
  • NPN pre driver is 2N3053 (Q112)
  • PNP pre driver is 2N4037 (Q110)
• Replaced misc carbon comp resistors with equivalent replacements
• Diode on pre-driver board tested good
• Tested emittor resistors - they are good
• output transistors are onsemi MJ21194G

Starting in on voltage checks:

• Power supply - Everything is close to spec
  • Voltage regulator transistor: B= .8v C= 5 v E= 1mv
  • [1]+70: 72v
  • [2]+37.5: 39v
  • [3]+23.5: 25v
  • [4]+25: 27v
• Board input voltages - Match power supply outs
  • Good Board
    • [1] +70: 71v
    • [2] 37.5: 39v
  • Bad Board
    • [1] +70: 71v
    • [2] 37.5: 39v
• Top output Transistor (Q 113) - starting to see noticeable differences between channels at Base and Emitter which are fed by pre-driver board
  • Bad: B= 69.5v C= 72.7 v E= 70v Collector voltage is good; Base and Emitter Voltages are 2x what they should be
  • Good B=.36.6 C= 72.7 E 36.6
• Bottom output transistor (Q114) - All my voltages are different
  • Bad: B = 0.1mv C= 70v E= 14.7mv - C is 2x spec seems like emitter is pushing to much voltage too
  • Good: B = .6mv C= 36 E= . 7mv
• Filter Caps -
  • Bad: + = 69.5 - = 25mv - Pos(+) volts are 2x what they are supposed to be
  • Good += 37 v - = 0.1 mv

All indicators would be a bad board. Seems like I'm putting out too much voltage into 213 on the bad board and too little voltage into the base of 214; Likewise too much voltage at the filter cap and Collector of 214.

I'll try to create a close up of the Pre-Driver board schematic tomorrow. Can't seem to find the PDF I had created.

 

[Moving Ahead]

If Q111 is 2N4401 it will burn up!
Q11&12 should be matched to-39 minimum
what have you done with Q09?

Q05&06 are protection they could be kicking in. No need to replace unless proven faulty.

Did you measure gain on the original Q07&08 What did it measure? What do the 2n4401 hfe replacements measure?

with all these tranny changes did you do correct lead orientation?

I hope theres enough in the 2N3053 & 2N4037. Ceo Cbo seems underwhelming. I suggest the NTE128/129- there would be more room to move.

Being a sherwood there will most likely be tantalums. What are C260 and c270? tantalum? Any other tantalums on main amp board should go. And if you replace with electrolytic- ensure correct orientation.

Sorry for all the questions. Maybe something will click.

 

[special_20]

Thanks Moving Ahead!

Transistors

• I tested Q109 and it seemed to behave (not shorted). It's a black plastic looking transistor of unknown material.
• I miss-typed above. Q111 and Q112 are the pre-driver transistors. These are where I put in the 2N3053 and 2N4037. I can pull and see if they are shorted. I triple checked orientation... I could be wrong as these were tricky to install.
• Q105, 106 and 108 all had identical transistors on both boards: manual spec was SPS41/v and SPS817. HFE for all was 60. The 2n4401.... crap HFE is 20! Maybe that's my screw up. I was prioritizing matching Vce, Vcb, Veb values; is Hfe more critical than I thought?.
• Q107 had a slightly different part (S4199) but had almost identical specs to the SPS41 (manual spec says low noise). Also has difference pin out.

Maybe back to the drawing board on transistors. NTE128/129 for pre-drivers and what do you think for Q105/106/108 and Q107? NTE123 was a part that came up repeatedly.

Capacitors

• C260 and C270 were and are definitely electrolytics
• Had to replace 1 ceramic cap that I couldn't get a reading off of; it was also on a burned transistor next to scorch
• Now that you prime me, there is One tantalum on there. Should i test it or just bin it?

I think this was enlightening (mostly that I still have a lot to learn). Thanks moving ahead!
Prescient comment that this was going to take multiple part orders...

