Luxkit A504 repair, long

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello everyone,

First of all, before I launch into this post, let me apologize for the length of this post and also for the fact that my first post in this community is a plea for help. Thanks in advance for those who are patient enough to read the whole thing and to those who can offer some assistance.

OK, here's the story:
A friend of mine shorted the output leads of his Luxkit A504 amp and blew up one of the output stages. (Actually, as he related it, "Flames shot out of the top." Output stage damage was my diagnosis -- not a hard assessment to make in the light of the cracked and blackened transistors.) Never one to turn away from the opportunity to disassemble something, I agreed to attempt a repair.

Let me digress for a moment to provide a little background. I have a degree in electrical engineering. As part of my job over the past ten years or so, I have designed a number of analog circuits (signal conditioning and data acquisition stuff) and several fairly high-power control circuits for inductive loads (motors and solenoid drivers). I'm not completely unfamiliar with the operation of transistors, diodes, bias circuits, etc, but I must be careful to point out that having never worked on an audio amp, my knowledge of them is academic, rather than practical.

OK, back to the story at hand. Faced with the damaged amp and an irresponsibly optimistic promise to attempt to fix it, I ordered a full set of replacement output transistors (3 each Toshiba 2SA1095 and 2SC2565) and books on amp design from Hood, Slone, and Self. I also attempted to locate, unsuccessfully, documentation for the amplifier. From what little I've been able to find on the web, the amp was a kit version of the Luxman A504 amplifier, and was sold only in Japan. I should add that I have no schematics.

As I mentioned, each output stage consists of 3 paralleled BJT pairs. The output transistors are mounted on a fairly complex "heat-pipe" type thermal sink. Circuit boards are single-layer, holes are not plated-through, all components are leaded, and circuit traces have that old-school hand-drawn look.Most solder joints appear to be functional but ugly -- lots and lots of solder (as is typical of boards without plated-through holes), and no apparent attempt by the builder to remove solder flux. The various boards in the amp are connected with point-to-point wiring.

My first attempt at a repair was simply to replace the output transistors and plug the amp in to the wall. Two things amazed me about that first attempt: 1. the 6" flames that leapt out of the vent holes in the case and 2. just how long the burning smell lingered.

Here is my second repair attempt: I ordered a new set of output transistors. I also began to pursue other damage in the VA stage upstream of the output stage. I found and replaced several 100 Ohm power resistors and a couple of small (TO-220 package) transistors that were blackened. With the power transistors uninstalled, I turned on the amp, and verified that the bias potentiometer adjusted the voltage across the bases of the output transistors. I should point out that the bias voltage swing was about 1 volt higher than the range of voltages that I observed in the amp's good channel. I felt comfortable ignoring this problem (perhaps, in hindsight, not a wise decision) because I expected the base currents to force an IR drop across series resistors in the base circuits. I turned the pot to its lowest voltage, unplugged the amp, installed the new output transistors, and turned on the amp. I suppose I can claim some some measure of success this time because there was no flame. Just lots of smoke, and the amp, she still doesn't work.

Guys and girls, it is clear to me now that this is beyond my capability. I am ready to concede that I will not, without a lot more effort and experience, be able to fix this amplifier. Which leads me, finally, to the topic of this post -- is there someone out there, preferrably in the LA or Orange County, CA area, who feels confident they could fix this thing? I am willing to pay a reasonable fee for the repair, as long as you are willing to discuss the problem and the repair with me. I'd like to get this thing fixed. I'd also like to learn a little more about amps. Maybe I can leverage your experience to help me climb this learning curve?

Thanks for your patience and consideration.

David Sprinkle
 
Hi David

I am not much into SS but have repaired a large number of amps and my experience in releasing fire and smoke from the tiny cans may be of some use to you.

The biggest problem of your approach is that you are assuming the amp is fixed before applying power. If you use a variac to gradually increase PS voltage while monitoring current or at least connect low value/low power resistors in series with the PS rails, you'll be able to greatly reduce the bodycount.

A more time consuming but surefire way of fixing is to test every single semicondutor (and probably quite a few resistors).
 
Hi dsprinkle,

You give up too easily! You must understand that the output stages of a power amplifier only do what they are told to do! The input stage , etc.. do the controlling. If they tell it to blow up it will keep blowing up every time you replace those power (read smoke and flames) controlling devices!

