SX-880 Restoration

Also, from my notes on the 880, R255/256 and R261/262 could be incorrect on the schematic. I've found those to be 390Ω and 120Ω instead of the marked 330Ω and 82Ω. If those were replaced as part of the rebuilding of the board, check those.
 
Also, from my notes on the 880, R255/256 and R261/262 could be incorrect on the schematic. I've found those to be 390Ω and 120Ω instead of the marked 330Ω and 82Ω. If those were replaced as part of the rebuilding of the board, check those.
I haven’t replaced any resistors, but I will check them to be sure.
 
Also, from my notes on the 880, R255/256 and R261/262 could be incorrect on the schematic. I've found those to be 390Ω and 120Ω instead of the marked 330Ω and 82Ω. If those were replaced as part of the rebuilding of the board, check those.
Well Watthour, you taught me a huge lesson with your questions. I cannot thank you enough! I was bound and determined that my problem had to be an active component. However, I had gone in and logged every voltage for the transistors you asked about, and then I went to measure the four resistors, and guess what? The resistors were all open...all four of them. So is the 390 and 120 correct, or should I use 330 and 82? I temporarily solderd in 330 and 150 (I missed your note and saw 150 ohms marked on the circuit board layout in the service manual). It fired right up, and the protection relay pulled in. I obviously want to get the right values in there and then adjust the VRs for offset and bias.

Another question. Given that I’ve replaced almost every transistor in this thing...especially the transistors in the amplifier chain and power supply, would you advise I put all of the old ones back, certain ones of the old ones back, or none of them? As I noted earlier, the 5 legged transistors all showed to be extremely low (5 and 7) on the hfe. Given that the specs say that these should be a low of 250hfe, I feel like those probably needed to be changed anyway. Is that accurate, or should I put the originals back there as well?

Thanks again for your help!
 
If you gain-matched the replacements reasonably closely, and thermally bonded them, they should be acceptable. Maybe a drop of Super-Glue on them in addition to the heat shrink just to be sure. Save the 5-leg pieces for later if their junctions still test as they should.

Use 390Ω and 150Ω replacements if the unit will bias and idle correctly with those installed, and ONLY if the E-B bias at Q9 and Q10 is 0.5 to 0.6V with them installed.

I had the 390/120 in my notes as the correct "upgrade" in the S and S/G types from some time after the first schematics were published. I haven't read/heard of any later updates, nor reasons to alter those from the latest versions.
 
Okay, dc offset to zero, and bias at 30mv. I replaced the dial lights with LEDs. I noticed that when bass notes hit and the volume is up a little, the lights dim. That can’t be right, can it? It sounds fine.
 
First, DC Balance (offset) should be set as close to zero as possible.

Idle current (bias) checks would require the meter negative lead on pins 10 and 16 for L and R channels. Adjust the hot amp for 20-30mV drop on the pair of emitter resistors, equating to an idle current of about 20-30mA in the emitters. If the output transistors are new and improved you may get by with the lower end of the range (20mA) without distortion.
 
Yes, draining the main PS caps can dim the lamps. Smaller amps do that more noticeably, but even larger amps can exhibit that. I have 450s and 550s that do that very noticeably, and even the 850/950 will dim a bit when driving 4Ω at elevated power levels.

Bumping the capacitance of the main filters might alleviate that a bit, but the transformer size (kVA) and main rectifiers must also be up to the task, so usually a 10-15% increase in capacitance is relatively safe for the power supply, but not always without consequences.
 
Yes, draining the main PS caps can dim the lamps. Smaller amps do that more noticeably, but even larger amps can exhibit that. I have 450s and 550s that do that very noticeably, and even the 850/950 will dim a bit when driving 4Ω at elevated power levels.

Bumping the capacitance of the main filters might alleviate that a bit, but the transformer size (kVA) and main rectifiers must also be up to the task, so usually a 10-15% increase in capacitance is relatively safe for the power supply, but not always without consequences.
Okay...well therein most likely lies the problem. Again, I'm a newbie and you can't believe everything you read on the internet. I saw some posts somewhere along the way (may not have been AK) that made it sound as if it would actually be better to raise the capacitance of the main filter caps. Based on that, I've been bumping them up more than the 10-15% increase you mentioned. I have been bumping them up to something the same diameter, which in this case went from the original 15000 to 22000. Based on your last post, I'm going to bet that extra capacitance is more than the transformer can take. I'll need to order a different set and figure out how to deal with the size difference.
 
