• Check out the Annual AK Secret Santa exchange Program, signup before November 23rd in the AK Charities & Christmas forum!

Finished Recapping my receiver but...

Stevescivic

Active Member
Okay so I finally finished in 1 evening a complete re-cap of my SX-3800's electrolytic caps. For the most part everything went quite well with the exception of one of the output transistor plugs. Not sure if its from heat stress but the plastic shell casing that encases the wires is completed cracked in half therefore there is a potentially hazardous loose connection.

Before I fire anything up I want to know how I can get a replacement plug.


Thanks,
Steve
 
Last edited:
Register to hide this ad
Okay so I finally finished in 1 evening a complete re-cap of my SX-3800's electrolytic caps. For the most part everything went quite well with the exception of one of the output transistor plugs. Not sure if its from heat stress but the plastic shell casing that encases the wires is completed cracked in half therefore there is a potentially hazardous loose connection.

Before I fire anything up I want to know how I can get a replacement plug.

Also I made a minor mistake with respect to my tone board. Where does the headphone jack ribbon cable plug into? I am pretty sure it goes to the FL display board and not to the connector on the tone assembly.


Thanks,
Steve

Yes the one ribbon will go to the display board and the other goes to the speaker terminal board.

As for the broken plug you can find one at mouser or digikey. I'll leave that search for you. In the meantime you can use some heatshrink over the exposed wire and pins. The plugs help in proper orientation so just be sure you have the base, emitter and collector connected properly.
 
Last edited:
Okay so I finally finished in 1 evening a complete re-cap of my SX-3800's electrolytic caps. For the most part everything went quite well with the exception of one of the output transistor plugs. Not sure if its from heat stress but the plastic shell casing that encases the wires is completed cracked in half therefore there is a potentially hazardous loose connection.

Before I fire anything up I want to know how I can get a replacement plug.


Thanks,
Steve

The originals can no longer be found, and if NOS is found, it too has deteriorated over time, with heat accelerating the damage.

I have been too busy to write up a thread to post this fix. It has been a long term problem, and AK'er SolutionRoom found some connectors in a small parts store chain. After a lot of digging I recently came up with the original that all the other companies were copying. And I found it at mouser.com. I needed a bunch to fix some 3800/3900/sx-d5000/sx-d7000 type amplifiers that had self destructed due to bad connectors and overheated amplifier current source transistors. They're working great and this is an elegant solution to the problem. It even fits onto the transistor while the old socket retainer and transistor hold down remains in place!! :D

So you don't know how LUCKY your timing is...

See the green things where your busted connector would be?

it's mouser.com part number 651-1757022
Phoenix Contact Fixed Terminal Blocks 5.08 mm spacing

They are $2.08 each, and worth EVERY PENNY.

You will need one per transistor, cut the wire going to the bad socket's pins RIGHT AT the end of the insulation, strip off 1/8 inch of insulation , being VERY careful NOT to nick the wire, push it all the way into the appropriate square hole and tighten the screw. when all three are on, push the socket onto the transistor, with the transistor leads going into the round holes.

A nicked wire will eventually break, we are NOT using the bare wire that was wrapped around the old socket pins, because the socket pins are DESIGNED to nick the wire, and there is one last turn NOT under tension around that pin to act as a strain relief.

DO NOT OPERATE THE AMP WITH BUSTED SOCKETS, although if ALL are disconnected from the power transistor, then there will be no harm as long as NO LOAD is applied to the amplifier. NOT EVEN HEADPHONES!!!

I have disassembled and scrutinized this new connector, and in my opinion it is superior to the original in many aspects. It is rated to 230v and 15 amps of current, and the round hole is expecting a round pin of approximately the same diameter as the width of the transistor's lead.

attachment.php


attachment.php
 

Attachments

  • d7000_transistor_phoenix_socket.JPG
    d7000_transistor_phoenix_socket.JPG
    93.8 KB · Views: 1,332
Last edited:
Very long - lots of questions for my final stages of my rebuild!

kick arse and lucky indeed!

I was so tired last night by the time I finished recapping my receiver that I failed to mention the following:

1. When I was desoldering the EQ assembly I pulled a bit too hard on a cap that wouldn't quite let go and delaminted about half of one of the cap holes from the board. The copper trace never broke but detached half of itself from the hole where it normally would be. I resoldered a new cap without a hitch BUT I'm not sure if that will cause me any grief down the road.

2. On the FL display board when I was trying desolder a cap I inadvertently thought that the cap wouldn't let go and I snipped what turned out to be the bottom portion of a hard wire jumper. I snipped the bent portion (that locks the jumper wire to the board) of the wire only to realize that I shouldn't have done that. In the end the leg of the jumper was still long enough to properly solder but again that was own botch up that I hope won't cause me issues.

3. One of the small caps on the power supply board was either popped or on the verge of popping. Thank god I replaced it before it fried anything else.

4. On the the power supply board it looks like a section of the board where a single FET looking transistor got quite warm b/c the board is discolored/burnt looking. I looked very closely and all the parts around it seem fine just the board is REALLY dark brown. I reflowed solder on the back of that transistor b/c it looked like heat stress was causing the solder to crack.

5. I have a Canadian market SX-3800 and there is a resistor on mattsd's list of caps that I needed to install on the board BUT when I looked at its location it had holes available but was blanked out (filled with solder from factory) so I'm assuming that its not needed for my unit. Don't know what that would be used for.

6. Do my two large 12, 000 uf caps need replacing? They don't appear to be leaking.

7. I was unsucessful at cleaning my boards of the thin film of smoke residue on it so I'm going to use this product to wash everything down. Looks like a safe bet, is non-conductive and will probably clean everything up nicely. Anyone have experience using this stuff? I want to just blast it clean and walk away from it. No brushing or wiping if possible.

http://www.mgchemicals.com/products/406b.html

8. I re-bent everything on the boards to straighten things out so that parts are not sitting too close together. Hoping that what I just did is frowned upon (I don't think I did any damage to anything).

Thanks!
 
1. probably not

2. use an ohmmeter to verify the connection is intact, and leave it alone

3. which cap - might be a symptom

4. which transistor - again - may be a symptom

5. which resistor - 3.4.5. you have to get more specific if you want REAL answers instead of smoke and mirrors... the service manual NAMES the parts on the schematic and parts lists, and PLACES the parts using the schematic and layout drawings.

6. the large caps have the greatest moisture reserve, fail infrequently and are expensive and difficult to replace, so usually only when they are provably bad.

7. smoke is the hardest thing to get rid of. I used to get a nicotine buzz from contact absorption cleaning taxi dispatch office phones in for refurbishing that started out brown and ended up white when clean. Until I started wearing rubber gloves. All I used was Fantastik...

8. just parts waving in the breeze, not resistors or diodes sitting close to the board surface...
 
Last edited:
Misc questions and answers

Hey mark I meant to say that the power supply board had a cap that was on the list and not my original description of a resistor. I will check when I go home later.


As far as the blown cap there was some juice leaking out of it so I am assuming that it's either popped or on the verge of popping. Things in the unit for the most part seem pretty good. I think photos that I will try to post will help.

Power caps do look ok so I will leave them alone.

I managed to get the Phoenix connectors at my local supply house so tonight I will go ahead and button everything back up :)

Just need some de-oxit for my pots so I can properly clean the dc offset and idle currents.
 
okay so here is the deal:

1. I've come to the conclusion that I suck at reading circuit diagrams. So bad in fact that I am trouble identifying things such as resistors and transistors within a circuit. Is there a trick to reading them?
2. I made sure everything was hooked up properly before I fired up the unit.
3. Receiver turned on and no burning so that's a good sign. Protection relay clicks so that's a good sign.
4. I can't for the life of me get 3 turns on the tuner dial. I tried following the service manual but I can't get the last loop on the dial. I have no more string and I know before I took it off it was just enough.

BUT... major problem:

When powered up the FL display shows the tuner as on all the time regardless of the input selection I make. Its relatively bright, looks like its tuning when the knob is not being turned AND when I select Phono or AUX the numeric display still shows things as if it were still tuning. Big time no no.

What on earth would cause this and should I be looking at the EQ board or the actual FL board?

Thanks
 
Sorry to cut in.

On the FL display problem, recheck the jumper wire that mistakenly cut to see if you really reconnected it. That was issue #2 in the 9:19AM post. On the tuner cord stringing, yes they can be frustrating! Now that you've had a break try again with fresh eyes and mind. Keep in mind there's usually a spring that keeps the cord taught that will need pulled out to get the cord on. The way I usually do that is to get the cord on all the way but the last pulley and just pull on that loop. The other end will need held in place as an "anchor", usually masking tape holding the dial pointer works. You'll get it!

As far as schematic reading I suggest finding a beginning electronics book, hopefully at your local library! You're an adventurous soul and I bet you'll enjoy the knowledge you gain.

Best wishes on getting the Pio sorted out!
:thmbsp::yes:
 
Last edited:
okay so here is the next observation.


I just did DC balance check and:
Measuring
terminal 23 and chassis ground = -23.5 DC volts (yes that is the right reading).
terminal 22 and chassis ground = +41.8 DC volts(yes that is the right reading).

Idle current:
between terminal 28 and 25 = + 4.77 DC volts
between terminal 17 and 20 = -48.4 DC volts

Given the numbers I'm seeing its nowhere near what its supposed to be.

I thought that the protection circuit wouldn't click in if there was something haywire going on.

Either way I'm now trying to figure out what went wrong. I doubled checked all the values of the caps and polarity and nothing was sparking or burning up when I lit it up.



Thanks,
 
I can sort of read circuit diagrams (as in understanding the transistors, resistors and caps and such. Its tracing it from the actual diagram to the actual piece.

I'll double check that jumper again although when I looked at it all seemed well.

With the tuner it will switch between AM and FM display but the tuning numbers make absolutely no sense. The signal meter works fine b/c in AM mode I know that 660 and 770 are strong AM stations here in town and the signal meter does properly reflect that its getting something.

But odd that when in phono or aux mode the display continues to stay lit!
 
Ah well, we tried! Sorry I'm not familiar with that model but it could be that the service manual is incorrect with where to measure offset and bias. Yes the protection circuit should keep the relay disengaged if there's too much offset so I bet it's in the ballpark already. There's a sticky post on adjusting DC offseet/bias that's well worth reading if you haven't already. Markthefixer will likely jump in and help ya out.
 
Just exactly how many wires did you disconnect and or break?

and how did you determine which wire went where when you changed the power transistor connectors .

at the minimum, read and post the 12 ground referenced voltages on the 4 power output transistors.

and if you pulled the motion transfer bars for the function switch, be sure they are all seated properly, one longer than another would "lock on" a function.

IIRC, I redid an entire sx-3800 without breaking or disconnecting almost any wires. that was years ago. Granted a lot of other mechanical stuff was disassembled to get it out of the way. I don't remember what I did with the shielded cables from the tuner to the frequency display board, and I did have to desolder the display board to get it up and away.
 
If you would believe it I didn't break a single wire and only "nipped" the one solid jumper wire and then quickly realized that I should've paid closer attention to that one jumper. Resoldered it so all is good and the connection is rock solid with full continuity.

I did de-solder out the ribbon cable that ties the tone board to the rear speaker terminals but that was a simple desoler / resolder job.

When taking things apart I was able to get each assembly out via dismantling the chassis and turning the boards just a bit to get a soldering iron and solder wick onto things. Amazingly this receiver has plenty of room to work in and the wires are all just long enough to "flip" boards out of the way to get access to the back of them.

As for the power transistor connectors I just followed the sequence in the proper orientation. yellow/red/green brown/orange/blue. I did one power transistor connector replacement and re-grease and re-mount at a time so I didn't screw that up.

I also verified the two little varisistor wires weren't broken (STVH2's IIRC). I've read so many threads where people have accidentally broken the fine wires that come off of those pieces. All was perfect and intact upon re-assembly.

The only thing I can think of that *might* be linked to my issues is the use of the ECQ poly caps in place of the electrolytics. IIRC the .33 and .22 uf caps that mattsd suggested using were of the ECQ part #s and not the typical UPW/UKL designation.

Those ECQ's were used on the power amplifier assembly and another one on the FL display board. Since those poly caps aren't polarized I simply soldered them in.

The motion transfer bars are working perfectly and everything is aligned as it was before I took it apart.

In saying all this I do know that I took my time replacing each cap one cap at a time. During the installation process I carefully examined the values and sizes of each cap (and cross referenced it with mattsd's original list) so I'm confident that I did match everything up carefully. If polarity of the cap was an issue then I would've blown things up by now as the amplifier was powered on continuously for at least 10 minutes.

Even the output transistors were making some heat just idling (same as before I did the re-work).
 
just did DC balance check and:
Measuring
terminal 23 and chassis ground = -23.5 DC volts (yes that is the right reading).
terminal 22 and chassis ground = +41.8 DC volts(yes that is the right reading).
Not looking good..:scratch2:
:lurk:
 
So fair have verified the eq power supply and power amp sections for incorrect cap values and so far nothing. Everything looks right so I am baffled as to what else would be causing this. What drives the fl display to shut off the tuner display when aux or phono is selected? I am assuming the eq board since that is where the switches are.

Thanks
 
If my memory serves me, since you did the recap, I don't remember seeing any power supply voltages listed.

That is step ONE. Both amps misbehaving? Check the power supply voltages.

Use a piece of bare chassis metal (aluminum?) as your ground reference.

I am primarily interested in these on the GWR-122 board:
(copy this and past your numbers after the examples)
pin 20 +47v
pin 18 -47v
pin 22 +32v
pin 23 -32v

pin 24 -21v
pin 25 -14v

pin 15 +12.8v
pin 17 +5.5v when tuner active, 0v when not,
pin 17 is controlled by 12.8v or 0v on pin 16 when the AM or FM is selected by the function switches.
pin 16 12.8v or 0v

pin 13 +21v protection power on/off sense voltage

The awm-227 board (am/fm/phono/aux function selector mezzanine board with phono/equalizer amp) has two diodes, d1(AM) and d2(FM) that connect to pin 10 that goes to the GWR-122 power supply board pin 16. +12.2v am/fm, 0v phono/aux.

See all those ribbon cables crossing around? One end is soldered, the other end is in a connector with a snap down arm/wedge that forces two metal "leaves" together for contact. They could need a bit of cleaning or some may have pulled loose, also check for breaks at the solder end.


The power amplifier has an offset compensation transistor that is connected to the +42 and -42 volt supplies, and has a narrow band of operation, which is very upset when the power supply voltages change.

Your early observation of:
4. On the the power supply board it looks like a section of the board where a single FET looking transistor got quite warm b/c the board is discolored/burnt looking. I looked very closely and all the parts around it seem fine just the board is REALLY dark brown. I reflowed solder on the back of that transistor b/c it looked like heat stress was causing the solder to crack.
had me worried, but I let it pass.
NO LONGER...
Go BACK, FIND that part's NUMBER, and post it. AND the board number that it is on.

NO LONGER will you be allowed to get away with "a single FET looking transistor ", you will NAME the part

DO NOT do or try anything else. Jumping ahead causes damage, and it is now clear that I should have taken closer control of everything from the start.

After we get the power supply trustworthy, we may get the amp's voltages tamed, THEN and only then we shall address other things one at a time. Like WHY the protection kicked in, when it obviously should not have.
 
Last edited:
in 20 +47v = 54.8v
pin 18 -47v = -40.7v
pin 22 +32v = 39.9v
pin 23 -32v = -25.8v

pin 24 -21v = -15.1v
pin 25 -14v = -6.95v

pin 15 +12.8v = 18.75v
pin 17 +5.5v when tuner active, 0v when not, = 6.1v tuner active / 5.3v in aux mode
pin 17 is controlled by 12.8v or 0v on pin 16 when the AM or FM is selected by the function switches.
pin 16 12.8v or 0v = 18.2v tuner active / 5.80v in aux mode

pin 13 +21v protection power on/off sense voltage = 29.7v

The three parts on the control board that made the board look browned/corroded around the metal heatsink edges that contact the board on GWR-122 appears to be Q204/Q208/Q209
 
Last edited:
Back
Top Bottom