You must seek the culprit earlier on .. in the control stages. If the bias is high you must find the reason. This is your mission... because you want to know?


Cheers,
Greg
 
David:

The A504 was never sold as a finished product, AFAIK. It was _only_ sold as a kit, so naturally some were built well and worked well, while others were not.

Also, the design was quite flexible, which in the right hands made for an interesting amp, but in the wrong hands, could be the source of niggling construction errors, or outright mayhem. ;) At the very least the A504 could be configured as class A or class AB, single amp or BTL, with global NFB or without. There may have been other options, for example battery operation, but I don't remember whether this applied to the A5xx series or only the later A9xx series.

Also, these amps were originally sold in the early 1980s, and especially given the possibility of class A operation, they weren't known for running cool. The electrolytics on the pcbs should probably be all replaced for safety, and prior to inserting new electrolytics, the pcbs should also be inspected for copper trace damage due to leaking electrolyte.

Given the range of possible configurations and construction quality, I think you should try to get a schematic at the very least, and preferably the building manual. OTOH, the documentation may be only in Japanese, which could be a problem for the building manual, but won't affect the schematic.

If you contact Luxman, they should be able to provide at least photocopies of the documentation.

http://www.luxman.co.jp/

Some of the Luxman kit amps were sold in North America under the "Fuji" brand (I think), so perhaps this may be another lead to pursue.

The various boards in the amp are connected with point-to-point wiring.

Now that's an expanded definition of P2P wiring that I haven't encountered too frequently. Sure you haven't also worked in marketing? :D

hth, jonathan carr
 
carlmart said:
Probably a stupid question: but after the first attempt of replacing all three output pairs, shouldn't it be more cautious replacing just one pair at the time until it stabilizes?

Isn't the reason for paralleling transistors at the output just taking care of heavier loads?


Carlos

Thank you, Carlos, for pointing this out. Yes, I considered just replacing only one output pair, but remember that I had convinced myself that proper bias required the base current to provide an IR drop across the base resistors. As I mentioned in my original post, this was in hindsight not a wise decision.
 
dsprinkle said:


Thank you, Carlos, for pointing this out. Yes, I considered just replacing only one output pair, but remember that I had convinced myself that proper bias required the base current to provide an IR drop across the base resistors. As I mentioned in my original post, this was in hindsight not a wise decision.

Yes, bias should be moved to the least current position. Or at least stay in class B or AB grounds before going further on.

JCarr's suggestion to get a schematic and/or building plans, as well as looking for 'lytic leaks are very important.

Did you check if the "fire-cracking" effects did not cause any pcb problems or if that might be the reason for the fire?

In any case you should certainly put 5 or 10 watt resistors in series with both supplies to check for problems before total power connection.


Carlos
 
Jonathan, thanks for the pointer about the manufacturer. I had a native Japanese coworker call Luxman today. We were able to order the original build docs for a small fee. Should arrive in a couple of weeks.

Carlos, thanks for the suggestion about installing a power resistor in series with the supply rails. Shoulda thought of that myself.

Finally, any suggestions for variac specs? I have a chance to pick up a 7.5A Variac model 116. Am I correct that this seems a bit too small?

Thanks again.

-dave
 
Administrator
Joined 2004
Paid Member
Hi Dave,
Definitely not! It's kinda large really unless you are working on large class A amps. I mostly use a 2A model, I have 10A units I can drag out when needed (CJ Premier 1 for example).

If it's a good price, pick it up. I find an ammeter and voltmeter permanantely connected is very helpful. You can easily watch for problems.

-Chris
 
dsprinkle:

I'm sorry to say that but I think you are suffering a tremendous lack of inventiveness and creativity, no matter if you have an EE degree (and blowing output devices in groups of six is definitely not good for anybody's budget).

There are thousands of tricks to avoid such catastrophic failures, being the simplest of them to wire a 100W or 200W light bulb in series with the mains cord of the amplifier the first time you plug it after repairing (my favourite). Among other methods, you can insert power resistors or 500mA fuses in the supply rails, or test the amplifier without output devices (and without load!!) in order to check input stage and VAS, etc...

