Lexicon repairs... who can do it?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

dustbro

Well-known member
Joined
Feb 1, 2006
Messages
665
Location
New York - USA
I have a 480L that needs some love. I sent it to Harmond/Lexicon for repairs, but they were unable to fix it. They said that the "backplane" was broken and they no longer carry that part. their recommendation was to contact Jim Fabiano, but he's still out of business.
Anyone out there have a suggestion?? or maybe has this part laying around in their shop?
thanks
Dan
 
I do know someone who can get it fixed for good.
seriously he is the man!
He lives here in London though...
If you wish to get in touch with him drop me a PM and I'll give you his details.


Best,
Mattia.
 
I could look at it and probably get it working for you. If the backplane has broken traces, it can be repaired but it takes a lot of work and time. Normally these are repaired by attaching small wires neatly to the back. You can email me at mics (@ at ) 10000cows (dot) com. Jim would be a better choice since he works on them all the time, but if he's not working on them or so far behind... I have a backlog of only two or three weeks of repairs, but for a 480L I can probably move that up to the front of the line :wink: Sometimes it takes a while to get parts so the first look is to figure out what is needed, then get the parts, then put them in. Is the backplane totally 'bashed up' or is it just likely one or two bad traces? Symptoms? Do you have the extra sampler module installed?

I've actually done up new custom PC boards for reverbs like this - actually, I did up a replacement filter PC board to replace the Murata filters that seem to go bad on PCM60's, 70's, 480's, and some of the Roland units. I also redid some of the PC boards in my console. The problem with this approach is that redoing the backplane is likely to cost around $1000 or so for the boards, and probably ten or so hours of redesign time. It is, however, possible. If it can be 'patched' that is way better.

For that matter, if someone was desperate for an ARU, CMU, or MMU chip, I could make one up if required. The first one would be expensive but the second one would be about $30.

-Dale
 
Send it to Lexicon. They have a flate rate of $450.00 and
will repair anything the 480L needs short of replacing the LARK.
They also will update it to the current firmware release.

(801) 568-7567

hope this helps.
RonL
 
[quote author="rlaury"]

(801) 568-7567

[/quote]

Just realized they are about 20 minutes from my house :shock: :shock: :shock:
Not that it matters, of course... I even don't have one :wink:
Just nice to know in such a small place as SLC to be surrounded with Wilson Audio, Kimber Kable, Aco Pacific, and now...

Never mind.
 
[quote author="dustbro"]I sent it to Harmond/Lexicon for repairs, but they were unable to fix it. They said that the "backplane" was broken and they no longer carry that part.[/quote]

[quote author="rlaury"]Send it to Lexicon.[/quote]
 
As in most Lexicons prior to the Lexichip custom IC, the 480 (and PCM60/70/300) use a PCM53 for that function. Of course, that's a DAC, but you can use it as an ADC if you add some extra chips. Those extra chips include a precision comparator, and the 'CMU' chip - it's a successive approximation controller. In a 60/70, a single PCM53 is used for three functions - A to D, plus left D-A and right D-A. In a 480, four are used. One is used for A to D on each input, then one is used for each output (right AUX and MAIN; left AUX and MAIN). I am somewhat surprised that Lexicon didn't use only two PCM53's (they were expensive at the time) but that's how they did it.
 
so no seperate AD... quantec had it though :)

well if the plastic pcm53 were expensive then the ceramic pcm75 was called outrageous :twisted:
 
At the price that a 480 went for, you'd better be buying something good.

The computing power, or more appropriately, memory transfer rate, in the 480 is impressive. I was calculating out memory cycle timings, and the four processing cores in a 480 - with essentially the same processor as a PCM60, but four of them - can out-read the SDRAM used on PC's these days, at least in a random read/write cycle as used in a 'verb.
 
well quantec had this remark too, only recent technologies can match what they put together in 80's from, I guess, bitslice.
 