If the transformer does not get inordinately hot (above 150°F) and the main rectifiers remain below that same range, don't lose any sleep over it. I've done the same kind of thing, but be aware that in some cases, there is such a thing as too much.
 
If the transformer does not get inordinately hot (above 150°F) and the main rectifiers remain below that same range, don't lose any sleep over it. I've done the same kind of thing, but be aware that in some cases, there is such a thing as too much.
Well these LED lamps are dimming a lot more than would be acceptable by any consumer. I'll check the transformer temp. When I put the LEDs in, I added a bridge rectifier to do away with the 60 cycle blink. Any ideas on how to deal with this problem? Is there anything else I might need to look at to make sure everything is okay, and not some issue that's causing them to dim? I would think that if the incandescent bulbs were still in there, they would be dimming significantly as well. Do I maybe need to come up with a different place to pull power for them?

Oh one other thing that's off...the rail voltage should be 41.5, but it's 45 or 46 something like that. Is that a problem? What would cause that?

I just realized that those resistors I replaced are Fusible resistors. I replaced them with 1/4 watt metal film 1% resistors as that's all I had. I've looked at Mouser for fusible resistors in the values I need, but they don't have any. Obviously those original fusible resistors "blew" for a reason, and I'm sure that saved the unit from a much worse fate. Any suggestions on what to do for proper replacements?
 
Last edited:
The original spec for these was RD¼PSF390J and RD¼PSF120J.

These are 120Ω:
https://www.digikey.com/product-detail/en/yageo/FRM50SJR-52-120R/120DVCT-ND/2813214


The "standard" metal film will go 5 seconds at 5/8W before failing. Adding a fiberglass sleeve may cause it to fail faster, but axial lead fuse resistors are getting tougher to find. This sleeve might help keep it cooking:
https://www.mouser.com/ProductDetai...2M8jXxUgU9BDHAZFO10OiV%2b0jlk%2bh8tPjYOi4jA==

However, a 10' long piece might be more than a lifetime supply.

This is a pretty tight spec "standard" film resistor:
https://www.mouser.com/ProductDetai...=sGAEpiMZZMu61qfTUdNhG0IXHLFuiNnd/GPKq32Vt9o=
 
Regarding the voltages, is this a multi-volt unit (Type S/G)? If so, the voltage selector plug might be checked. Utilities seem to be running their line voltages closer to 250V on residential feeders these days, at least in the U.S. that can affect some of these older "117V" rated devices.
 
N
Regarding the voltages, is this a multi-volt unit (Type S/G)? If so, the voltage selector plug might be checked. Utilities seem to be running their line voltages closer to 250V on residential feeders these days, at least in the U.S. that can affect some of these older "117V" rated devices.
No, that was my first thought too, so I checked it and it’s a standard unit set up for 120V.
 
At design conditions in the S-S/G model shows ±44.5V on the main supply. Your measurements seem reasonably close.
 
At design conditions in the S-S/G model shows ±44.5V on the main supply. Your measurements seem reasonably close.
Okay, thanks. I just needed a sanity check on that :)

While I'm waiting on some other parts to come in for it, I pulled out an SX-737 that's in good condition, but needs to be recapped and gone through a bit. It does the dimming bit as well, just not quite as badly, so I see what you mean. Maybe that's normal for the SX-880, and I'm just not used to that sort of behavior. The transformer isn't getting hot, but the two regulators are too hot to hold your finger on. You can see them in the Flir image below with the red arrow pointing towards them. Maybe because of the oversized filter caps? I ordered some 15000uF from Mouser that should still fit the mounting rings, so I'll swap those in on Monday when they arrive.