Also, your understanding of audio amplifiers would improve a lot shall you take a pencil and a piece of paper and face the job of drawing a complete schematic of your broken amplifier, so that you can analyze it thoughtfully while resting on your favorite armchair or bed (I learnt this way). Such an analysis may reveal interesting facts, for example: when output devices fail, drivers, pre-drivers and the bias circuit will be subject to overcurrent, overvoltage or reverse bias, and will also tend to fail (so by replacing only the output devices and pluging the amplifier directly to mains you are going to get smoke quite often).

Finally, remember to test all semiconductors against shorted junctions before powering a repaired amplifier, this takes little time and doesn't require to unsolder componentes but will reveal most blown devices since the most common failure mode is a short.
 
Dear Dsprinkle,

Whilst I commiserate with your predicament, unfortunately I am not qualified to assist.

You may, however, hold the answer to my own A504 problem.
I have three of these 504 amps; two in working condition and one not. I have full schematics and build instructions - but in Japanese. Two of my amps are european 220V models (marketed as the Z504) and one, which I imported from Japan earlier this year, has a 100V transformer.
I have emailed Luxman Japan on numerous occasions seeking English build instructions but never received a reply. I approached the local University to see if I could get them translated, but to no avail. If you are successful in obtaining build instructions in a language other than Japanese I would be obliged if you could forward copies. Obviously I would pay for them.

Meanwhile, if you want copies of the schematics, although in Japanese, I will gladly forward these to you. I have the sheet scanned into 4 units.

As I mentioned I have no electronic experience but my brother in England used to build amps for Musical Fidelity so if I can get the build instructions I will ship the broken amp to him from here in Northern Ireland.

Some extra data you may be interested in:-

'Luxman Japan' built factory finished 504s and marketed them as A504. In Europe a company called 'Avance' in Belgium and France sold the Z504 in kit form. In the USA 'Monarchy Engineering' sold the 504 as the A504 in kit form. Here is a link to the full Z/A range of the 1980's.
http://my.reset.jp/~inu/ProductsDataBase/Products/LUX/LUXKIT1982/LUXKIT 1982.htm

I have 3 x 504s, 1 x A506 electronic crossover, 1 x Z502 pre-amp, 1 x Z503 tuner.

Best regards Stephen
 
Hi dsprinkle and Reeler,

I'm also an owner of a Lux Z504 (same amp) as well as the companion Z502 pre-amp; and I too, am looking for a circuit diagram at the very least.

Is there any chance that I could somehow renumerate one (or both -- given they might be different copies) of you for a decient quality copy of a schematic?


In return I can offer you:

- an average quality copy of the original product brochure
- an average quality copy of a review of the amp by Sound Canada Magazine
- a good quality copy of the Owners Manual
- Wiring instructions for turning the amp into either:
  • Bridged Mono class-AB (600w)
  • bridged mono pure class-A (60w)
  • stereo class-AB (200w /ch)
  • stereo pure class-A (60w /ch)
Hopefully one of you will read this thread again and can help me out! :)




Oh, and Eva, you made some very good suggestions for pre-testing an amp. (Definately making a note of those for my own future use!). However, if you saw how many components were in this thing, you might reconsider your other suggestion to 'just' reverse engineer a schematic from it. Seriously, days and days of work for this one! (and a resultant high chance of errors obviously)
 
Reeler said:

I have emailed Luxman Japan on numerous occasions seeking English build instructions but never received a reply. I approached the local University to see if I could get them translated, but to no avail. If you are successful in obtaining build instructions in a language other than Japanese I would be obliged if you could forward copies. Obviously I would pay for them.

Meanwhile, if you want copies of the schematics, although in Japanese, I will gladly forward these to you. I have the sheet scanned into 4 units.



Luxman is probably not interested in customers that bought those units, particularly recently.

You probably did, but have you scanned small portions of the Japanese text and tried Babelfish on it?

Is there any chance you put those schematics here?


Carlos
 
Is there any chance that I could somehow renumerate one (or both -- given they might be different copies) of you for a decient quality copy of a schematic?

Hello, Magestic,

Thank you for your reply to my post. Yes. I have a copy of the schematic, and I'll be glad to send it to you, either a hard copy via snail-mail or as a high-res scan via email. Please let me know which option (maybe both?) you would prefer.