You can get very good performance from static RAM these days (sub-10ns read/write cycle time) but the current crop of big memory (SDRAM) only shines when doing sequential access. A DRAM page size of 4k or 8k would be much better as that would cover the sizes of most diffusors and delays in a reverb, but that is usually larger than a typical page size which is more like 512 bytes to 2k.

For random access (which is what you need for a Lexicon-style 'verb), the memory cycle time is actually slower than - or no faster than - a DRAM from the late 1980's and early 1990's. The 480 runs around a 160ns memory cycle time, but with four parallel processors, it can do the equivalent of around 40 ns per memory cycle if it was given a single memory bus and single processor. An SDRAM IC takes seven clock cycles to do a random-location read-recharge cycle, or 52 ns. In this case, reverb is memory speed limited and not really processor limited. On any DSP or even a fast PC or Mac, you can get lots of instruction cycles in here but the classical program/data + cache system is not efficient for this type of algorithm. That is why up until recently, Lexicon's claim that harware was more efficient and cost-effective than even general-purpose DSP's was (and might still be) absolutely correct.

I don't think it's a case where the 1980's technology was better. I think it's actually more of a case where in the 1980's, hardware and software designers really understood bottlenecks and performance limitations, and how to get around them. At that time, parallel processing was how you pretty much had to do it. At the time, memory and processing were about synchronized in terms of speed. Now, memory is not significantly faster (except for the burst transfer modes which don't count in this application), but processing is between five and twenty times faster.

My feeling today is that software programmers have an attitude of making something look nice and shiny, make it look organized with this structured programming stuff, and let the hardware guys catch up. Also, I don't think that many software programmers really understand deep-down how memory timing and caching works. This approach might work on a PC, but for a hard real-time system (such as a real-time reverberator, or my day job, engine management system design), programmers must understand that the correct answer that's too late is a wrong answer.

Of course, this doesn't help with who to send a 480 to but it's interesting to me anyways.

-Dale
 
http://davidsonelectronics.com

is located in Plainview, out on Lonk Eyeland. They are not "authorized" but they fix anything. They have done all of my Lex' 41, 42, 61, 71, and 224 weird problems.
I recommend them highly; much, MUCH more than any of the studio equipment service shops in the city. They were planning on some sort of weekly NYC drop, and you can send or messenger it if they have not.
They rock the vintage synth world as well.

"A broken backplane"? It takes a while to extract it, but once out it can be cleaned, re-soldered, de-fluxed, and continuity tested. Unless the PCB is cracked, and even then with a thorough re-tracking, it should fix-up fine. Get a new fan while you are at it- the whole thing will

Did not know that Jim is "out of business", "still"?
Mike
 
[quote author="dale116dot7"]
I've actually done up new custom PC boards for reverbs like this - actually, I did up a replacement filter PC board to replace the Murata filters that seem to go bad on PCM60's, 70's, 480's, and some of the Roland units.

For that matter, if someone was desperate for an ARU, CMU, or MMU chip, I could make one up if required. The first one would be expensive but the second one would be about $30.

-Dale[/quote]

Dale,
Very interesting stuff.
I have a PCM 60 that I've been wanting to get working for YEARS. :sad:
You can faintly here the reverbs working but they're very low in volume & masked by noise. I'd have to get it out again to give a better description if needed. It does seem to pass audio OK in BYPASS!

If you think it could be one of the chips you said you made I'd like to get one.

I had a good tech friend of mine check it out a few years back & he said it sounded like one of the "Proprietary chips" you can't get any more went bad... he has since passed away.

Please let me know if you can help me & if you need any more detailed info or specific tests done to help evaluate it.

Thanks,
Kevin
 
If you have access to a 'scope and you can look at a few waveforms, you can tell pretty quickly. Keep in mind that all of the proprietary chips are in the digital section, and they are usually relatively well protected from outside stuff.

My PCM60 needed to have its DRAM replaced. It worked fine after that. Another common problem with PCM60's (and so many other machines like it) are broken ceramic filters. You can troubleshoot this with a 'scope. Look for signal at TP6. That is the filtered signal before going into the D-A section. Look for signal at TP12 and TP13. Those are the signals coming out of the D-A. Look for signal at TP15 and TP16. If the signal disappears between there, you've got a toasted ceramic filter. TP16 should be 1/2 the amplitude of TP15.

