The No.1 Website for Pro Audio
LM394 replacement?
Old 25th December 2009
  #1
Gear Guru
 
u b k's Avatar
 

LM394 replacement?

These pups are gonna be obsolete in Summer 2010, it'd really be nice to have a drop-in replacement with a manufacturing future, but I'm not seeing anything that fits the bill.

Ideas?


Gregory Scott - ubk
Old 25th December 2009
  #2
Old 26th December 2009
  #3
Gear Guru
 
u b k's Avatar
 

Thanks... unfortunately, both of those devices are discontinued as well.

The thing is, I actually like the slurry darkness of the 394.


Gregory Scott - ubk
Old 26th December 2009
  #4
Here for the gear
 
[silent:arts]'s Avatar
 

would be interesting to know what John Hardy will do with his 990 ...
Old 26th December 2009
  #5
Gear Guru
 
u b k's Avatar
 

I was wondering the same thing; seventh circle will eventually face the same issue.


Gregory Scott - ubk
Old 26th December 2009
  #6
Not a problemo. Resellers like America II electronics and Part Miner can source NOS parts for you. You may pay, but there are plenty of parts out there.

Matched pairs are also available from Toshiba and Hitachi although Hitachi is also now out of the discrete transistor biz.

There are also matched arrays from that corp, the 300 series. I've had little luck with those, lower betas and higher noise than I would use.

Even if you match up a pair of discretes on a curve tracer, the thermal differences will still cause gain drift. You will need to either use a monolythic solution or maybe try to enclose the pair with silicon thermal grease and a copper clip to attach them face to face. dbx used to use some adhesive backed aluminum foil.

Jim Williams
Audio Upgrades
Old 26th December 2009
  #7
Gear Maniac
 

I was given conflicting information by people at Analog Devices, but the most recent information is this:

1. The FAB where the MAT02 was made was closed. Therefore, they decided to discontinue the MAT02.

2. Subsequently, there was strong demand for it, so they changed their minds.

3. They decided to redesign the part. I'm not sure if it was because of tooling issues, machinery incompatibilities, newer and better equipment or what, but they decided to redesign the part.

4. The new part will be known as the "MAT12". Samples in March, 2010. Production in June, 2010.

John Hardy
Old 26th December 2009
  #8
Lives for gear
 
analogtodd's Avatar
 

So does this mean the MAT12 will be a modern, in production, replacement for the LM394?
Old 26th December 2009
  #9
Gear Maniac
 

Quote:
So does this mean the MAT12 will be a modern, in production, replacement for the LM394?
It all depends on how Analog Devices redesigns the MAT02, and the capabilities of the FAB where the MAT12 will be manufactured. There are plenty of ways that the new design could become screwed up, but it could also turn out to be superior. Hopefully Analog Devices still has people that know how to design analog parts, and they still have manufacturing facilities that can make great analog parts. It is Analog Devices, after all.

If the MAT12 is as good, or better than, the MAT02, it will be an excellent part and a suitable replacement for the LM394. Of course, any time one part is substituted for another, the performance should be carefully studied and verified.

John Hardy
Old 26th December 2009
  #10
Here for the gear
 
[silent:arts]'s Avatar
 

John, thanks for the information.
for sure I'm hoping the MAT12 will suite your needs.
Old 27th December 2009
  #11
Lives for gear
 
dcollins's Avatar
Quote:
Originally Posted by John Hardy View Post
If the MAT12 is as good, or better than, the MAT02, it will be an excellent part and a suitable replacement for the LM394. Of course, any time one part is substituted for another, the performance should be carefully studied and verified.
How have past users of the LM-394 based 990 been able to cope with the "slurry darkness" ascribed to this part, and will a replacement NPN pair be sonically compatible with existing devices in the field?

What parameter in the transistor causes "slurry darkness," anyway? Ft, Rbb?

???


DC
Old 27th December 2009
  #12
Gear Guru
 
u b k's Avatar
 

Quote:
Originally Posted by dcollins View Post
What parameter in the transistor causes "slurry darkness," anyway? Ft, Rbb?

I don't have a clue, and I suspect that as always the sound is not inherent to the component itself, it's due more to the implementation of it and the way it reacts to what comes before and what goes after.