 

[special_20]

Found my notes on the signal transistors. Is HFe the most important feature or is matching Vcb/Vce/Veb? Is Hfe being off enough concern to pull and replace?

 

[Moving Ahead]

Just a quick reply as I'm at work, will respond later
however Q11 &12 are both NPN (S409F) and same part.
2N4037 is PNP and can never be used in this location.

 

[special_20]

ahh, wrote while I too was at work and toggling between views. Just did a visual on the boards and against "good board". Q111 and Q112 have npn transistors (2N3053 ); Q110 is pnp (2n4037). Sorry for confusion and thanks.

 

[Moving Ahead]

OK back to the drawing board.

Q107 had a slightly different part (S4199) but had almost identical specs to the SPS41 (manual spec says low noise)

Where did you read this? I'm really coming up zip with data sheets.
SPS4199- 30v/40v/5v/500mA/360mW/250MHz/60minhfe. That is typical spec for that location.
the other trannies I cant get any info.


What I'm thinking is Q11&12 NPN drivers-S409F. The shape is to-66 (but not) That smacks of high Pd and perhaps more importantly a high voltage, which your 2N3053 are not. They will also be low gain. Do you have any TO-220 high volt low gain transistors laying around. maybe pop them in.
If I remember right they had a socket for mounting. Did you remove that and are the boards marked clearly ecb so you mounted correctly?

Make sure you are using a dim bulb tester.
Check your work for correct cap & trans orientation.

Also

Q105, 106 and 108 all had identical transistors on both boards: manual spec was SPS41/v and SPS817. HFE for all was 60. The 2n4401.... crap HFE is 20

Do you have a tester to indicate hfe? (Can you go through and give me some numbers of the original transistors you've pulled out?) I tested some of my 2n4401 and I was getting 300. I would think this is probaby twice the original sps41.

I was prioritizing matching Vce, Vcb, Veb values; is Hfe more critical than I thought?

Yes. Mess with it too much and there goes the sound signature. In addition oscillation can occur.

 

[special_20]

The Q110, Q111 and Q112 were a bit of a mixed bag of info. Good board has sockets (but also non-spec transistors). Bad board had sockets removed and non-spec transistors.

Here is a quick breakdown of what's in manual vs. on the boards (and the NTE equiv)

Item Manual Good Board Bad Board Type Descr Notes NTE Equiv
q210 s522 s522 rca40394 pnp predriver 129
q211 s409f 2n4239 sk3024 npn sames as 212; driver 128
q212 s409f 2n4239 sk3024 npn 128

Now that I review specs, I think I based selection around the worst possible option (that RCA 40394). in reviewing my notes, it's pretty obvious these specs are off

I'll get some NTE128/129's ordered up here this week and see if that does the trick. This level of cross referencing is a fisrt for me. I've done a few but it's always been a parts match (not a specs match).

As for the other transistors (Q105, 106 and 108). Old ones were all labeled TP109 on the parts. Two still measured ok and I got HFE readings of 528 and 498 with "uf" of 630 for both. The 2n4401 I have had an JFE of 289 and uf of 625mv. Still think the 2n4401 is underrated and worth speccing higher?

As far as orientation, the boards are all labeled on back silk screen (B, C, E). Not fun, but I'm fairly confident in my lead placement. I've checked them a few times (as well as against orientation on the good board) since this is not the most straightforward board layout.


Thanks for all the help Moving Ahead!

 

[Moving Ahead]

You're in luck. I was cleaning out the office and came across my old notes and research(before you had a break)
The recommendations are:
outputs s356 > MJ21194
Main amp board.
drivers Q11/12 (NPN) S409F > 2N3439 or 2N3440 these are TO-39( you could add a heat sink to these and give it that original look.
or if not after that look try: 2N5655G (TO-126) or a T0-220 package with low gain.
***I don't think the NTE128 is right for the workload at this location.
**** high voltage/ high Pd / low gain

Q10 (PNP) S522> NTE129
Q09 (NPN) S2104 > NTE 128
***complimentary / low noise

predrivers:
SPS41 > BC546A / 2N3904 (ebc pin-out)
SPS4199 & SPS817 >2n4400 , MPSA05, MPSA06 (ebc pin-out)

Pre-amp/ Tone
S169N > KSC1845F (ecb pin-out) . (if you find sps41N here confirm pin-out.)