Before doing any of those checks, when in normal (non-bypassed) mode do you get dry signal? If not, you gotta get that working first. The dry signal does not go through any of the digital reverb section, unlike on many newer 'verbs. The PCM70 I thought was the last to run dry through a separate analogue path.

Also, check for +/- voltage across any of the op-amps (5532's). Note that the bypass switch is actually a relay-based bypass.

-Dale
 
[quote author="dale116dot7"]

I don't think it's a case where the 1980's technology was better. I think it's actually more of a case where in the 1980's, hardware and software designers really understood bottlenecks and performance limitations, and how to get around them. At that time, parallel processing was how you pretty much had to do it. At the time, memory and processing were about synchronized in terms of speed. Now, memory is not significantly faster (except for the burst transfer modes which don't count in this application), but processing is between five and twenty times faster.

My feeling today is that software programmers have an attitude of making something look nice and shiny, make it look organized with this structured programming stuff, and let the hardware guys catch up. Also, I don't think that many software programmers really understand deep-down how memory timing and caching works. This approach might work on a PC, but for a hard real-time system (such as a real-time reverberator, or my day job, engine management system design), programmers must understand that the correct answer that's too late is a wrong answer.

-Dale[/quote]

This is so often true and reminds me of the wonderful piece of years back, "Real Programmers Don't Write Pascal", patterned after the best-seller Real Men Don't Eat Quiche. Available I see (ahh search engines be blessed!) here: http://www.desc.okayama-u.ac.jp/~jyam/Personal/PC/RPNWP.html

Some choice excerpts:

"My first task in the Real World was to read and understand a 200,000-line FORTRAN program, then speed it up by a factor of two. Any Real Programmer will tell you that all the Structured Programming in the world won't help you solve a problem like that - it takes actual talent. Some quick observations on Real Programmers and Structured Programming:

* Real Programmers aren't afraid to use GOTO's.

* Real Programmers can write five-page long DO loops without getting confused.

* Real Programmers like arithmetic IF statements - they make the code more interesting.

* Real Programmers write self-modifying code, especially if they can save 20 nanoseconds in the middle of a tight loop.

* Real Programmers don't need comments - the code is obvious.

...

OS/370 is a truly remarkable operating system. It's possible to destroy days of work with a single misplaced space (actually this is also true of UNIX), so alertness in the programming staff is encouraged.

...

No, the Real Programmer wants a ``you asked for it, you got it'' text editor - complicated, cryptic, powerful, unforgiving, and dangerous. TECO, to be precise.

...

It has been observed that a TECO command sequence more closely resembles transmission-line noise than readable text [4]. One of the more entertaining games to play with TECO is to type your name in as a command line and try to guess what it does.

...

Some of the most awesome Real Programmers of all work at the Jet Propulsion Laboratory in California. Many of them know the entire operating system of the Pioneer and Voyager spacecraft by heart. With a combination of large ground-based FORTRAN programs and small spacecraft-based assembly language programs, they were able to do incredible feats of navigation and improvisation - hitting ten kilometer wide windows at Saturn after six years in space, repairing or bypassing damaged sensor platforms, radios, batteries. Allegedly, one Real Programmer managed to tuck a pattern-matching program into a few hundred bytes of unused memory in a Voyager spacecraft that searched for, located, and photographed a new moon of Jupiter.

"

I love that last anecdote about Voyager :green:

Even Mark G., my friend and collaborator from time to time, now mostly writes compiled (C) code for low-bandwidth PIC uC applications. It's interesting how quickly that approach runs out of gas even for what seems at first glance to be very low-octane tasks. But yes it's easier to inspect and modify and maintain yada yada.

Oh well. It will be interesting to see what happens to Lexicon now that Harman (whose misspellings are proliferating of late and probably contributing to its deterioration) seem to be on the rocks. :sad:
 

Latest posts

Back
Top