Fast PNP?...

P

Phil Hobbs

Guest
Hi, all,

A couple of years ago the last mainstream maker (NXP) stopped making the
last fast PNP--the 5-GHz BFT92--thus depriving some of us of valuable
design space, e.g. for improving the transconductance of FET bootstraps
by wrapping a much faster PNP around it.

A BF862 or CPH3910 with a BFT92 wraparound makes a 750-MHz FET with 10x
the transconductance but about the same noise. Magic.

I have ~9k BFT92s in stock, so I\'m set for anything I might want to do
myself, but for customer designs it\'s more of a worry. Not too long
ago, the NTE2403 was announced, but stock seems to be, *ahem*, thin.

I ran across this datasheet:
<http://www.unisonic.com.tw/datasheet/2SA1977.pdf>, which is marked
\"preliminary\" and dated 2021. It\'s apparently a closeish second source
for the long lost Renesas/NEC 2SA1977, which was even faster than the
BFT92 (8 GHz f_T).

Anybody know anything about this outfit? I\'ve never heard of them.

Thanks

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
 
On Tue, 10 May 2022 10:20:56 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

Hi, all,

A couple of years ago the last mainstream maker (NXP) stopped making the
last fast PNP--the 5-GHz BFT92--thus depriving some of us of valuable
design space, e.g. for improving the transconductance of FET bootstraps
by wrapping a much faster PNP around it.

A BF862 or CPH3910 with a BFT92 wraparound makes a 750-MHz FET with 10x
the transconductance but about the same noise. Magic.

I have ~9k BFT92s in stock, so I\'m set for anything I might want to do
myself, but for customer designs it\'s more of a worry. Not too long
ago, the NTE2403 was announced, but stock seems to be, *ahem*, thin.

I ran across this datasheet:
http://www.unisonic.com.tw/datasheet/2SA1977.pdf>, which is marked
\"preliminary\" and dated 2021. It\'s apparently a closeish second source
for the long lost Renesas/NEC 2SA1977, which was even faster than the
BFT92 (8 GHz f_T).

Anybody know anything about this outfit? I\'ve never heard of them.

Thanks

Phil Hobbs
Click to expand...

Here too:

https://www.digikey.com/en/products/detail/renesas-electronics-america-inc/2SA1977-T1B-A/11550285

I\'ve had to redesign a few products that used fast pnp\'s in SOT-89.



--

Anybody can count to one.

- Robert Widlar
 
On Tue, 10 May 2022 10:20:56 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

Hi, all,

A couple of years ago the last mainstream maker (NXP) stopped making the
last fast PNP--the 5-GHz BFT92--thus depriving some of us of valuable
design space, e.g. for improving the transconductance of FET bootstraps
by wrapping a much faster PNP around it.

A BF862 or CPH3910 with a BFT92 wraparound makes a 750-MHz FET with 10x
the transconductance but about the same noise. Magic.

I have ~9k BFT92s in stock, so I\'m set for anything I might want to do
myself, but for customer designs it\'s more of a worry. Not too long
ago, the NTE2403 was announced, but stock seems to be, *ahem*, thin.

I ran across this datasheet:
http://www.unisonic.com.tw/datasheet/2SA1977.pdf>, which is marked
\"preliminary\" and dated 2021. It\'s apparently a closeish second source
for the long lost Renesas/NEC 2SA1977, which was even faster than the
BFT92 (8 GHz f_T).

Anybody know anything about this outfit? I\'ve never heard of them.

Thanks

Phil Hobbs
Click to expand...

Here\'s the output stage of our P400 pulse/delay generator, the one you
have.

https://www.dropbox.com/s/gj5252vstfiw9gw/P400_Output_A.pdf?dl=0

It\'s packed with obsolete parts. I replaced it with the GaN fet
mouse-bite thing for the next generation.



--

Anybody can count to one.

- Robert Widlar
 
jlarkin@highlandsniptechnology.com wrote:
On Tue, 10 May 2022 10:20:56 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

Hi, all,

A couple of years ago the last mainstream maker (NXP) stopped
making the last fast PNP--the 5-GHz BFT92--thus depriving some of
us of valuable design space, e.g. for improving the
transconductance of FET bootstraps by wrapping a much faster PNP
around it.

A BF862 or CPH3910 with a BFT92 wraparound makes a 750-MHz FET with
10x the transconductance but about the same noise. Magic.

I have ~9k BFT92s in stock, so I\'m set for anything I might want to
do myself, but for customer designs it\'s more of a worry. Not too
long ago, the NTE2403 was announced, but stock seems to be, *ahem*,
thin.

I ran across this datasheet:
http://www.unisonic.com.tw/datasheet/2SA1977.pdf>, which is
marked \"preliminary\" and dated 2021. It\'s apparently a closeish
second source for the long lost Renesas/NEC 2SA1977, which was even
faster than the BFT92 (8 GHz f_T).

Anybody know anything about this outfit? I\'ve never heard of
them.


Here too:

https://www.digikey.com/en/products/detail/renesas-electronics-america-inc/2SA1977-T1B-A/11550285
Click to expand...
Yeah, there are a fair amount of NOS Renesas ones out there, apparently.
For licensed designs it would be more comforting to be able to point to
current production.

> I\'ve had to redesign a few products that used fast pnp\'s in SOT-89.

I have a reel of BFG31s too. (That\'s only a thousand pieces, of
course.)

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
 
jlarkin@highlandsniptechnology.com wrote:
On Tue, 10 May 2022 10:20:56 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

Hi, all,

A couple of years ago the last mainstream maker (NXP) stopped making the
last fast PNP--the 5-GHz BFT92--thus depriving some of us of valuable
design space, e.g. for improving the transconductance of FET bootstraps
by wrapping a much faster PNP around it.

A BF862 or CPH3910 with a BFT92 wraparound makes a 750-MHz FET with 10x
the transconductance but about the same noise. Magic.

I have ~9k BFT92s in stock, so I\'m set for anything I might want to do
myself, but for customer designs it\'s more of a worry. Not too long
ago, the NTE2403 was announced, but stock seems to be, *ahem*, thin.

I ran across this datasheet:
http://www.unisonic.com.tw/datasheet/2SA1977.pdf>, which is marked
\"preliminary\" and dated 2021. It\'s apparently a closeish second source
for the long lost Renesas/NEC 2SA1977, which was even faster than the
BFT92 (8 GHz f_T).

Anybody know anything about this outfit? I\'ve never heard of them.

Thanks

Phil Hobbs

Here\'s the output stage of our P400 pulse/delay generator, the one you
have.

https://www.dropbox.com/s/gj5252vstfiw9gw/P400_Output_A.pdf?dl=0

It\'s packed with obsolete parts. I replaced it with the GaN fet
mouse-bite thing for the next generation.
Click to expand...

Nice gizmo, though.

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
 
Am 10.05.22 um 17:35 schrieb Phil Hobbs:
jlarkin@highlandsniptechnology.com wrote:
On Tue, 10 May 2022 10:20:56 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

Hi, all,

A couple of years ago the last mainstream maker (NXP) stopped
making the last fast PNP--the 5-GHz BFT92--thus depriving some of
us of valuable design space, e.g. for improving the
transconductance of FET bootstraps by wrapping a much faster PNP
around it.

A BF862 or CPH3910 with a BFT92 wraparound makes a 750-MHz FET with
10x the transconductance but about the same noise.  Magic.

I have ~9k BFT92s in stock, so I\'m set for anything I might want to
do myself, but for customer designs it\'s more of a worry.  Not too
long ago, the NTE2403 was announced, but stock seems to be, *ahem*,
thin.

I ran across this datasheet:
http://www.unisonic.com.tw/datasheet/2SA1977.pdf>, which is
marked \"preliminary\" and dated 2021.  It\'s apparently a closeish
second source for the long lost Renesas/NEC 2SA1977, which was even
faster than the BFT92 (8 GHz f_T).

Anybody know anything about this outfit?  I\'ve never heard of
them.
Click to expand...

I think Renesas is the sum of NEC semi, Hitachi, Harris, Intersil.

There is still the HFA3096, a transistor array, Si on isolator.
Some members of this family have 5.5 GHz PNPs, still active.
It was easy to get them space qualified a few years ago
for a dual slope time stretcher. In a ceramic flatpack where
nobody dared to shorten the extra long leads. :)


<
https://www.digikey.de/de/products/filter/transistoren-bipolar-bjt-hf/283?s=N4IgTCBcDaIBYDMCGBmADATgGwgLoF8g
Here too:

https://www.digikey.com/en/products/detail/renesas-electronics-america-inc/2SA1977-T1B-A/11550285



Yeah, there are a fair amount of NOS Renesas ones out there, apparently.
 For licensed designs it would be more comforting to be able to point to
current production.

I\'ve had to redesign a few products that used fast pnp\'s in SOT-89.

I have a reel of BFG31s too. (That\'s only a thousand pieces, of
course.)
Click to expand...

#me too. Ask me if you are desperate. I don\'t need production quantities.

Is quantitties a word? It should be. Sugestive as hell. :)

cheers, Gerhard
 
Gerhard Hoffmann wrote:
Am 10.05.22 um 17:35 schrieb Phil Hobbs:
jlarkin@highlandsniptechnology.com wrote:
On Tue, 10 May 2022 10:20:56 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

Hi, all,

A couple of years ago the last mainstream maker (NXP) stopped
making the last fast PNP--the 5-GHz BFT92--thus depriving some of
us of valuable design space, e.g. for improving the
transconductance of FET bootstraps by wrapping a much faster PNP
around it.

A BF862 or CPH3910 with a BFT92 wraparound makes a 750-MHz FET with
10x the transconductance but about the same noise.  Magic.

I have ~9k BFT92s in stock, so I\'m set for anything I might want to
do myself, but for customer designs it\'s more of a worry.  Not too
long ago, the NTE2403 was announced, but stock seems to be, *ahem*,
thin.

I ran across this datasheet:
http://www.unisonic.com.tw/datasheet/2SA1977.pdf>, which is
marked \"preliminary\" and dated 2021.  It\'s apparently a closeish
second source for the long lost Renesas/NEC 2SA1977, which was even
faster than the BFT92 (8 GHz f_T).

Anybody know anything about this outfit?  I\'ve never heard of
them.

I think Renesas is the sum of NEC semi, Hitachi, Harris, Intersil.

There is still the HFA3096, a transistor array, Si on isolator.
Some members of this family have 5.5 GHz PNPs, still active.
It was easy to get them space qualified a few years ago
for a dual slope time stretcher. In a ceramic flatpack where
nobody dared to shorten the extra long leads.  :)
Click to expand...

Yeah, the package parasitics and especially the R_ee\' are pretty
disappointing. I looked at them carefully for laser noise canceller
use, but they weren\'t much use. A pity.

https://www.digikey.de/de/products/filter/transistoren-bipolar-bjt-hf/283?s=N4IgTCBcDaIBYDMCGBmADATgGwgLoF8g


Here too:

https://www.digikey.com/en/products/detail/renesas-electronics-america-inc/2SA1977-T1B-A/11550285



Yeah, there are a fair amount of NOS Renesas ones out there, apparently.
  For licensed designs it would be more comforting to be able to point to
current production.

I\'ve had to redesign a few products that used fast pnp\'s in SOT-89.

I have a reel of BFG31s too. (That\'s only a thousand pieces, of
course.)

#me too. Ask me if you are desperate. I don\'t need production quantities.

Is quantitties a word? It should be. Sugestive as hell.   :)

cheers, Gerhard
Click to expand...

You need to get out more. ;)

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
 
On Tue, 10 May 2022 11:40:07 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

jlarkin@highlandsniptechnology.com wrote:
On Tue, 10 May 2022 10:20:56 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

Hi, all,

A couple of years ago the last mainstream maker (NXP) stopped making the
last fast PNP--the 5-GHz BFT92--thus depriving some of us of valuable
design space, e.g. for improving the transconductance of FET bootstraps
by wrapping a much faster PNP around it.

A BF862 or CPH3910 with a BFT92 wraparound makes a 750-MHz FET with 10x
the transconductance but about the same noise. Magic.

I have ~9k BFT92s in stock, so I\'m set for anything I might want to do
myself, but for customer designs it\'s more of a worry. Not too long
ago, the NTE2403 was announced, but stock seems to be, *ahem*, thin.

I ran across this datasheet:
http://www.unisonic.com.tw/datasheet/2SA1977.pdf>, which is marked
\"preliminary\" and dated 2021. It\'s apparently a closeish second source
for the long lost Renesas/NEC 2SA1977, which was even faster than the
BFT92 (8 GHz f_T).

Anybody know anything about this outfit? I\'ve never heard of them.

Thanks

Phil Hobbs

Here\'s the output stage of our P400 pulse/delay generator, the one you
have.

https://www.dropbox.com/s/gj5252vstfiw9gw/P400_Output_A.pdf?dl=0

It\'s packed with obsolete parts. I replaced it with the GaN fet
mouse-bite thing for the next generation.

Nice gizmo, though.

Cheers

Phil Hobbs
Click to expand...

Everybody should have one.

--

If a man will begin with certainties, he shall end with doubts,
but if he will be content to begin with doubts he shall end in certainties.
Francis Bacon
 
On 10/05/2022 3:20 pm, Phil Hobbs wrote:
Hi, all,

A couple of years ago the last mainstream maker (NXP) stopped making the
last fast PNP--the 5-GHz BFT92--thus depriving some of us of valuable
design space, e.g. for improving the transconductance of FET bootstraps
by wrapping a much faster PNP around it.

A BF862 or CPH3910 with a BFT92 wraparound makes a 750-MHz FET with 10x
the transconductance but about the same noise.  Magic.

I have ~9k BFT92s in stock, so I\'m set for anything I might want to do
myself, but for customer designs it\'s more of a worry.  Not too long
ago, the NTE2403 was announced, but stock seems to be, *ahem*, thin.

I ran across this datasheet:
http://www.unisonic.com.tw/datasheet/2SA1977.pdf>, which is marked
\"preliminary\" and dated 2021.  It\'s apparently a closeish second source
for the long lost Renesas/NEC 2SA1977, which was even faster than the
BFT92 (8 GHz f_T).

Anybody know anything about this outfit?  I\'ve never heard of them.

Thanks

Phil Hobbs
Click to expand...

Well I have heard of Unisonic and seen their parts in other gear but
have no direct experience.

There is still the 2SA1669 at 3GHz?

piglet
 
piglet wrote:
On 10/05/2022 3:20 pm, Phil Hobbs wrote:
Hi, all,

A couple of years ago the last mainstream maker (NXP) stopped making
the last fast PNP--the 5-GHz BFT92--thus depriving some of us of
valuable design space, e.g. for improving the transconductance of FET
bootstraps by wrapping a much faster PNP around it.

A BF862 or CPH3910 with a BFT92 wraparound makes a 750-MHz FET with
10x the transconductance but about the same noise.  Magic.

I have ~9k BFT92s in stock, so I\'m set for anything I might want to do
myself, but for customer designs it\'s more of a worry.  Not too long
ago, the NTE2403 was announced, but stock seems to be, *ahem*, thin.

I ran across this datasheet:
http://www.unisonic.com.tw/datasheet/2SA1977.pdf>, which is marked
\"preliminary\" and dated 2021.  It\'s apparently a closeish second
source for the long lost Renesas/NEC 2SA1977, which was even faster
than the BFT92 (8 GHz f_T).

Anybody know anything about this outfit?  I\'ve never heard of them.



Well I have heard of Unisonic and seen their parts in other gear but
have no direct experience.

There is still the 2SA1669 at 3GHz?
Click to expand...

Just from brokers--no current production.

As I say, I have three reels of BFT92s, from which I can probably build
$10M worth of TIAs if need be. (Just ask!) ;)

The issue is licensed designs, for which customers naturally ask
questions about long term parts availability.

Maybe Unisonic will come through where NTE apparently whiffed.

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
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top