Gregory Scott - ubk
Old 27th December 2009
  #13
I'd like to find a pair of MAT-02 to swap out with the LM394's that are in my Ashly SC-50. It's one of the few components I haven't changed.
Old 29th December 2009
  #14
Gear Maniac
 

I have rarely heard any comments about sonic differences between the LM394, MAT02 and SSM2210 in a 990. "Slurry darkness" is not a comment that I have heard before. Sounds like a good band name.

I have always used the LM394 out of respect for Deane Jensen who designed the 990 circuit. I have no problem with using another device, and will obviously have to use another device at some point since the LM394 will be discontinued in mid 2010. Surplus stock will be available for some time, but eventually another device will have to be used.

I will contact the folks at Jensen to see what their concerns are. It may require a product name change, it may not. Perhaps I will need to change the color of the markings, or some other method of differentiating LM394-based 990s from MAT12-based 990s, or whatever other matched pair ends up being used.

Perhaps this will be the beginning of endless discussions of the sonic virtues of "old" 990s vs. "new" 990s. I believe the 990 will still be an excellent discrete op-amp regardless of what matched-pair of transistors is used.

John Hardy
Old 29th December 2009
  #15
I began replacing LM394's the month the MAT02 came out. They went into MCI tape machine repro amps, series 600 console mic preamps, Pacific Recorders 990 opamps, anywhere a 394 was found. Yes, I hear the "slurry darkness" of the LM394, an excellent description. Take gain on these and differences to other transistors are not hard to discern. Mic preamps are the best test bed as they tend to exhibit the sonic differences at higher gains. A socket and some quality discretes will be informative.

I believe AD could reproduce the MAT02 without problems. They have the dies and masks, any bipolar fab should be able to do it. I believe they dropped it because they can do a much better new part now. They have some very advanced layout tools that were not available just a few years ago. They have been producing advanced linear devices using new thermal and layout matching techniques that compensate for thermal induced drift in the die itself. They have been producing excellent opamps with .9 nv/hz/sq noise specs along with well over 80 db open loop gain at 10 k hz. The dc precision is stunning on those. I can take 46 db of gain without any blocking caps and still have under 2 mv dc offset.

I expect them to produce a .7 nv noise or less part with unprecedented dc precision. They will exceed the MAT02 ft. If they lower the input bias current that would be a plus. I expect a superior device as I've seen what those guys can do.
The big if is whether they can get the .5 nv noise specs a 20 cent Japanese transistor can already do.

Jim Williams
Audio Upgrades
Old 30th December 2009
  #16
Gear Guru
 
u b k's Avatar
 

Quote:
Originally Posted by Jim Williams View Post
They have been producing excellent opamps with .9 nv/hz/sq noise specs along with well over 80 db open loop gain at 10 k hz. The dc precision is stunning on those. I can take 46 db of gain without any blocking caps and still have under 2 mv dc offset.

It's crazy how far op-amps have come in the last few years. How do those AD's compare to something like a 49720?


Gregory Scott - ubk
.
Old 30th December 2009
  #17
Lives for gear
 
brianroth's Avatar
 

Wow, I finally received the last 10 LM394 parts that Arrow had. Stash for future....

Bri
Old 4th January 2010
  #18
Lives for gear
 
dcollins's Avatar
Quote:
Originally Posted by John Hardy View Post
I have rarely heard any comments about sonic differences between the LM394, MAT02 and SSM2210 in a 990. "Slurry darkness" is not a comment that I have heard before. Sounds like a good band name.
It's confusing. Like the guy that wouldn't work with Doug Sax because he used 2N3391A's in his console.


DC
Old 5th January 2010
  #19
Quote:
Originally Posted by dcollins View Post
It's confusing. Like the guy that wouldn't work with Doug Sax because he used 2N3391A's in his console.
DC
That was ME! No, just kidding. I did use Doug a lot back in the early 80's before CD's. He was always helpful in an extra way to me as I didn't have much money and built my own stuff to record with. Doug and Lincoln were very impressed with what I was doing on a shoestring budget, more knowledge than cash. He gave me nice comps for the sounds I brought. We would talk before the sessions discussing electronics, my custom console and reverbs (they were springs). We would always set up a 10 am session so I would be out the door by noon and then I would fly across Hollyweird to Sheffield labs for the 1 o'clock bath.
That was to preserve the transients of the laquers as they start to soften right after cutting.

We discussed the head amps in his MCI decks. He showed them to me. He used Rel Cap polyprops and polystyrenes and those 2N3391A transistors. I suggested the 2N5089 as a superior replacement, that's what I switched to in the mid 70's after dumping the 3391's in those on board guitar preamps I designed for Rex Bogue and used by Zappa and Mahavishnu.

