Help I need to locate a part


I have owned Spectral amps DMA 50, 80, and still own a 90.
(I find the Audio Magic digitals sound like the 80)
Spectral, when last I spoke with them would not touch something if anyone else had been in it. This was in the days when they told folks that non-MIT cables could damage the amps.
I had a chip in the 50 blow, and an ex teck named Crosby helped me locate it as he and spectral had bought out the market on that chip.
Well, my 90 is literally giving me static and my current teck needs to
replace the I.C. and we really need to locate, either:
*the chip,
*any contact info for Mr. Crosby (who might know the chip intimately) or
*someone who knows cross referencing and may know a better alternative to
a very hard to find chip.
I'm not technical when it comes to circut specs.
But this is the info on the chip:

number NPD5566.
The other number which we believe is a date code only is M9384.

So if anyone can please help me, email me any leads, I will be very
grateful.
This amp already has upgraded RCAs, binding posts, I.E.C. input and output
wire, and is sounding closer to the next version as well as nearly like
the next size up.
Thank You all in advance for any assistance you may provide.
Spiro
spiro
The NPD5566 is National Semiconductor's version of the N-ch JFET 2N5566
You might find Siliconix data sheet here http://www.vishay.com/docs/70254/70254.pdf
(Siliconix belongs to Vishay now)

Hi Folks,
I will take strongly the advice to keep away from fet-rolling :(
My experience with Spectral was under the following conditions:
"Hi, I have a DMA 50, I love it. But my teck tells me that it needs a
motorola part that can no longer be found. I've alot of upgrades in it,
like top Vampire (was the brand that they used) RCAs and binding posts
(that
they previously told me *they wouldn't do as an upgrade). Can you help me
with this?"
"No, we will not acept back any product that anyone else has been inside."
Now, i didn't demand to talk to Rick, and didn't like the attitude that the woman gave me, though she might really have been an upper crust *lady. I contacted "Crosby" and he said that he bought up all the rest of that Motorola chip that Spectral hadn't and he sold me 2, for 50 a piece.
My experience with Pass Labs (6 months later) was, "oh yeah, you can upgrade lots of parts in the Aleph 3, diagrams are on the net, or we can mail them to you... or I can look up the vaues if you want."
so after a decade with Spectral, I spent the next decade giving money to
Pass Labs in my upgrade fever. If the folks at Spectral are no longer commanding MIT cables (as I've heard) and will do what even the lofty Krell will, (fix and charge you, and do it again if you like) then my impressions are out dated. For any outdated info I convey, I apologize. But it turned me off big time.
Thanks for all of your comments folks.
BTW: Husky, told me that New Jersey Semi Conductor has them. The part in
the
amp *has the National Semi conductor logo on it, and calling NJSC
found 40 in stock.
Thank you all very much.
Mr.s K & S- You'll note I prefaced my note with a, "supposedly". My brief glance at a substitution handbook at work was obviously TOO brief. Please accept my profound, "OOPS!"
Rodman, your link is to an OPA627, and that's an opamp, NOT a dual JFET. They're, uh . . . different. Using it as a sub would be like trying to substitute a brake caliper for a carburetor.

These things were kinda in vogue in the early 1980s, and were used in some Urei compressor/limiters. Last time I worked on one, the 2N5566 was out of production and a pain in the butt to find -- and that was well over a decade ago. But that was also in the Internet's relative infancy, so it may actually be easier to find these days.

Subbing small-signal JFETS in a power amplifier is NOT a journeyman-technician's task . . . best left to somebody very experienced - "JFET-rolling" is a very bad idea! Small differences in junction capacitances from one type to another could at best make the amp not sound like it should, or at worst turn it into a power-oscillator, destroying both the amp and any speaker attached.

I'm still blown away that Spectral won't give factory service for your amp - that's absolutely the best option.
Thanks folks.
After I posted, I tried google and found the mention of the n-ch jfet as well.
But may I ask, regarding the B.Browns, what improvement would they make in an amplifier?
Are they a straight replacement in that app to the NPD?
Also, John, C. I saw hat and couldn't get into the thread on the given link.
So thank you all.
Anyone have info on Mr. Crosby, the "Crosby Quad's" technician?
Thank you.
http://www.diyaudio.com/forums/showthread.php?postid=155479

This fellow has 230 of them to sell.

John C.
(http://search.digikey.com/scripts/DkSearch/dksus.dll?lang=en&site=US&keywords=opa627) Supposedly- these are equivalents to the 2N5566(NPD5566). I'd prefer the Burr-Brown, the data-sheet of which can be downloaded here:(http://www.datasheetcatalog.com/datasheets_pdf/O/P/A/6/OPA627.shtml)
Spectral, when last I spoke with them would not touch something if anyone else had been in it.
Wow, that's really lame support. If I was in their position, I'd WANT it back so I could be sure that it was working the way it was supposed to. Of course if somebody had made a mess inside I'd charge accordingly . . .

But for your part. IIRC this is the same as a 2N5566, which is a dual matched JFET. If you google it into the ground, you can probably find one. If not, you might contact Erno Borbely, he's passionate about using matched-JFETs in his own designs, and may have some ideas for substitutes. Also there's a company called Linear Systems that makes matched JFETs . . . I've tested some of their samples and they seemed to work well.

www.borbelyaudio.com

www.linearsystems.com

If your technician is making substitutions, make sure that he/she is properly equipped to check for parasitic oscillations at a wide range of frequencies and power levels. Checking for IMD isn't a bad idea, either. I'm assuming that Spectral is using this part in an input differential amplifier, so substitutions/problems here can definitely affect performance.