As for the other transistors (Q105, 106 and 108). Old ones were all labeled TP109 on the parts. Two still measured ok and I got HFE readings of 528 and 498 with "uf" of 630 for both. The 2n4401 I have had an JFE of 289 and uf of 625mv. Still think the 2n4401 is underrated and worth speccing higher?

In schematic these are SPS41. Whats on the original working board?
Can you get a hfe reading on a SPS41?

 

[special_20]

awesome moving ahead. I've got the NTE pair coming in and just ordered up some a 2n3440. Hopefully they'll be here by weekend.

Regarding the SPS41, I'll have to go back and check if I can get a reading later this evening. I was getting some spotty readings and a few were fried. The TP109 labeled parts are on both boards (might have just been equivalents or OEM parts?).

 

[special_20]

Still a head scratcher. I replaced Q09 and Q10 with NTE parts. I replaced Q11 with a 2n3440. I confirmed pinouts and confirmed their placement on the board. Still no joy; still just a hum out of channel. I think this one is going to need to go to a more experienced tech or someone able to trace a circuit better than I. That and it's spring and there are some other project that I need to attend to. Thanks for all your help guys. I'll post up if I get this bird singing

 

[Binkman]

sounds like a ground off a rail. like a low hum off the ac frequency. dreaded 50hz hum to our euro or aussi scabs! D)

Could be power supply off the channel. Might check all chassis screws and any grounds to pcb's. could be a filter cap off a rail. Flip is transistors amplify.. that is what they do.. including 'hum'.

 

[jpreza]

sounds like a ground off a rail. like a low hum off the ac frequency. dreaded 50hz hum to our euro or aussi scabs! D)

Could be power supply off the channel. Might check all chassis screws and any grounds to pcb's. could be a filter cap off a rail. Flip is transistors amplify.. that is what they do.. including 'hum'.

@special_20 Did you ever get your Sherwood S-8900 fixed? Mine works somewhat - FM radio works ok. When the unit is first turned on, sometimes it has a "warbling" or humming sound. The PHONO input has humming as well.

 

[sansuibutch]

Moving ahead,

Working through your suggestions.

• Flipped the plug around in wall socket: No change.
• Check chassis grounds: reflowed solder grounds as well as capacitor ground connections in power supply and filter cap: no change
• Connect outside preamp: afraid this unit doesn't have jumpers between pre-amp and amp. Not sure how to do this with board connections.
• Replace the 4000 ufd couplig caps: already done. The one on the bad channel had gone open. Replaced with 4700ufd caps
• Swapping boards: getting close to doing...

Mostly so I can keep track of what I've done:

Recap on repairs and parts

• Replaced all caps in the power supply
• Replaced both filter caps
• Tested power supply diodes and everything checked out
• replaced transistors on burned board
  • Small signal transistors are Fairchild 2N4401's all around (Q105,106, 107,Q 108 and Q111)
  • NPN pre driver is 2N3053 (Q112)
  • PNP pre driver is 2N4037 (Q110)
• Replaced misc carbon comp resistors with equivalent replacements
• Diode on pre-driver board tested good
• Tested emittor resistors - they are good
• output transistors are onsemi MJ21194G

Starting in on voltage checks:

• Power supply - Everything is close to spec
  • Voltage regulator transistor: B= .8v C= 5 v E= 1mv
  • [1]+70: 72v
  • [2]+37.5: 39v
  • [3]+23.5: 25v
  • [4]+25: 27v
• Board input voltages - Match power supply outs
  • Good Board
    • [1] +70: 71v
    • [2] 37.5: 39v
  • Bad Board
    • [1] +70: 71v
    • [2] 37.5: 39v
• Top output Transistor (Q 113) - starting to see noticeable differences between channels at Base and Emitter which are fed by pre-driver board
  • Bad: B= 69.5v C= 72.7 v E= 70v Collector voltage is good; Base and Emitter Voltages are 2x what they should be
  • Good B=.36.6 C= 72.7 E 36.6
• Bottom output transistor (Q114) - All my voltages are different
  • Bad: B = 0.1mv C= 70v E= 14.7mv - C is 2x spec seems like emitter is pushing to much voltage too
  • Good: B = .6mv C= 36 E= . 7mv
• Filter Caps -
  • Bad: + = 69.5 - = 25mv - Pos(+) volts are 2x what they are supposed to be
  • Good += 37 v - = 0.1 mv

All indicators would be a bad board. Seems like I'm putting out too much voltage into 213 on the bad board and too little voltage into the base of 214; Likewise too much voltage at the filter cap and Collector of 214.

I'll try to create a close up of the Pre-Driver board schematic tomorrow. Can't seem to find the PDF I had created.

Moving ahead,

Working through your suggestions.

• Flipped the plug around in wall socket: No change.
• Check chassis grounds: reflowed solder grounds as well as capacitor ground connections in power supply and filter cap: no change
• Connect outside preamp: afraid this unit doesn't have jumpers between pre-amp and amp. Not sure how to do this with board connections.
• Replace the 4000 ufd couplig caps: already done. The one on the bad channel had gone open. Replaced with 4700ufd caps
• Swapping boards: getting close to doing...

Mostly so I can keep track of what I've done:

Recap on repairs and parts

• Replaced all caps in the power supply
• Replaced both filter caps
• Tested power supply diodes and everything checked out
• replaced transistors on burned board
  • Small signal transistors are Fairchild 2N4401's all around (Q105,106, 107,Q 108 and Q111)
  • NPN pre driver is 2N3053 (Q112)
  • PNP pre driver is 2N4037 (Q110)
• Replaced misc carbon comp resistors with equivalent replacements
• Diode on pre-driver board tested good
• Tested emittor resistors - they are good
• output transistors are onsemi MJ21194G

Starting in on voltage checks:

• Power supply - Everything is close to spec
  • Voltage regulator transistor: B= .8v C= 5 v E= 1mv
  • [1]+70: 72v
  • [2]+37.5: 39v
  • [3]+23.5: 25v
  • [4]+25: 27v
• Board input voltages - Match power supply outs
  • Good Board
    • [1] +70: 71v
    • [2] 37.5: 39v
  • Bad Board
    • [1] +70: 71v
    • [2] 37.5: 39v
• Top output Transistor (Q 113) - starting to see noticeable differences between channels at Base and Emitter which are fed by pre-driver board
  • Bad: B= 69.5v C= 72.7 v E= 70v Collector voltage is good; Base and Emitter Voltages are 2x what they should be
  • Good B=.36.6 C= 72.7 E 36.6
• Bottom output transistor (Q114) - All my voltages are different
  • Bad: B = 0.1mv C= 70v E= 14.7mv - C is 2x spec seems like emitter is pushing to much voltage too
  • Good: B = .6mv C= 36 E= . 7mv
• Filter Caps -
  • Bad: + = 69.5 - = 25mv - Pos(+) volts are 2x what they are supposed to be
  • Good += 37 v - = 0.1 mv

All indicators would be a bad board. Seems like I'm putting out too much voltage into 213 on the bad board and too little voltage into the base of 214; Likewise too much voltage at the filter cap and Collector of 214.

I'll try to create a close up of the Pre-Driver board schematic tomorrow. Can't seem to find the PDF I had created.

I'm really hoping you figured this out as I'm having the same issues. I know this is an old thread but hey worth a shot. I've went over this thing with a fine tooth comb and cant figure out what is causing the voltage to be very similar to yours in the same spots you mentioned.