I should point out that the schematic appears to be a copy several generations old and in a smaller format than the original. Which is to say that it can be slightly difficult to read. More significant, though is that some of the little "dots" signifying an electrical node between 2 crossing signals have faded out of existence. In sum, though, this problem is much easier to address than not having any schematic at all.

Oh, and Eva, you made some very good suggestions for pre-testing an amp. (Definately making a note of those for my own future use!). However, if you saw how many components were in this thing, you might reconsider your other suggestion to 'just' reverse engineer a schematic from it. Seriously, days and days of work for this one! (and a resultant high chance of errors obviously)

That observation is absolutely correct. Before I received the schematic from Luxman, I spent several days attempting to reverse-engineer the schematic. I didn't complete the entire amp, and what I did complete had significant errors when compared with the proper schematic. Eva, maybe you wouldn't have had any trouble with the amp, but it turns out I'm a bottom-feeder on the amp-guru food chain and the circuit presented a substantial challenge.

For those of you still keeping score, here's one of the lessons I've learned:

Most of the transistors in this amp are NLA. Typical replacements seem to work OK, for the most part, with one notable exception, which had its pins reversed from the original! (ECB vs. BCE) Took about a week to debug that one. In the future, I'll just do a simple diode check. Tough way to learn a simple lesson.
 
504 schematics

Hi Majestic,

I apologise for the late reply, I am re-locating to sunnier climates and have only recently found the box containing your requirements.
Drop me an email and I can forward you good quality schematics of the 504. I have 4 images scanned in which should allow you to print out sections as required.

As I get settled here in Spain I will go through all my notes to see if there is anything else I can help you with. What is your 'Transformer' number (on the side beginning with P), I believe I may have separate wiring diagrams for the various transformers. Infact, I have a Japanese transformer which I will never use if you are interested. I was never able to track down a 240v UK transformer, but that is of little consequence as I'm in 220v teritorry now

By the way, have you had any problems with the turntable output on your 502? I have a niggling feeling that I am not benefitting from all that could be. (sound not particularly clear and coherent).


Hi Carlman, I've tried the bablefish direction but the scanner can't recognise the hyroglifics (spelling?), even bought a software program to help me - with no luck. I'll have a look around this site to see how to post up the Schematics etc.


More later.

Regards Reeler
 
LuxKit A504 repair, llong

Hi there & Reeler.
I am a owner and collector of Luxkit since the 70's.
I am looking for the Japanese instruction and schematics of the A504 as one of mine broke down a while ago.
I understand Japanese very well and am a English speaker.
I can help translate Reeler's Japanese A504 papers into English for free as long as I can have a scan/photo copy of the instruction book. If Reeler permits, I can post the translated version here.
I tried to contact Reeler via email but I am not allowed as I am a new member. Any one can help me to contact Reeler I will be much appreciated.
I am an acoustical consultant with a degree in electronics engineering, I started audio diy since the 70's.when I was in high school. I will try to help anyone have problem with the Luxkit A501 & A504 when I have the instruction and schematics. (I have complete papers of A501).
Thanks so much
 
Hello everybody,
Like most of the family own, quality concern Japanese business, Luxman went into financial difficulties in the late 80s. (It was sad to see how many Japnese audio electronics manufacturers did the same and disppeared in one night at that period, like San-Ei Musan, Sansui, Tango Transformer etc.) Luxman was acquired by a Korean business, relocated to Korean and eliminated most of the Luxman heritage. Business didn't improved under the Korean management. Sadly Luxkit disappeared forever since then. The former Luxman management and employees jointed together and bought back Luxman in 1999, launched the first product of the Luxman SACD/DVD-A/CD player and sold very well in Japan,app. 300+ were sold at 1,000,000yan, USd10,000 eqivalent. They got a good start and now have complete lines of S.S. and Tube amps and player at very high quality/price in Luxman tradition.
Considering the time went by, change of people and even locations, I don't think the current Luxman or people there can provide any info of the 20+ years old product especially that was not their main stream one.
I have experience of looking for schematics of a preamp kit purchased from Japan in 1977, CR100 from Ken Audio. Although when I visit them physically, the company is still there and the people didn't change much, when I asked them for a copy of the schematics and their answer was "NO! That was too old."
Matt
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.