High Vbr(ebo) transistors?

C

Chris Carlen

Guest
Greetings:

I came up with a real simple way to clamp an op-amp integrator that you
folks probably tried already. For clamping the negative swing for
instance, connect NPN emitter to opamp output, base to the desired clamp
voltage +0.6V (so to clamp at -10V, base goes to -9.4V through a
resistor). Collector goes to the opamp inverting input.

When the integrator swings to -10V, the NPN turns on, bypassing current
around the integrating capacitor preventing it from charging further.

Do a similar thing with a PNP to clamp the positive swing. It works!

The trouble is, for an integrator that can swing (after clamping) over a
fairly wide range, say +-10V, then the EB junctions of the transistors
get reverse biased beyond their ratings.

Should I try to find transistors with higher Vbr(ebo) ratings, or find
another integrator limiting method? (Please don't suggest another
specific method yet!)


Thanks for input.



Good day!




--
_______________________________________________________________________
Christopher R. Carlen
Principal Laser/Optical Technologist
Sandia National Laboratories CA USA
crcarleRemoveThis@BOGUSsandia.gov
NOTE, delete texts: "RemoveThis" and "BOGUS" from email address to reply.
 
Chris Carlen wrote:

Greetings:

I came up with a real simple way to clamp an op-amp integrator that you
folks probably tried already. For clamping the negative swing for
instance, connect NPN emitter to opamp output, base to the desired clamp
voltage +0.6V (so to clamp at -10V, base goes to -9.4V through a
resistor). Collector goes to the opamp inverting input.

When the integrator swings to -10V, the NPN turns on, bypassing current
around the integrating capacitor preventing it from charging further.

Do a similar thing with a PNP to clamp the positive swing. It works!

The trouble is, for an integrator that can swing (after clamping) over a
fairly wide range, say +-10V, then the EB junctions of the transistors
get reverse biased beyond their ratings.

Should I try to find transistors with higher Vbr(ebo) ratings, or find
another integrator limiting method? (Please don't suggest another
specific method yet!)

Thanks for input.

Good day!
Click to expand...
The base emitter breakdown is an avalanche effect IIRC. Usually occurs around
5-7 V I think.

I think there's likely an inherent problem here.

Graham
 
Hello Chris,

Should I try to find transistors with higher Vbr(ebo) ratings, ...
Click to expand...
That might win you a Nobel prize. Or a patent which you could then trade
in for your own tropical island with a landing strip for your new Lear Jet.

...or find another integrator limiting method?
Click to expand...
I am afraid so.

... (Please don't suggest another specific method yet!)
Click to expand...
Now that is like showing a piece of sausage to our dogs and then putting
it away again...

Regards, Joerg

http://www.analogconsultants.com
 
Joerg wrote:
Hello Chris,

Should I try to find transistors with higher Vbr(ebo) ratings, ...

That might win you a Nobel prize. Or a patent which you could then trade
in for your own tropical island with a landing strip for your new Lear Jet.

...or find another integrator limiting method?

I am afraid so.
Click to expand...
I get the picture.

... (Please don't suggest another specific method yet!)

Now that is like showing a piece of sausage to our dogs and then putting
it away again...
Click to expand...
I have a picture of how to do it with a pair of opamps and MOSFETs.
I'll tinker with that a while, and see if I come up with any other
ideas, then post something someday opening the door to
critiques/enhancements/alternatives.


Thanks for the input.


Good day!


--
_______________________________________________________________________
Christopher R. Carlen
Principal Laser/Optical Technologist
Sandia National Laboratories CA USA
crcarleRemoveThis@BOGUSsandia.gov
NOTE, delete texts: "RemoveThis" and "BOGUS" from email address to reply.
 
Hello Chris,

I have a picture of how to do it with a pair of opamps and MOSFETs. I'll
tinker with that a while, and see if I come up with any other ideas,
then post something someday opening the door to
critiques/enhancements/alternatives.
Click to expand...
FETs would give you more voltage range. Most small ones are rated for
+/-20V gate swings. But their Vth varies a bit from lot to lot.

Or use a differential approach. For really precise clamp points you can
still get arrays, mostly from JPN mfgs nowadays.

Regards, Joerg

http://www.analogconsultants.com
 
In article <d6ifus0kr1@news3.newsguy.com>,
Chris Carlen <crcarleRemoveThis@BOGUSsandia.gov> wrote:
Greetings:

I came up with a real simple way to clamp an op-amp integrator that you
folks probably tried already. For clamping the negative swing for
instance, connect NPN emitter to opamp output, base to the desired clamp
voltage +0.6V (so to clamp at -10V, base goes to -9.4V through a
resistor). Collector goes to the opamp inverting input.
Click to expand...

How about:

----+------ Op amp inverting input
GND !
! !
\ !
/ !
\ !
/ ! C
! !/
+----! NPN
! !\ Q2
\ ! E
/ !
\ !
/ !
! ! E
! !/
Vlimit --!<-+----! NPN
D1 !\ Q1
! C
----------- Swing being clamped

Q1 is being run in inverted mode so its gain sucks, and its "collector"
breakdown will be about 5V but its "emitter" breakdown will be high.

Q2 prevents the "collector" of Q1 from seeing much voltage.

D1 is just for fun to partly compensate the Q1 "e-b" drop away.

--
--
kensmith@rahul.net forging knowledge
 
Ken Smith wrote:
In article <d6ifus0kr1@news3.newsguy.com>,
Chris Carlen <crcarleRemoveThis@BOGUSsandia.gov> wrote:

Greetings:

I came up with a real simple way to clamp an op-amp integrator that you
folks probably tried already. For clamping the negative swing for
instance, connect NPN emitter to opamp output, base to the desired clamp
voltage +0.6V (so to clamp at -10V, base goes to -9.4V through a
resistor). Collector goes to the opamp inverting input.



How about:

----+------ Op amp inverting input
GND !
! !
\ !
/ !
\ !
/ ! C
! !/
+----! NPN
! !\ Q2
\ ! E
/ !
\ !
/ !
! ! E
! !/
Vlimit --!<-+----! NPN
D1 !\ Q1
! C
----------- Swing being clamped

Q1 is being run in inverted mode so its gain sucks, and its "collector"
breakdown will be about 5V but its "emitter" breakdown will be high.

Q2 prevents the "collector" of Q1 from seeing much voltage.

D1 is just for fun to partly compensate the Q1 "e-b" drop away.

Actually, Zetek makes some transistors with rather high inverse beta.
Click to expand...
 
dagmargoodboat@yahoo.com wrote:
Robert Baer wrote:
Actually, Zetek makes some transistors with rather high inverse
beta.

Yes, they do. The FMMT618, for example, which has a reverse beta
higher than a lot of transistors' forward beta.

This simplified circuit, however, does the original job with
just one extra component:


to summing node
|
|
|/
-Vlimit ---R1---+---| NPN
+ Vbe | |>. Q1
- |
D1 ^ |
'------+-------- opamp output


Cheers,
James Arthur
Click to expand...
But this forbids a positive integrator output. Ideally we can produce
an arbitrary (but more likely an arbitrary but symmetrical) clamp
voltage about zero.


Good day!



--
_____________________
Christopher R. Carlen
crobc@bogus-remove-me.sbcglobal.net
SuSE 9.1 Linux 2.6.5
 
On Sun, 22 May 2005 18:15:51 -0700, Chris Carlen
<crobc@BOGUSFIELD.sbcglobal.net> wrote:

dagmargoodboat@yahoo.com wrote:
Robert Baer wrote:
Actually, Zetek makes some transistors with rather high inverse
beta.

Yes, they do. The FMMT618, for example, which has a reverse beta
higher than a lot of transistors' forward beta.

This simplified circuit, however, does the original job with
just one extra component:


to summing node
|
|
|/
-Vlimit ---R1---+---| NPN
+ Vbe | |>. Q1
- |
D1 ^ |
'------+-------- opamp output


Cheers,
James Arthur


But this forbids a positive integrator output. Ideally we can produce
an arbitrary (but more likely an arbitrary but symmetrical) clamp
voltage about zero.


Good day!
Click to expand...
Just add one more diode and you'll be parasitic action free...

to summing node
|
V D2
|
|
|
|/
-Vlimit ---R1---+---| NPN
+ Vbe | |>. Q1
- |
D1 ^ |
'------+-------- opamp output

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| E-mail Address at Website Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.
 
On Sun, 22 May 2005 18:15:51 -0700, Chris Carlen wrote:

dagmargoodboat@yahoo.com wrote:
Robert Baer wrote:
Actually, Zetek makes some transistors with rather high inverse
beta.

Yes, they do. The FMMT618, for example, which has a reverse beta
higher than a lot of transistors' forward beta.

This simplified circuit, however, does the original job with
just one extra component:


to summing node
|
|
|/
-Vlimit ---R1---+---| NPN
+ Vbe | |>. Q1
- |
D1 ^ |
'------+-------- opamp output


Cheers,
James Arthur


But this forbids a positive integrator output. Ideally we can produce
an arbitrary (but more likely an arbitrary but symmetrical) clamp
voltage about zero.
Click to expand...
Well, if all you want to do is clamp to an arbitrary, but widely toleranced
voltage, these guys come in a bipolar version:
http://www.vishay.com/diodes/protection-tvs-esd/trans-zorb/

Cheers!
Rich
 
Rich Grise wrote:
On Sun, 22 May 2005 18:15:51 -0700, Chris Carlen wrote:

dagmargoodboat@yahoo.com wrote:
Robert Baer wrote:
Actually, Zetek makes some transistors with rather high inverse
beta.

Yes, they do. The FMMT618, for example, which has a reverse beta
higher than a lot of transistors' forward beta.

This simplified circuit, however, does the original job with
just one extra component:


to summing node
|
|
|/
-Vlimit ---R1---+---| NPN
+ Vbe | |>. Q1
- |
D1 ^ |
'------+-------- opamp output


Cheers,
James Arthur

But this forbids a positive integrator output. Ideally we can produce
an arbitrary (but more likely an arbitrary but symmetrical) clamp
voltage about zero.

Well, if all you want to do is clamp to an arbitrary, but widely toleranced
voltage, these guys come in a bipolar version:
http://www.vishay.com/diodes/protection-tvs-esd/trans-zorb/
Click to expand...
Yes, zeners are an option. At this point I am going to use Zeners
because an adjustable option is added complexity I can't afford space
for on the board. A later daughterboard add on with a little uC for
thermal management will perhaps also include potentiometer settable
current limits.

But Rich, Transorbs are a little hefty for this sort of application.

The Vishay BZX55-C series has 2% tolerances available in stock from
Newark. I think a small signal zener is a better choice here.


Good day!



--
_____________________
Christopher R. Carlen
crobc@bogus-remove-me.sbcglobal.net
SuSE 9.1 Linux 2.6.5
 
"Chris Carlen" <crobc@BOGUSFIELD.sbcglobal.net> a écrit dans le message de
news:d6taut02unm@news3.newsguy.com...
Rich Grise wrote:
On Sun, 22 May 2005 18:15:51 -0700, Chris Carlen wrote:

dagmargoodboat@yahoo.com wrote:
Robert Baer wrote:
Actually, Zetek makes some transistors with rather high inverse
beta.

Yes, they do. The FMMT618, for example, which has a reverse beta
higher than a lot of transistors' forward beta.

This simplified circuit, however, does the original job with
just one extra component:


to summing node
|
|
|/
-Vlimit ---R1---+---| NPN
+ Vbe | |>. Q1
- |
D1 ^ |
'------+-------- opamp output


Cheers,
James Arthur

But this forbids a positive integrator output. Ideally we can produce
an arbitrary (but more likely an arbitrary but symmetrical) clamp
voltage about zero.

Well, if all you want to do is clamp to an arbitrary, but widely
toleranced
voltage, these guys come in a bipolar version:
http://www.vishay.com/diodes/protection-tvs-esd/trans-zorb/

Yes, zeners are an option. At this point I am going to use Zeners
because an adjustable option is added complexity I can't afford space
for on the board. A later daughterboard add on with a little uC for
thermal management will perhaps also include potentiometer settable
current limits.

But Rich, Transorbs are a little hefty for this sort of application.

The Vishay BZX55-C series has 2% tolerances available in stock from
Newark. I think a small signal zener is a better choice here.
Click to expand...
Have a look at AD8036-37 + fixed gain.

--
Thanks,
Fred.
 
Hello Chris,

Yes, zeners are an option. At this point I am going to use Zeners
because an adjustable option is added complexity I can't afford space
for on the board. A later daughterboard add on with a little uC for
thermal management will perhaps also include potentiometer settable
current limits.

But Rich, Transorbs are a little hefty for this sort of application.

The Vishay BZX55-C series has 2% tolerances available in stock from
Newark. I think a small signal zener is a better choice here.
Click to expand...
Over the years I had to design out a lot of zeners for clients.
Tolerances became unbearable. Other times their data sheets appeared to
be a bit on the optimistic side, at least for high volume production.
Mostly I use things like the TLV431 instead, maybe you could concoct
something around two of these.

Regards, Joerg

http://www.analogconsultants.com
 
Fred Bartoli wrote:
Have a look at AD8036-37 + fixed gain.
Click to expand...
Yes, I did give that a look. Awfully high current consumption,
obviously since it is far wider band than I need. But the main spoiler
is low voltage range.

It's too bad there are no limiting amps in lower power and "GP"
implementations.

Good day!



--
_____________________
Christopher R. Carlen
crobc@bogus-remove-me.sbcglobal.net
SuSE 9.1 Linux 2.6.5
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top