• Please note that there are a few updates and clarifications made in the Audiokarma Rules, mostly relating to advertising and the addition of the new "Paying it Forward" & "Giving back" forums in the AudioKarma Audio Marketplace section.

Pioneer SX-880 2SK34 FET Replacement

Brivan

Brivan

Well-Known Member
Hi, all

I've got a Pioneer SX-880 that apparently has one bad 2SK34 in the power supply and I'm planning on replacing it with an NTE312 per some threads I read here. I've discovered from the datasheets that the source and gate leads are swapped between the 2SK34 and NTE312. The transistor installs with all the pins in one row, so swapping two of them is going to be awkward. Just wanted to check on this before trying it. BTW I looked for other (better) alternative replacements for the 2SK34 and everything else was backordered. Thank you!
 
Register to hide this ad
check with ak'er eastpoint...

the two in the power supply can be subbed by current regulator diodes (CRD).
the mouser 954-e-202 is a 2mA at 100v CRD for $0.97 each.
that's out of stock
954-E-272 has 79 at 2.7 mA and $0.97 as a reasonable substitute.

the muting circuit does need jfets.
 
Last edited:
Watthour said:
Any luck finding a BF256B? That should work in that application.
Click to expand...
They were a few available on eBay, but they wanted plenty of money for them, so I opted for the NTE312 since it had been used successfully by other AKers. After looking at the datasheet for the BF256B, it appears the pin configuration is the same as the NTE312, with the gate and source pins swapped in comparison to the 2SK34. I'll probably just add some heat shrink to those two legs of the NTE312, then cross them and solder them in place.
 
markthefixer said:
check with ak'er eastpoint...

the two in the power supply can be subbed by current regulator diodes (CRD).
the mouser 954-e-202 is a 2mA at 100v CRD for $0.97 each.
that's out of stock
954-E-272 has 79 at 2.7 mA and $0.97 as a reasonable substitute.

the muting circuit does need jfets.
Click to expand...
I did see the 954-E-202 was out of stock, but I didn't realize there was another alternative in the 954-E-272. Thank you for the info!
 
Well, duh, the source and gate leads are tied together on the PC board, so it doesn't matter they're reversed from the original 2SK34. A little bummed that the protection relay isn't engaging after replacing both 2SK34s in the power supply with NTE312. Prior to this, I had replaced both D28 and Q28, which tested bad. I'm showing correct voltages at both replaced FETs and everywhere else I've checked. I've also been able to set both DC offset and idle current to factory specs. Now I'm on to the PA3004 and I'd like to verify some voltages. Here's what I have:

PA3004
Pin 1: 13.8v
Pin 2: N/A
Pin 3: 15v
Pin 4: -12mV
Pin 5: -0.8mV
Pin 6: -7.2v
Pin 7: 643mV
Pin 8: 6.2v

Obviously pin 3 is high since the PA3004 doesn't think the protection relay should be engaged. Everything else looks about normal, but I'm not sure about pin 7. Neither the schematic nor the board diagram gives a clue about what to expect, here. My schematic gets crappy when trying to follow the connections away from pin 7. Can someone enlighten me?
 
After doing some more thread searching, it appears I should have seen around 7vac on pin 7 of the PA3004. I traced this back to a faulty 1-amp circuit breaker I had clipped to the terminals for FU2 for the -52v supply. I had blown several 1-amp fuses while diagnosing the problem, so I subbed in what I thought was a good circuit breaker - not so much, as it turns out. I was excited to find out all I needed to do was replace the bad breaker with a 1-amp fuse one more time but, alas, the joy was not to last. I'm finding myself back at square one with this since the DBT lit up brightly after the fuse was replaced. I need to isolate the first stages of the amp section from the power supply and/or remove the PA3004 to see if the problem goes away.
 