I don't know if Doug ever did swap those transistors, but there are some real wing-dingers out now days... the Toshiba 2SC3329 is an excellent part.

Jim Williams
Audio Upgrades
Old 5th January 2010
  #20
Quote:
Originally Posted by u b k View Post
It's crazy how far op-amps have come in the last few years. How do those AD's compare to something like a 49720?
Gregory Scott - ubk
.
Better. Lower noise, higher loop gain, greater bandwidth, better precision, better output drive, more details, less transistor sound. They suck much more current so power is an issue. LME's sound just a bit closed off next to AD's best efforts. That includes the BB OPA1610's as well.

I'm finding less and less to complain about. That is a very good sign.

Jim Williams
Audio Upgrades
Old 5th January 2010
  #21
Lives for gear
 
JohnRoberts's Avatar
 

Quote:
Originally Posted by Jim Williams View Post
just a bit closed off next to AD's best efforts. That includes the BB OPA1610's as well.

I'm finding less and less to complain about. That is a very good sign.

Jim Williams
Audio Upgrades
More evidence that the end is near...

JR
Old 5th January 2010
  #22
Quote:
Originally Posted by JohnRoberts View Post
More evidence that the end is near...

JR
Dec. 21st, 2012. Not long to go now...

Jim Williams
Audio Upgrades
Old 16th February 2010
  #23
Here for the gear
 

A possible replacement is the LS312 from Linear Systems in California. It is pin for pin compatible and has similar specs. The datasheet is quite sparse for info though. It lists the Hfe matching at 5% typical, but their engineering support says that he has consistently achieved less than 1% over the past 15 - 20 yrs.
Old 20th February 2010
  #24
Lives for gear
 
Thermionic's Avatar
 

How come no one's mentioning the THAT 300 series: Transistor Arrays

?

Justin
Old 21st February 2010
  #25
Lives for gear
 
brianroth's Avatar
 

The packages are 14 pin, vs the 6/8 pin arrangements of the original.

Best,

Bri
Old 21st February 2010
  #26
Lives for gear
 
Thermionic's Avatar
 

Quote:
Originally Posted by brianroth View Post
The packages are 14 pin, vs the 6/8 pin arrangements of the original.

Best,

Bri
I did notice that, funnily enough :-) Unless it has to be a small footprint opamp, I don't see an issue. I just wondered if anyone had tried them, as I've been meaning to for some time..

I've watched the harmonics and x-talk increase on an FFT as discrete transistors in an LTP warm up. It certainly would be a hell of a task to get similar CMR using a discrete pair (there are other factors, I know - such as rail voltage, bias etc. - some designs will experience more drift than others).

Justin
Old 21st February 2010
  #27
Gear Guru
 
u b k's Avatar
 

Quote:
Originally Posted by Thermionic View Post
How come no one's mentioning the THAT 300 series: Transistor Arrays

I haven't tested those myself because they seem like they'd be way too noisy compared to the lm394/mat012.


Gregory Scott - ubk
Old 21st February 2010
  #28
Lives for gear
 
Thermionic's Avatar
 

Quote:
I haven't tested those myself because they seem like they'd be way too noisy compared to the lm394/mat012.

http://www.thatcorp.com/datashts/300data.pdf

http://www.national.com/ds/LM/LM194.pdf
Old 22nd February 2010
  #29
I played with the that 300 series arrays. They are pinned out for a AD MAT 04, but the center pins are the substrate meant to be grounded. They have even lower betas than the SSM parts and I've not had much luck subbing MAT 04's with them. Only 60 for NPN and 75 for PNP. What are 394's, 1200 or so?

Jim Williams
Audio Upgrades
Old 22nd February 2010
  #30
Lives for gear
 
Thermionic's Avatar
 

Quote:
Originally Posted by Jim Williams View Post
I played with the that 300 series arrays. They are pinned out for a AD MAT 04, but the center pins are the substrate meant to be grounded. They have even lower betas than the SSM parts and I've not had much luck subbing MAT 04's with them. Only 60 for NPN and 75 for PNP. What are 394's, 1200 or so?

Jim Williams
Audio Upgrades
That's what makes me curious about the 300. Why would they release something with such low hfe? Aside from beta, the specs look good. Even the humble old 2cent BC550C gives 5-600. I must be missing something...

Justin
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump