Well, it happened--the last fast PNP is EOL

[Robert Baer]

Phil Hobbs wrote:

On 06/04/2018 12:03 PM, Gerhard Hoffmann wrote:
Am 04.06.2018 um 17:54 schrieb Phil Hobbs:
BFT92, RIP.

NXP, you lousy bastards, you just took away about a quarter of my design
space. Get 'em while they last.



https://www.intersil.com/en/products/space-and-harsh-environment/harsh-environment/transistor-arrays/HFA3096.html


Let's hope that at least _these_ stay for some time..

cheers, Gerhard


Yeah, true, there are those. Unfortunately their Rbb' and Ree' are the
pits.

I just bought Newark's last reel of BFT92s, so we'll be okay for our own
stuff, but I can't use them in custom or licensed designs any more.

Which is a great pity--fast PNP wraparounds are good for a lot of things.

Cheers

Phil Hobbs

Stupid idea: get a fab company to make a custom part..

 

On Tuesday, June 5, 2018 at 1:54:21 AM UTC+10, Phil Hobbs wrote:

BFT92, RIP.

NXP, you lousy bastards, you just took away about a quarter of my design
space. Get 'em while they last.

I just made my prototype oscilloscope amplifier probe using these. The next best devices I could find are ON's MMBTH81 ft=600MHz(min) @ 1/8 speed. How else is one suppose to provide drive to the positive rail !!!

Is this the first sign of the great technological unwinding of humanity.

 

[John Larkin]

On Sun, 29 Jul 2018 08:59:38 -0700 (PDT), pcdhobbs@gmail.com wrote:

On Sunday, July 29, 2018 at 11:40:25 AM UTC-4, adam.s...@gmail.com wrote:
On Tuesday, June 5, 2018 at 1:54:21 AM UTC+10, Phil Hobbs wrote:
BFT92, RIP.

NXP, you lousy bastards, you just took away about a quarter of my design
space. Get 'em while they last.


I just made my prototype oscilloscope amplifier probe using these. The next best devices I could find are ON's MMBTH81 ft=600MHz(min) @ 1/8 speed. How else is one suppose to provide drive to the positive rail !!!

Is this the first sign of the great technological unwinding of humanity.

No, we're well down the curve. It was IBM's discontinuing OS/2 that started it.

You used to be able to get 8.5 GHz PNPs (I have 100 or so of them).

Cheers

Phil Hobbs

The early days of analog IC design was all NPNs. There were some truly
awful all-NPN opamps.

Back to the future, I suppose.

I spent most of my adulthood perfecting a pulse generator output stage
circuit. When I finally got it almost right, the critical PNPs went
EOL.

And as noted here, a billion-dollar product line has stopped because a
50 cent NEC transistor went EOL. That's either tragic or hilarious.





--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

On Sunday, July 29, 2018 at 11:40:25 AM UTC-4, adam.s...@gmail.com wrote:

On Tuesday, June 5, 2018 at 1:54:21 AM UTC+10, Phil Hobbs wrote:
BFT92, RIP.

NXP, you lousy bastards, you just took away about a quarter of my design
space. Get 'em while they last.


I just made my prototype oscilloscope amplifier probe using these. The next best devices I could find are ON's MMBTH81 ft=600MHz(min) @ 1/8 speed. How else is one suppose to provide drive to the positive rail !!!

Is this the first sign of the great technological unwinding of humanity.

No, we're well down the curve. It was IBM's discontinuing OS/2 that started it.

You used to be able to get 8.5 GHz PNPs (I have 100 or so of them).

Cheers

Phil Hobbs

 

[Phil Hobbs]

On 07/29/2018 12:15 PM, John Larkin wrote:

On Sun, 29 Jul 2018 08:59:38 -0700 (PDT), pcdhobbs@gmail.com wrote:

On Sunday, July 29, 2018 at 11:40:25 AM UTC-4, adam.s...@gmail.com wrote:
On Tuesday, June 5, 2018 at 1:54:21 AM UTC+10, Phil Hobbs wrote:
BFT92, RIP.

NXP, you lousy bastards, you just took away about a quarter of my design
space. Get 'em while they last.


I just made my prototype oscilloscope amplifier probe using these. The next best devices I could find are ON's MMBTH81 ft=600MHz(min) @ 1/8 speed. How else is one suppose to provide drive to the positive rail !!!

Is this the first sign of the great technological unwinding of humanity.

No, we're well down the curve. It was IBM's discontinuing OS/2 that started it.

You used to be able to get 8.5 GHz PNPs (I have 100 or so of them).


The early days of analog IC design was all NPNs. There were some truly
awful all-NPN opamps.

Which ones? Even the 709 had lateral PNPs.

The LM318 was probably the apogee. Of course it also had lateral PNPs
in it, but to get it to go fast you had to bypass the level shifters by
putting capacitors to the Vos adjust pins.

National had an app note where they tied both inputs to VEE and used a
JFET matched pair with its collectors going to the Vos pins.

Back to the future, I suppose.

I spent most of my adulthood perfecting a pulse generator output stage
circuit. When I finally got it almost right, the critical PNPs went
EOL.

And as noted here, a billion-dollar product line has stopped because a
50 cent NEC transistor went EOL. That's either tragic or hilarious.

Both, depending on whether you're considering the folks that are going
to get laid off or the ones whose heads are exploding (but will probably
keep their jobs).

There are workarounds, such as real vs. folded cascodes and current
source pullups with real or simulated inductors in series, but they all
cost headroom, current, complexity, performance, or (usually) all four.

OTOH that all-NPN cascoded White follower circuit I posted last year
works much better than the PNP wraparound, so all is not necessarily
lost.

Cheers

Phil Hobbs


--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[Phil Hobbs]

On 07/29/2018 02:32 PM, Phil Hobbs wrote:

On 07/29/2018 12:15 PM, John Larkin wrote:
On Sun, 29 Jul 2018 08:59:38 -0700 (PDT), pcdhobbs@gmail.com wrote:

On Sunday, July 29, 2018 at 11:40:25 AM UTC-4, adam.s...@gmail.com
wrote:
On Tuesday, June 5, 2018 at 1:54:21 AM UTC+10, Phil Hobbs wrote:
BFT92, RIP.

NXP, you lousy bastards, you just took away about a quarter of my
design
space.  Get 'em while they last.


I just made my prototype oscilloscope amplifier probe using these.
The next best  devices I could find are ON's MMBTH81 ft=600MHz(min)
@ 1/8 speed. How else is one suppose to provide drive to the
positive rail !!!

Is this the first sign of the great technological unwinding of
humanity.

No, we're well down the curve.  It was IBM's discontinuing OS/2 that
started it.

You used to be able to get 8.5 GHz PNPs (I have 100 or so of them).


The early days of analog IC design was all NPNs. There were some truly
awful all-NPN opamps.

Which ones?  Even the 709 had lateral PNPs.

The LM318 was probably the apogee.  Of course it also had lateral PNPs
in it, but to get it to go fast you had to bypass the level shifters by
putting capacitors to the Vos adjust pins.

National had an app note where they tied both inputs to VEE and used a
JFET matched pair with its collectors going to the Vos pins.

Ah, OK, the uA702 was all-NPN, and it was pretty horrible.

The old TI datasheet for their version has a schematic:
<https://pdf1.alldatasheet.com/datasheet-pdf/view/108088/TI/UA702M.html>

The tail current generator is interesting.

Cheers

Phil Hobbs

Back to the future, I suppose.

I spent most of my adulthood perfecting a pulse generator output stage
circuit. When I finally got it almost right, the critical PNPs went
EOL.

And as noted here, a billion-dollar product line has stopped because a
50 cent NEC transistor went EOL. That's either tragic or hilarious.

Both, depending on whether you're considering the folks that are going
to get laid off or the ones whose heads are exploding (but will probably
keep their jobs).

There are workarounds, such as real vs. folded cascodes and current
source pullups with real or simulated inductors in series, but they all
cost headroom, current, complexity, performance, or (usually) all four.

OTOH that all-NPN cascoded White follower circuit I posted last year
works much better than the PNP wraparound, so all is not necessarily
lost.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[John Larkin]

On Sun, 29 Jul 2018 14:44:40 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

On 07/29/2018 02:32 PM, Phil Hobbs wrote:
On 07/29/2018 12:15 PM, John Larkin wrote:
On Sun, 29 Jul 2018 08:59:38 -0700 (PDT), pcdhobbs@gmail.com wrote:

On Sunday, July 29, 2018 at 11:40:25 AM UTC-4, adam.s...@gmail.com
wrote:
On Tuesday, June 5, 2018 at 1:54:21 AM UTC+10, Phil Hobbs wrote:
BFT92, RIP.

NXP, you lousy bastards, you just took away about a quarter of my
design
space.  Get 'em while they last.


I just made my prototype oscilloscope amplifier probe using these.
The next best  devices I could find are ON's MMBTH81 ft=600MHz(min)
@ 1/8 speed. How else is one suppose to provide drive to the
positive rail !!!

Is this the first sign of the great technological unwinding of
humanity.

No, we're well down the curve.  It was IBM's discontinuing OS/2 that
started it.

You used to be able to get 8.5 GHz PNPs (I have 100 or so of them).


The early days of analog IC design was all NPNs. There were some truly
awful all-NPN opamps.

Which ones?  Even the 709 had lateral PNPs.

The LM318 was probably the apogee.  Of course it also had lateral PNPs
in it, but to get it to go fast you had to bypass the level shifters by
putting capacitors to the Vos adjust pins.

National had an app note where they tied both inputs to VEE and used a
JFET matched pair with its collectors going to the Vos pins.

Ah, OK, the uA702 was all-NPN, and it was pretty horrible.

The old TI datasheet for their version has a schematic:
https://pdf1.alldatasheet.com/datasheet-pdf/view/108088/TI/UA702M.html

The tail current generator is interesting.

Cheers

Phil Hobbs

GE made an opamp that used a zener internally as a level shifter. It
made a nice noise generator.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[Simon S Aysdie]

On Tuesday, June 5, 2018 at 7:10:43 AM UTC-7, Phil Hobbs wrote:

On 06/04/2018 05:41 PM, Jim Thompson wrote:
On 4 Jun 2018 14:04:31 -0700, Winfield Hill <hill@rowland.harvard.edu
wrote:

bitrex wrote...


Should start a s.e.d. fabless semiconductor company and get them made
again, China will make whatever you like. You could advertise it as
exactly that "The Last Fast PNP" like the Last of the Mohicans or something.

It's be nice to have it available in a SOT-323 SC70 package.

Maybe you and Hobbs should buy a wafer or two? Then, as time moves
on, package them to suit the era?

...Jim Thompson


Since they're unique parts, perhaps we could persuade Lansdale or
Rochester to make them--the litho resolution would be doable with very
old equipment. Do you still have contacts there? I'd really like to be
able to get the NE97733 again, too (8.5 GHz PNP).

Cheers

Phil Hobbs

I realize this is very stale.

Anyway, I had a call for a JFET recently. The J232 and perhaps 2N4116A looked okay for what I needed. Only Interfet makes the J232 anymore, and I've never had experience with them.

OnSemi still has legacy stuff but has EOL'd some things. It seems that Central Semi, Linear Systems, and Interfet live on legacy, and even welcome it. Central Semi and Linear Systems do bipolar parts, unlike Interfet. It made me think of this old thread. (I don't know the fabs they use.)

I haven't really done much with Central Semi & Linear Systems, so I don't know what they are capable of. Did you consider a query to them on the BFT92?