Searching \ for '[EE]: NPN Transistor with low switch on volta' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=npn+transistor+with
Search entire site for: 'NPN Transistor with low switch on volta'.

No exact or substring matches. trying for part
PICList Thread
'[EE]: NPN Transistor with low switch on voltage?'
2006\05\12@092954 by Edward Cooper

flavicon
face
Hi all

I'm finding it hard to search transistors by Vbe ON voltage,  
basically I need a NPN with a fairly high sensitivity, ie saturates  
at a small voltage on the base.

Any ideas, nothing to specialised though, just cheap and fairly  
available.

Thanks
Ed

2006\05\12@100243 by Spehro Pefhany

picon face
At 02:29 PM 5/12/2006 +0100, you wrote:
>Hi all
>
>I'm finding it hard to search transistors by Vbe ON voltage,
>basically I need a NPN with a fairly high sensitivity, ie saturates
>at a small voltage on the base.
>
>Any ideas, nothing to specialised though, just cheap and fairly
>available.
>
>Thanks
>Ed

That's because it's pretty much the same for *any* silicon BJT
at reasonable base currents (around 0.7V at room temperature).

>Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
spam_OUTspeffTakeThisOuTspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
->>Test equipment, parts OLED displys http://search.ebay.com/_W0QQsassZspeff


2006\05\12@100649 by olin piclist

face picon face
Edward Cooper wrote:
> I'm finding it hard to search transistors by Vbe ON voltage,

That is basically a diode and is mostly dictated by the semiconductor
material.  Doping levels do have some effect, but the rest is largely an
exponential equation.  Basically silicon diodes are going to have 500-700mV
drop for most "normal" operating uses.

If you really need less, you have to use a different semiconductor material,
like germanium.  Germanium transistors are pretty scarce these days, but
there are probably some specialty ones out there somewhere.  The volumes
will be low, so be prepared to pay several times the price of a similar
silicon transistor.

> basically I need a NPN with a fairly high sensitivity, ie saturates
> at a small voltage on the base.

This really smells like the wrong pre-supposed solution to a problem.  The
right answer is probably to step back and take another look at the problem
you are trying to solve two levels up.


******************************************************************
Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
consultant in 2004 program year.  http://www.embedinc.com/products

2006\05\12@101809 by Edward Cooper

flavicon
face
I wondered if it was something like that, I'll look into swapping my  
diode for a
schottky I really want to keep this simple though.

(Its a fan controller for a compact amp I've built and I'm aiming to  
get the fan
running at something near full speed at 1 v rms output - which is  
pretty loud)

Cheers

Ed

On 12 May 2006, at 15:13, Spehro Pefhany wrote:

{Quote hidden}

> --

2006\05\12@102047 by Edward Cooper

flavicon
face

On 12 May 2006, at 15:08, Olin Lathrop wrote:
> This really smells like the wrong pre-supposed solution to a  
> problem.  The
> right answer is probably to step back and take another look at the  
> problem
> you are trying to solve two levels up.

Probably is, I'll go thinking again - how would you switch a load on  
with the
presence of a 1v rms AC signal or greater?

Ed

2006\05\12@102651 by Rich Graziano

picon face
You need a high beta.

----- Original Message -----
From: "Spehro Pefhany" <speffspamKILLspaminterlog.com>
To: "Microcontroller discussion list - Public." <.....piclistKILLspamspam.....mit.edu>
Sent: Friday, May 12, 2006 10:13 AM
Subject: Re: [EE]: NPN Transistor with low switch on voltage?


{Quote hidden}

> --

2006\05\12@105840 by olin piclist

face picon face
Edward Cooper wrote:
> I wondered if it was something like that, I'll look into swapping my
> diode for a
> schottky I really want to keep this simple though.

Huh?  You originally asked about a transistor.  What kind of circuit do you
have where you can replace a NPN transistor with a Schottky diode?