I did the recommended modifications to the regulator section in the 880, including reversing and slightly raising the value of the cap that was backwards, and putting in a greater value resistor (Purple arrow) to bring the voltage down for the regulator a bit more. As a result, Q30 (Orange Arrow) is running very cool. I wonder if I might get less voltage drop coming off of that 12V supply to feed the LED backlights (so that they won't dim with the music), or if that would be too much extra current draw? Another options is I could tap in to the 55-ish volts ahead of that resistor and add in my own regulated DC supply for the LEDs. Do you think either of those would be a good idea to try?
20180518T172826.jpg
 
Another observation, and I'm hoping someone with experience across several different Pioneer units can weigh in here (Maybe you Watthour?). I have an SX-980 that I went through and replaced all of the caps, noisy transistors, etc. I love the way it sounds. It has crisp highs, but it also has very strong pronounced lows, especially when the Loudness switch is activated. The SX-737 that I'm working on, to me, doesn't have quite as good a sound quality, but it does have nice firm lows along with relatively crisp highs. The SX-880 that I'm working on has nice crisp highs, and the lows are there, but not nearly as pronounced as even the SX-737. I'm wondering if this is normal for the SX-880, or if mine needs some further work in the preamp area possibly? I've read of people replacing the SIP opamps with more modern opamps, like the OPA2604. I really want this thing to sound as good as possible so that my son can get the full late 70's/early 80's quality stereo experience. I really expected the SX-880 to sound a lot like the SX-980, albeit with slightly less power.
 
Generally speaking, the 20/30 series had what some classify as a "warmer" sound with more tonal coloring. Some attribute it to slower components or circuit designs rapidly rolling the highs off into softer tones. The 50 series typically had a little less "color" in the sound and a bit more even reproduction. The 80 series was the more "clinical" or "technical" sound, reproducing whatever was input in it's truest possible form. It's been discussed several hundred times here, usually ad nauseam, and with much opinion and speculation. There are several RC networks in these amps, some intentional, and some probably unintentional, which affect what frequencies are passed through and resonance.

A few things on the 80 series which might affect the apparent lack of lower frequency power are the power supplies, the quality and design of components in the audio path, the reliability of connections in the circuits in things like the high/low filter switches, pot wipers, and other signal path switches, and the speed of the amplification and output devices (and their damping). Non-inductive resistors might sound different that plan wirewounds, faster switching transistors may be less likely to roll the sustain the lower frequencies and roll off the higher frequencies. Rolled foil e-caps may resonate a little more than stacked films. And on, and on.
 
The SX-880 that I'm working on has nice crisp highs, and the lows are there, but not nearly as pronounced as even the SX-737. I'm wondering if this is normal for the SX-880, or if mine needs some further work in the preamp area possibly?

I have a SX-880 that I bought from the original owner who took extremely good care of it. So everything is original AFAIK although there was an IC chip replaced via a service bulletin shortly after he bought it. The low end sounds quite good on mine.
 
An update on the project. I received the 15,000uF 80V capacitors today. I crossed my fingers that maybe the larger filter caps were the root of the problem causing the dial LEDs to dim with the music. Sadly, this was not the problem. After replacing the caps and redoing the DC offset and bias adjustments, it behaves exactly the same way. I'm going to measure to see how much current the tuner board pulls through Q26, and check my LEDs to see if adding them to this supply would stay within the current limits of that regulator. I noted that even when really cranking up the volume, this voltage source only drops around 200-300mV.
 
Generally speaking, the 20/30 series had what some classify as a "warmer" sound with more tonal coloring. Some attribute it to slower components or circuit designs rapidly rolling the highs off into softer tones. The 50 series typically had a little less "color" in the sound and a bit more even reproduction. The 80 series was the more "clinical" or "technical" sound, reproducing whatever was input in it's truest possible form. It's been discussed several hundred times here, usually ad nauseam, and with much opinion and speculation. There are several RC networks in these amps, some intentional, and some probably unintentional, which affect what frequencies are passed through and resonance.

A few things on the 80 series which might affect the apparent lack of lower frequency power are the power supplies, the quality and design of components in the audio path, the reliability of connections in the circuits in things like the high/low filter switches, pot wipers, and other signal path switches, and the speed of the amplification and output devices (and their damping). Non-inductive resistors might sound different that plan wirewounds, faster switching transistors may be less likely to roll the sustain the lower frequencies and roll off the higher frequencies. Rolled foil e-caps may resonate a little more than stacked films. And on, and on.
Watthour,
As far as I can tell, everything is now set up correctly on this SX-880. It sounds crystal clear at lower volume levels, however, when I crank it up to where the bass notes are deflecting the VU meters up into the 10 - 30 watt range, I can definitely hear distortion. It gets exponentially worse past that point. This receiver is supposed to have a clean 60 Watts per channel into 8 ohms. What I'm seeing is not normal behavior is it? Any ideas on what could be causing this? Thanks in advance.
 
Back
Top Bottom