Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Modern Transistor for Avalanche in SMT
235 views
Skip to first unread message
Owen Roberts
Oct 8, 2012, 10:32:49 AM10/8/12
to
Ok, I'm tired of hunting old mesa transistors on Ebay for pulse
generators. I also want to get my loop area down to shorten the pulse.
Anyone have a SMT part that is available at Digikey or Mouser that
avalanches well?
Say 3$ a unit max? Only other requirement is a packaging that a 40
year old pair of eyes can handle.
I know about the $14 to 20$ Zetex/Diodes Inc parts.
Thanks,
Steve
generators. I also want to get my loop area down to shorten the pulse.
Anyone have a SMT part that is available at Digikey or Mouser that
avalanches well?
Say 3$ a unit max? Only other requirement is a packaging that a 40
year old pair of eyes can handle.
I know about the $14 to 20$ Zetex/Diodes Inc parts.
Thanks,
Steve
John Larkin
Oct 8, 2012, 10:59:32 AM10/8/12
to
On Mon, 8 Oct 2012 07:32:49 -0700 (PDT), Owen Roberts <o...@case.edu>
wrote:
What kind of pulse do you want to make? There are many ways to make
fast pulses these days. Avalanche is sort of hit-or-miss, involving
selecting transistors usually.
Modern epitaxial transistors don't avalanche well. Older
diffused-junction types sometimes do. The Zetexes are made in Russia,
presumably on an old fab line.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
wrote:
fast pulses these days. Avalanche is sort of hit-or-miss, involving
selecting transistors usually.
Modern epitaxial transistors don't avalanche well. Older
diffused-junction types sometimes do. The Zetexes are made in Russia,
presumably on an old fab line.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
dagmarg...@yahoo.com
Oct 8, 2012, 3:28:00 PM10/8/12
to
I tested a bunch of varieties. I'd have to dig through old notes.
I got pretty good yields, but today you might as well use ordinary
logic. It's faster, cheaper, easier, and makes better pulses. Unless
you're trying to make amps, that is.
--
Cheers,
James Arthur
Jim Thompson
Oct 8, 2012, 4:16:54 PM10/8/12
to
On Mon, 8 Oct 2012 12:28:00 -0700 (PDT), dagmarg...@yahoo.com
wrote:
>On Oct 8, 10:32 am, Owen Roberts <o...@case.edu> wrote:
>> Ok, I'm tired of hunting old mesa transistors on Ebay for pulse
>> generators. I also want to get my loop area down to shorten the pulse.
>> Anyone have a SMT part that is available at Digikey or Mouser that
>> avalanches well?
>> Say 3$ a unit max? Only other requirement is a packaging that a 40
>> year old pair of eyes can handle.
>>
>> I know about the $14 to 20$ Zetex/Diodes Inc parts.
>>
>> Thanks,
>>
>> Steve
>
>The old standard was 2n2369.
2N2369 is gold-doped and fast out of saturation, but I don't remember
it being used in avalanche mode.
>I've also used 2n2222, I think it was.
>I tested a bunch of varieties. I'd have to dig through old notes.
>
>I got pretty good yields, but today you might as well use ordinary
>logic. It's faster, cheaper, easier, and makes better pulses. Unless
>you're trying to make amps, that is.
...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
wrote:
>On Oct 8, 10:32 am, Owen Roberts <o...@case.edu> wrote:
>> Ok, I'm tired of hunting old mesa transistors on Ebay for pulse
>> generators. I also want to get my loop area down to shorten the pulse.
>> Anyone have a SMT part that is available at Digikey or Mouser that
>> avalanches well?
>> Say 3$ a unit max? Only other requirement is a packaging that a 40
>> year old pair of eyes can handle.
>>
>> I know about the $14 to 20$ Zetex/Diodes Inc parts.
>>
>> Thanks,
>>
>> Steve
>
>The old standard was 2n2369.
it being used in avalanche mode.
>I've also used 2n2222, I think it was.
>I tested a bunch of varieties. I'd have to dig through old notes.
>
>I got pretty good yields, but today you might as well use ordinary
>logic. It's faster, cheaper, easier, and makes better pulses. Unless
>you're trying to make amps, that is.
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
Jeroen
Oct 8, 2012, 4:42:46 PM10/8/12
to
lower voltage than a 2n2369. It's faster too.
Jeroen Belleman
Owen Roberts
Oct 8, 2012, 5:45:52 PM10/8/12
to
We have a little informal contest here on campus for fastest hand made
pulser under 20$. Right now I'm loosing to the EE Dept's tech by 50
pS.
He had time during the summer to hand select devices. Only avalanche
and SRD is allowed.
Why Else? Convenience....
I use pulsers like this in a ultrafast laser lab for testing
photodetectors. They drive 1.3 micron VECSELs with a little bias
network. Since the budget has to came out of my own pocket, I like
the simplicity. I can build them dead bug.
While no where near as fast as a femtosecond laser, its good enough
to see if the PD is still good, if the termination is correct etc.
Steve
pulser under 20$. Right now I'm loosing to the EE Dept's tech by 50
pS.
He had time during the summer to hand select devices. Only avalanche
and SRD is allowed.
Why Else? Convenience....
I use pulsers like this in a ultrafast laser lab for testing
photodetectors. They drive 1.3 micron VECSELs with a little bias
network. Since the budget has to came out of my own pocket, I like
the simplicity. I can build them dead bug.
While no where near as fast as a femtosecond laser, its good enough
to see if the PD is still good, if the termination is correct etc.
Steve
John Larkin
Oct 8, 2012, 6:23:09 PM10/8/12
to
On Mon, 8 Oct 2012 14:45:52 -0700 (PDT), Owen Roberts <o...@case.edu>
wrote:
>We have a little informal contest here on campus for fastest hand made
>pulser under 20$. Right now I'm loosing to the EE Dept's tech by 50
>pS.
How fast is he getting? How much voltage?
The fastest commercially available SRD that I know of is 30 or maybe
35 ps. They are a little hard to drive, slamming a bunch of current
into them before the stored charge is depleted.
Tunnel diodes, ca 1965 or so, had 25 ps risetimes.
There are ECL/CML gates with risetimes in the 12-15 ps range, way out
of the $20 budget.
Fast stuff is fun.
I did a little laser driver recently...
https://dl.dropbox.com/u/53724080/Circuits/Laser_Drivers/T165_5_c.jpg
https://dl.dropbox.com/u/53724080/Circuits/Laser_Drivers/T165_160_ps.JPG
and I can get 160 ps light pulses, if I tease the laser just right.
We've seen 50 ps light from a fancier driver and the right laser.
At higher current drives, we tend to see this:
https://dl.dropbox.com/u/53724080/Circuits/Laser_Drivers/T165_tail.JPG
where we think the tail is inherent to the laser.
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
wrote:
>We have a little informal contest here on campus for fastest hand made
>pulser under 20$. Right now I'm loosing to the EE Dept's tech by 50
>pS.
The fastest commercially available SRD that I know of is 30 or maybe
35 ps. They are a little hard to drive, slamming a bunch of current
into them before the stored charge is depleted.
Tunnel diodes, ca 1965 or so, had 25 ps risetimes.
There are ECL/CML gates with risetimes in the 12-15 ps range, way out
of the $20 budget.
Fast stuff is fun.
I did a little laser driver recently...
https://dl.dropbox.com/u/53724080/Circuits/Laser_Drivers/T165_5_c.jpg
https://dl.dropbox.com/u/53724080/Circuits/Laser_Drivers/T165_160_ps.JPG
and I can get 160 ps light pulses, if I tease the laser just right.
We've seen 50 ps light from a fancier driver and the right laser.
At higher current drives, we tend to see this:
https://dl.dropbox.com/u/53724080/Circuits/Laser_Drivers/T165_tail.JPG
where we think the tail is inherent to the laser.
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation
Phil Hobbs
Oct 8, 2012, 7:07:53 PM10/8/12
to
John Larkin wrote:
>
> On Mon, 8 Oct 2012 14:45:52 -0700 (PDT), Owen Roberts <o...@case.edu>
> wrote:
>
> >We have a little informal contest here on campus for fastest hand made
> >pulser under 20$. Right now I'm loosing to the EE Dept's tech by 50
> >pS.
>
> How fast is he getting? How much voltage?
>
> The fastest commercially available SRD that I know of is 30 or maybe
> 35 ps. They are a little hard to drive, slamming a bunch of current
> into them before the stored charge is depleted.
>
> Tunnel diodes, ca 1965 or so, had 25 ps risetimes.
>
But only if you get the 100-200 mA ones--they have like 200 pF of
>
> On Mon, 8 Oct 2012 14:45:52 -0700 (PDT), Owen Roberts <o...@case.edu>
> wrote:
>
> >We have a little informal contest here on campus for fastest hand made
> >pulser under 20$. Right now I'm loosing to the EE Dept's tech by 50
> >pS.
>
> How fast is he getting? How much voltage?
>
> The fastest commercially available SRD that I know of is 30 or maybe
> 35 ps. They are a little hard to drive, slamming a bunch of current
> into them before the stored charge is depleted.
>
> Tunnel diodes, ca 1965 or so, had 25 ps risetimes.
>
junction capacitance, so the 1 mA ones are depressingly slow.
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics
160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
Tim Williams
Oct 8, 2012, 7:18:02 PM10/8/12
to
I've always found 2N3904s work well, surprising for an epitaxial. Fast
and RF transistors tend to work as well. Jim Williams was fond of the
2N2369, as suggested; I've avalanched a, I think it was 2SA1016 (if I have
that wrong, it's a nearby number, with specs something like 20V, 100mA, fT
> 1GHz).
Tim
--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
"Owen Roberts" <o...@case.edu> wrote in message
news:f4a0f59f-cb03-4b0e...@x14g2000yqh.googlegroups.com...
and RF transistors tend to work as well. Jim Williams was fond of the
2N2369, as suggested; I've avalanched a, I think it was 2SA1016 (if I have
that wrong, it's a nearby number, with specs something like 20V, 100mA, fT
> 1GHz).
Tim
--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
"Owen Roberts" <o...@case.edu> wrote in message
news:f4a0f59f-cb03-4b0e...@x14g2000yqh.googlegroups.com...
Jamie
Oct 8, 2012, 8:13:45 PM10/8/12
to
15nS pulse, variable up to 250nS. It uses an inductor on the (+) input
as the time constant (500uh). Input generator set to match what a
Cmos version of the 555 timer can do on the raise and full time to
trigger an interval.
I wonder how a HEMT would do for you?
Jamie
aobo...@gmail.com
Sep 22, 2015, 8:56:10 PM9/22/15
to
在 2012年10月8日星期一 UTC-7下午1:42:46，Jeroen写道：
May I know the BFG541's avalanche switch's resistor parameters you are using? Charging resistor and base resistor value?
Thanks
Thanks
Jeroen Belleman
Sep 23, 2015, 3:56:19 AM9/23/15
to
resistor.
Jeroen Belleman
0 new messages