I am a little stuck with hopefully the final stages of repairing a Luxman L-430. Some assistance would be excellent if someone feels like casting their eye over this.
Attached is the schematic and the full service manualcan be had from here, with a couple of spurious pages for a different amp.
I have replaced one blown output device in one channel and a couple of cooked resistors in both the power supply and the output stage. When analysed statically with a multimeter the problematic channel appears identical to the good, however the bias current won't adjust to any more than 10mA or so.
I am reluctant to poke around with a probe without more idea of where to look, but supply rails appear fine (61V or so), the only difference I can measure between channels is that the voltage on R122 is much lower on the problem channel and LED801 is barely lit.
Some initial discussion of this amp here.
Attached is the schematic and the full service manualcan be had from here, with a couple of spurious pages for a different amp.
I have replaced one blown output device in one channel and a couple of cooked resistors in both the power supply and the output stage. When analysed statically with a multimeter the problematic channel appears identical to the good, however the bias current won't adjust to any more than 10mA or so.
I am reluctant to poke around with a probe without more idea of where to look, but supply rails appear fine (61V or so), the only difference I can measure between channels is that the voltage on R122 is much lower on the problem channel and LED801 is barely lit.
Some initial discussion of this amp here.
If led801 is barely lit compared with the normal channel, there is less current running through it supplied by R812 (typo?) with an odd value 55k (56k is E12).
Look for bad solderings causing a higher resistance in this track.
Due to the configuration around Q809-810, this lesser current is not immidiately causing a lower bias current.
Bias is controlled with Q911-912 (the collector of Q911should be connected to the collector of Q912, or the circuit runs beserk).
Are leds 803 to 806 normal lit (same as good channel)?
Vbias range should be:
(2.7k + 1.0k + 0.58k) / (1.0k + 0.58k) * 1.2V = 3.2V minimum
(2.7k + 0.58k) / (0.58k) * 1.2V = 6.8V maximum
With lower values, it means there's something wrong here (another bad soldering?).
Lots of typo's in this drawing (two times Q802... 2SK270 is a dual fet)
Look for bad solderings causing a higher resistance in this track.
Due to the configuration around Q809-810, this lesser current is not immidiately causing a lower bias current.
Bias is controlled with Q911-912 (the collector of Q911should be connected to the collector of Q912, or the circuit runs beserk).
Are leds 803 to 806 normal lit (same as good channel)?
Vbias range should be:
(2.7k + 1.0k + 0.58k) / (1.0k + 0.58k) * 1.2V = 3.2V minimum
(2.7k + 0.58k) / (0.58k) * 1.2V = 6.8V maximum
With lower values, it means there's something wrong here (another bad soldering?).
Lots of typo's in this drawing (two times Q802... 2SK270 is a dual fet)
I replaced:
One of the 2SA1265 output devices
R905 on the output PCB
R825
R824
R822
Some others around here show signs of getting warm but test good.
I haven't been able to find any problems with semis in this area.
All relating to the same channel (b)
I have resoldered a fair number of joins, it was surprisingly bad for high end gear. I will check again and look at R812. All other LED's look pretty normal.
There is a manual that's perhaps been corrected on HiFi Engine but it's a terrible copy and illegible in places.
One of the 2SA1265 output devices
R905 on the output PCB
R825
R824
R822
Some others around here show signs of getting warm but test good.
I haven't been able to find any problems with semis in this area.
All relating to the same channel (b)
I have resoldered a fair number of joins, it was surprisingly bad for high end gear. I will check again and look at R812. All other LED's look pretty normal.
There is a manual that's perhaps been corrected on HiFi Engine but it's a terrible copy and illegible in places.
Last edited:
I can't find a 2SA1265, in the schematic Q904-906 are 2SA1146. May be replaced in the past (?).
Replaced resistors indicate a failing output transistor (2SA1265 or 2SA1146) indeed. Check for Q811, Q901, R901 and Q902 too, they're in this 'failpath'.
Is connector CT901 clean and neat?
Correction: is NOT neccessary, will work with this open collector. (Sorry!)
It appears to have had at least a couple of different versions over its life, both output devices are noted in the manual.
Q811 is indeed shorted, thank you. It took removing it from circuit to tell and I was trying to avoid doing that to every single semiconductor.
It is a 2SC1904b. I have the non b variant here, hopefully if I try that I will not produce more smoke, the datasheet doesn't help with this.
R812 is indeed 55k - it is four 220k in parallel.
No, no bias current at all with the C1904 in place.
I have these as when I was first repairing the amp I was looking at replacing all the semiconductors on the output PCB, or at least having spares for them and the non-b version was all I could get. I subbed in these 1904's briefly but had a similar no current problem in that position. I'll have a bit of a google and look for the proper version, or feel free to advise a modern sub.
I have these as when I was first repairing the amp I was looking at replacing all the semiconductors on the output PCB, or at least having spares for them and the non-b version was all I could get. I subbed in these 1904's briefly but had a similar no current problem in that position. I'll have a bit of a google and look for the proper version, or feel free to advise a modern sub.
There's something broken in the 'failpath'; checkcheck the mentioned parts. And around them. Hissing sounds?
G'day mate, but being on the 'wa-swan side' for three month's just ten years ago, I'm not tuned to the oz slang anymore.
Have a look up for the 2SC1904... no just a single npn in TO126 package.
(annoying next typo in the schematic: see Q808, a 'C1904' but's a PNP, should be a 2SA 899!) Becoming a blindfolding repair... if it's hot, turn it off.
Burn this Luxman-witch on the stake (no gender-variation offence intended in any interpretation). What a mess. Typo's, odditys, cuneiform drawings, more typo's, bad solderings, unknown/unregistred variants. Did I mention typo's? The Q809-810 trick is smart, but now I consider it 'borrowed' or better just copied it from a smarter brand or inventor now. Nice casings these however, 'looks' sales. Commercial crap, my opinion.
Four 220k in parallel... no, don't fetch a 56k/1W instead. Way more sophisticated. Not our smack. Dinosaurs. Something is itching between the two sides of my fundamentals. I have to obey the rules of this platform.
Moderators: please do erase abusive dictions in this contribution with... oh just ...
I certainly will, but I am also reasonably certain that 2SC1904 is not a good sub for the 2SC1904b that was in there. Before the amp cooked itself the second time I changed Q911 and Q912 from the 1904b was there to 1904 replacements as a precaution - no bias with these. Changed back to 1904b and bias set up perfectly, with no other changes made.
The fun bit is that the 1904b part appears to be completely unavailable, so I'd love to know what I can sub in.
(And yes, Q911 and Q912 are noted in the both versions of the manual as 2SC1903 - that's not what came in the amp though. Previous repair? Factory change? Who knows.)
That's a road to frustration. I hear you though.
No great argument from me, I don't have a dog in the fight. I was given the amp and when it works it sounds nicer than anything else I own. Does catch fire a bit though.
Alright, some progress.
I went back and checked all my work yesterday. Late last night I apparently managed to put in the 2SC1904 backwards, which is a little embarrassing. With it in the right way around and all nearby devices checked the unit biases up properly and it's making low volume music at the moment.
There is still an intermittent pop and hiss out of that channel, which I haven't heard previously. I'm hoping it's a dodgy solder join, there has been plenty of poking and pulling to make such a thing worse, so I'll go over it again. Any other ideas on that welcome though.
Thanks MarsBravo for your help.
I went back and checked all my work yesterday. Late last night I apparently managed to put in the 2SC1904 backwards, which is a little embarrassing. With it in the right way around and all nearby devices checked the unit biases up properly and it's making low volume music at the moment.
There is still an intermittent pop and hiss out of that channel, which I haven't heard previously. I'm hoping it's a dodgy solder join, there has been plenty of poking and pulling to make such a thing worse, so I'll go over it again. Any other ideas on that welcome though.
Thanks MarsBravo for your help.
Determine if it's a pre or power amp issue. Connect RCA's cables at PRE-OUT and MAIN-IN's but cross, ie, Left Pre OUT to Right MAIN-IN. Set signal processing switch at rear. Does noise stay on same channel, indicates a power amp issue. Preamp issue if noise changes channel.
Tentatively we're good. Noise was unchanged by volume setting so that narrowed it to the power amp section and tapping with rubber sidecutter ends narrowed things down further to an area I had not previously resoldered. A hundred or so joins later and it's silent and currently playing Blood Sugar Sex Magic rather well.
Fingers crossed it stays that way.
Fingers crossed it stays that way.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.