LM9022 on ebay

[Grahame Marsh]

Hi

The LM9022 IC which is a fairly niffty VFD driver has turned up on ebay
item 171472657676

No affiliation.

Grahame

[Quixotic Nixotic]

I never understood why I was still ordering and getting free samples, long after everyone on this group was saying they were unobtanium. Oh well.

John S

[Dekatron42]

I found this comment over at Texas instruments forum:

"The LM9022 (VFD Filament Driver) was a marketing spin of the LM4871 (3W Audio Power Amplifier)

http://www.ti.com/product/lm4871"

http://e2e.ti.com/support/power_management/isolated_controllers/f/188/p/289255/1009125.aspx#1009125

Does anyone here at the forum know if this is a substitute for the LM9022, it was a Ti employee who wrote that comment (check the forum thread).

/Martin

[threeneurons]

Have to dig into the details, but I could see it. I've driven VFD filaments from the much older LM1877, before National got swallowed up by TI. Its a stereo amp, which I rigged in a "bi-phase" configuration, much like the LM4871. All these parts belonged to National. I wonder if that employee was a National employee, and came along with the acquisition ?  

[taylorjpt]

The two parts look nearly identical from the datasheets.  Since it was common to simply add an application diagram to an existing datasheet instead of creating a new part number, the LM9022 was probably a way to use LM4871s that failed some of the audio amplifier specs that would be critical to an audio aplication but would be of no concern for driving a VFD.

I have 200 of the LM9022s, I'll get some of the LM4871s and compare them.

[John Rehwinkel]

> The two parts look nearly identical from the datasheets. Since it was common to simply add an application diagram to an existing datasheet instead of creating a new part number, the LM9022 was probably a way to use LM4871s that failed some of the audio amplifier specs that would be critical to an audio aplication but would be of no concern for driving a VFD.
>
> I have 200 of the LM9022s, I'll get some of the LM4871s and compare them.

I'm tempted to get one of each and have a friend decap and examine them.

- John

[taylorjpt]

//I'm tempted to get one of each and have a friend decap and examine them.
// -John

That's actually not a bad idea.  When the LM4871s come in I'll do just that.

[John Rehwinkel]

> //I'm tempted to get one of each and have a friend decap and examine them.
> // -John
>
> That's actually not a bad idea. When the LM4871s come in I'll do just that.

Oh, you have the facilities for that? Because I'm all curious now.

- John

[Dekatron42]

Why not ask TI directly? Anyone here who can do that?

/Martin

[threeneurons]

From reading both datasheets, the specs are so close, if not identical, they are the same part. Even the pinout is the same ! So use the LM4871.

Cutting the top off of one, and looking at the guts, may not reveal anything. Chip makers periodically change the mask set of all ICs, to keep up with the current processes, that they have on the line. I had a supervisor, once, who worked for a semi house. He said they changed mask sets at least once a year. And this is for old established parts. A LM358, today, won't look like one made in 2003, and both different from one made in 1983. They can only keep so many lines, and the wafers just get bigger.

[taylorjpt]

Pulled the die from one of my ca 2001 LM9022s and it was indeed an LM4871 (Picture later).  This makes me think that the LM9022 is a way to absorb fallout from the LM4871 line, and if they don't have enough fallout they could just use good LM4871s instead.

jt

[taylorjpt]

Images of LM9022 die...

http://ppl.ug/PCAptU3GN6g/
http://ppl.ug/5DTx4xMnVT4/
http://ppl.ug/dDfWdwPaa4I/

[Dekatron42]

That is most interesting! Thank you for taking the time and effort to show us this!

/Martin

[Nick]

You get special points for that - very interesting!

Cheers

Nick

[threeneurons]

That's pretty definitive ! A LM9022 is a LM4871.

Yes, extra brownie points, for opening one up, and the photos ! I didn't expect to see the actual part number ! An internal code, that translates to a part number, but not the actual part number.

http://www.mouser.com/Semiconductors/Audio-ICs/Audio-Amplifiers/_/N-4gxtfZscv7?P=1z0z7v5&Keyword=LM4871&FS=True

Mouser has over 4000 of them. Not bad ones-n-two-zees prices. $1.42 on a reel tape, $1.71 in a rigid tube. And a good price break if you get 10 or more. Probably same prices at Digi-Key and Newark (Farnell).

[threeneurons]

Mouser has over 4000 of them. Not bad ones-n-two-zees prices. $1.42 on a reel tape, $1.71 in a rigid tube. And a good price break if you get 10 or more. Probably same prices at Digi-Key and Newark (Farnell).

 http://www.newark.com/webapp/wcs/stores/servlet/Search?catalogId=15003&langId=-1&storeId=10194&st=lm4871&pageSize=25&showResults=true&sf=783

Newark is currently running a special on them ... 68 cents US, and they only have ~1200 of them !

[taylorjpt]

These are the exact same parts (Mechanically, perhaps not performance wise), right down to the mask set which has not changed since 2001:  Every layer in the set has the same revision between these two parts.

New LM4871 from Digikey:
http://ppl.ug/i4CUcbOKElY/

From the LM4871 data sheet:
http://ppl.ug/j1G65GS5644/

LM4871 date code is 32AC = 2003 or 2013, weeks 12 to 17, Die run "AC"
(Probably 2013 because Digikey has thousands of them.  It looks like these were introduced in 1997: http://ppl.ug/n7kz2l2RUmo/)

LM9022 date code is 449E = 2004, weeks 24 to 29, Die run "9E"
(Not 1984 or 1994 etc as the mask set is from 2001, not 2014 as I owned these many years ago.  It looks like these were released in 2001: http://ppl.ug/myU2AVIrnZc/)

jt