Okay, after removing the F lead wire to the amplifier, removing the PA3004 and Q27, I still had a brightly-glowing DBT so I started looking into the power supply and found D26 shorted. Replaced it with a 1N4004, then reinstalled and reconnected everything I had removed/disconnected previously. DBT dims down right away after power-up so we solved that problem. Still no protection relay click. All pins on the PA3004 measure the same as before, with a notable exception on pin 8 - it seems to be following a charge/discharge cycle of C417 - voltage will build to around 7 volts, then drop to 1 or 2 volts and start building back up to 7 volts, again. Now this may be due to the flaky main power in my building that causes anything higher-powered than an SX-780 to cycle the protection relay on/off (there is extensive thread on AK about this). Given that, I've always had the relay engage briefly, but not with this one.

When measuring 7vac on pin 7 of the PA3004, what should I have the negative lead of my meter connect to - ground? When grounded, I measure about 650mV AC. Interestingly, when I remove the negative lead from ground and let it float, I measure 7.4v AC.
 
Another note: the input on pin 7 of the PA3004 appears to come from a voltage divider using two 1m resistors between the secondary windings of the transformer via pin 72 and pin 73 of the main board. I'm measuring 40vac from pin 73 to ground, 0vac from pin 73 to ground and 40vac between pins 72 and 73. Testing across each of those resistors in-circuit, I'm seeing about 2500K-ohms.
 
Those readings should be ok for pin 7... 0.6VAC with respect to ground is about what it should be!

I remember another PA3004 thread on AK where they were seeing pin 8 charge up and then discharge without ever getting to the point of engaging the relay... It's probably worth digging up that thread.
 
alexhastings said:
Those readings should be ok for pin 7... 0.6VAC with respect to ground is about what it should be!

I remember another PA3004 thread on AK where they were seeing pin 8 charge up and then discharge without ever getting to the point of engaging the relay... It's probably worth digging up that thread.
Click to expand...
Thanks for the info! I don't recall if that was my own thread or not - could have been. In the meantime I've pulled PA3004 and subbed in a SIP socket for less abuse on that IC in the future. Voltage readings with the PA3304 pulled are:
Pin 1: 13.7v
Pin 2: N/A
Pin 3: 50v
Pin 4: 0v
Pin 5: 0v
Pin 6: -13v
Pin 7: 13.3vac
Pin 8: 260mV

I've replaced the PA3004 with a new spare and have the same results and voltage readings as I had before. About 0.6vac at pin 7 and a constant charge/discharge on pin 8. I did pull, test and replace C417, even though it tested good. Same problem. I did some searching for the thread describing the charge/discharge issue on pin 8 but, so far, haven't found it.
 
Q37 is probably blown, pin 3 of 50 volts is so far out of the pale it's not funny.
anything over 0.6v on pin 3 and Q37 is turned on.

pull q37 and see what happens with the pa3004 voltages.

Other posts of mine show an input impedance of pin 7 being about 10,000 ohms, which should have about 0.7 to 0.9v ac at pin 7.
I suspect your DBT is causing the charge / discharge because there isn't enough AC at pin 7.
If you see more than 2 or 3 volts of ac there, the chip is probably damaged.

first be sure all regulated voltages are correct.
 
markthefixer said:
Q37 is probably blown, pin 3 of 50 volts is so far out of the pale it's not funny.
anything over 0.6v on pin 3 and Q37 is turned on.

pull q37 and see what happens with the pa3004 voltages.

Other posts of mine show an input impedance of pin 7 being about 10,000 ohms, which should have about 0.7 to 0.9v ac at pin 7.
I suspect your DBT is causing the charge / discharge because there isn't enough AC at pin 7.
If you see more than 2 or 3 volts of ac there, the chip is probably damaged.

first be sure all regulated voltages are correct.
Click to expand...
Thanks, Mark. Just to verify - the 50v at pin 3 of the PA3004's position was with the the PA3004 pulled (or maybe you knew that). All regulated voltages were good last time I checked, but I'll do a re-check. I did try mains power to check this, but still got the cycling of the voltage at pin 8 (PA3004 installed). What I've learned from my goofy power is that relay cycling (clicking on and off) in receivers with the PA3004 chip tend to not exhibit the problem when connected via a single 60-watt bulb on the DBT. As soon as I start adding more bulbs in parallel (see my sig pic), I get relay cycling. Now, I don't know if the voltage cycling on pin 8 of the PA3004 in this particular receiver is related to the relay cycling problem I've witnessed on other receivers, but I'd be willing to bet it is. I'm wondering if I install a smaller-value cap connected to pin 8, if it would charge up and latch the relay before the PA3004 decides the power sucks and shuts things down, again.
 
Last thing I did yesterday was bypass the protection relay (since I had successfully set offset and idle current) and connect up my test speakers to the receiver. On power-up, I immediately got a glowing DBT, so I shut it down quickly. Hmmm. Well, after some overnight pondering and reviewing my previous posts, it looks pretty obvious that I adjusted DC offset and idle current while I still had the faulty circuit breaker connected in place of the fuse for the -52v supply. Pretty sure those adjustments are going to be way off now that I've repaired the power supply and have a good fuse back in place. Hopefully I didn't roast an output transistor or one of my bench speakers in the process. I'll check those, as well. Will report all my findings back here on Monday.
 
Okay, things got a little stranger this morning. For the first time, since I was fairly certain the power supply problem was solved, I plugged the receiver into mains power and gave it a try. The protection relay engaged, but it engaged instantly, briefly pegging the output meters (as expected). Back on DBT power (two 60-watt bulbs in parallel) the relay doesn't engage. I'm a little confused as to what's going on, so I've taken some measurements:
Q27
B: -42.3v
C: -54.7v
E: -41.7v

Q30
B: 43.4v
C: 53.3v
E: 42.8v

Q26
B: 14.4v
C: 35.1v - 35.6v (oddly, this seems to pulse at about the same rate as pin 8 of the PA3004)
E: 13.9v

Q37
B: 15.3v
C: 20.67v - 20.77v (again with the pulsing)
E: 0v

PA3004
1: 13.9v
2: N/A
3: 15.3v
4: -12.7mV
5: 0v
6: -6.87v - 6.97v (pulsing)
7: 0.621vac - 0.634vac (pulsing)
8: 1.6v - 6.2v (pulsing)

Protection Relay
20.8v & 47.1v = 26.3v across relay coil (on DBT power)
21.1v & 50.3v = 29.2v across relay (on mains power)
I've sometimes had the relay engage while on DBT power, sometimes not. I'm guessing somewhere in the 26v - 29v is the tipping point for engaging the relay.
 
Between Mark's suggestion and the odd voltage readings I was getting on Q37, I pulled and tested it and found it was open from B to E. D35 also didn't quite measure up resistance-wise to a replacement 1N4148, so it was replaced. I replaced Q37 with a KSC2383 and am happy to report that I now have a solid relay click a few seconds after power-up. Pin 8 of the PA3004 is a solid 7.2v. That bad circuit breaker I was subbing in for a 1-amp fuse really threw me on this one - lesson learned. As always, thanks to those who lent their advice!
 
I probably posted my closing remarks a little too soon. I now have very weak audio on the right channel. In fact, the only time I get audio in the right channel at all is when the loudness switch is turned on. It seems to be lacking in lower frequencies, so I'll pay close attention to the tone control portion of the preamp when diagnosing this. I'll post a separate thread on this if I run into a dead end.
 
Bad solder connection on the volume control pot. There are other solder connections (not mine!) that look suspect, as well. I think this thing was made on a Friday!
 
Brivan said:
I did see the 954-E-202 was out of stock, but I didn't realize there was another alternative in the 954-E-272. Thank you for the info!
Click to expand...

The Semitec E-202 CLDs were supposed to arrive at Mouser today. I'm waiting on a couple myself to replace a couple of 2SK373s in an A-70 Integrated Amp
 
Well, I'm back at this, again, with a different SX-880. I've got +58v & -58v in to each of the 2SK34 JFETs, but only about +3v & -3v out, respectively. I replaced both JFETs with NTE312 JFETs and E202 diodes and am getting the same result. I've checked for proper grounding, etc. but am baffled as to why my output voltages are so low. Obviously, the receiver is dead. Can anyone shed some light on this? Thanks!
 
You must log in or register to reply here.
Back
Top Bottom