> (Its a fan controller for a compact amp I've built and I'm aiming to
> get the fan
> running at something near full speed at 1 v rms output - which is
> pretty loud)

So what does that have to do with the B-E drop of a transistor, which is
what you originally asked about?  It sounds like you want low on voltage.
Why not a FET like IRLML2502, for example?


******************************************************************
Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
consultant in 2004 program year.  http://www.embedinc.com/products

2006\05\12@133506 by Vasile Surducan

face picon face
On 5/12/06, Rich Graziano <rgrazia1spamspam_OUTrochester.rr.com> wrote:
> You need a high beta.

Arguable. Beta or h21 defines the behaviour of the transistor in the
linear mode and not in the switching mode. This is the reason why a
switching transistor has usually a low measured beta. I've seen 2N2222
with beta between 10 and 30 working perfect, or this is a standard
value for beta for high power transistors running in linear mode at IC
>= 3A.
The Vbe voltage value (or the Ib current) requested for a switching
device is defined only by the manufacturing technology and by the
junction temperature.

Vasile

>
> {Original Message removed}

2006\05\12@134033 by Peter

picon face


On Fri, 12 May 2006, Edward Cooper wrote:

> I wondered if it was something like that, I'll look into swapping my diode
> for a
> schottky I really want to keep this simple though.
>
> (Its a fan controller for a compact amp I've built and I'm aiming to get the
> fan
> running at something near full speed at 1 v rms output - which is pretty
> loud)

How about two transistors with less than 50mV 'switch on voltage' ?:

         +--+----- +5V for fan
         |  +--+
        R2 FAN C2
         |  +--+
         |  C
       E-+-B  Q2 NPN
IN-C1+B     E
    R1 C    |
     +-+----+----- GND
      Q1 PNP

For a sine wave input the voltage on C2 will be 1/2 Vcc so you can use
12V to get 5V out.

Peter

2006\05\12@134308 by Spehro Pefhany

picon face
At 10:59 AM 5/12/2006 -0400, you wrote:
>Edward Cooper wrote:
> > I wondered if it was something like that, I'll look into swapping my
> > diode for a
> > schottky I really want to keep this simple though.
>
>Huh?  You originally asked about a transistor.  What kind of circuit do you
>have where you can replace a NPN transistor with a Schottky diode?
>
> > (Its a fan controller for a compact amp I've built and I'm aiming to
> > get the fan
> > running at something near full speed at 1 v rms output - which is
> > pretty loud)
>
>So what does that have to do with the B-E drop of a transistor, which is
>what you originally asked about?

Sounds like he needs a small audio transformer, some diodes (eg. a couple
of BAV99s), a film cap, a resistor, a zener and a logic-level MOSFET.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
@spam@speffKILLspamspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
->>Test equipment, parts OLED displys http://search.ebay.com/_W0QQsassZspeff


2006\05\12@151650 by Herbert Graf

flavicon
face
On Fri, 2006-05-12 at 14:29 +0100, Edward Cooper wrote:
> Hi all
>
> I'm finding it hard to search transistors by Vbe ON voltage,  
> basically I need a NPN with a fairly high sensitivity, ie saturates  
> at a small voltage on the base.
>
> Any ideas, nothing to specialised though, just cheap and fairly  
> available.

Since the forward voltage of the PN junction in an NPN transistor is
largely dependent on the material used, you're not going to be able to
find something much less then "normal" when sticking with silicon.

There are other options, germanium transistors used to exist, I don't
know how common they are these days.

If you're just switching something then low Vgs MOSFETs would make a
good solution.

TTYL

2006\05\12@211045 by Russell McMahon

face
flavicon
face
>> I'm finding it hard to search transistors by Vbe ON voltage,
>> basically I need a NPN with a fairly high sensitivity, ie saturates
>> at a small voltage on the base.

As Herbert says, bipolar silicon transistors all have essentially the
same turn on voltage.

Some conventional FETs have Vth below the turn on voltage for
bipolars.

There is a new process available (actually new technology) that allows
FETs to have arbitrarily low turn on voltages.

Also. You can bias a bipolar transistor so a small input voltage turns
it in.

Tell us how you are trying to do an unknown thing and we may not be
able to help.
Tell us what you want to achieve and we'll tell you how to achieve it.


       RM

2006\05\12@222057 by Edward Cooper

flavicon
face
>
> Tell us how you are trying to do an unknown thing and we may not be
> able to help.
> Tell us what you want to achieve and we'll tell you how to achieve it.

Hi Russell

I've sorted it now, changed how I was doing it, basically now a high  
gain
NPN after two diodes from the left and right channels, with it  
connected to the
base of a PNP(power), which controls the supply to the fan, with some  
caps and
resistors in there to smooth fan operation out.  Fan runs at varying  
speeds from
0.7v rms to 1.5v rms so thats quite nice.

I was just confused why I couldn't search by Vbe ON/Saturated  
voltages, but I
understand why now.

Cheers, have a great weekend all.

Ed

2006\05\12@222301 by Spehro Pefhany

picon face
At 12:38 PM 5/13/2006 +1200, you wrote:


>Tell us how you are trying to do an unknown thing and we may not be
>able to help.
>Tell us what you want to achieve and we'll tell you how to achieve it.
>         RM

I think what he wants to do is to take an audio signal (line level?)
and modulate a fan controller based on the expected power dissipation
in an audio amplifier, so that the fan tends to be quiet when the
music is quiet, and loud when the music is blasting.

Something like a precision rectifier (couple of op-amps) into a PIC ADC
and PWM output would fill the bill. The front end might not have to be
than complex, of course.

>Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
KILLspamspeffKILLspamspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
->>Test equipment, parts OLED displys http://search.ebay.com/_W0QQsassZspeff


2006\05\12@231208 by Gus Salavatore Calabrese

face picon face

Would not a fan based on heat sensing make more sense ?
( poorly positioned pun intended )
AGSC POD  Gus S Calabrese
000000000000000000000000000000000000000000

On 2006-May 12, at 8:20 PM, Edward Cooper wrote:

>
> Tell us how you are trying to do an unknown thing and we may not be
> able to help.
> Tell us what you want to achieve and we'll tell you how to achieve it.

Hi Russell

I've sorted it now, changed how I was doing it, basically now a high
gain
NPN after two diodes from the left and right channels, with it
connected to the
base of a PNP(power), which controls the supply to the fan, with some
caps and
resistors in there to smooth fan operation out.  Fan runs at varying
speeds from
0.7v rms to 1.5v rms so thats quite nice.

I was just confused why I couldn't search by Vbe ON/Saturated
voltages, but I
understand why now.

Cheers, have a great weekend all.

Ed

2006\05\13@152103 by Vasile Surducan

face picon face
On 5/13/06, Spehro Pefhany <RemoveMEspeffTakeThisOuTspaminterlog.com> wrote:
> At 12:38 PM 5/13/2006 +1200, you wrote:
>
>
> >Tell us how you are trying to do an unknown thing and we may not be
> >able to help.
> >Tell us what you want to achieve and we'll tell you how to achieve it.
> >         RM
>
> I think what he wants to do is to take an audio signal (line level?)
> and modulate a fan controller based on the expected power dissipation
> in an audio amplifier, so that the fan tends to be quiet when the
> music is quiet, and loud when the music is blasting.


This simple trick and needs one: transistor, termistor (Kohm range),
resistor, capacitor, ventilator.

Thermistor is sensing the radiator's temperature on which final
transistors are mounted. Transistor's load is a low current ventilator
connected from colector to Vcc, an RC from the transistor base to
ground and the thermistor from base to Vcc. Heating the termistor will
decrease his resistance so the transitor will be turned ON and the
ventilator will blow faster. It's a common scheme used in many
computer power supplies.

Vasile



{Quote hidden}

> -

2006\05\13@234636 by Rich Graziano

picon face
I find this one rather confusing.  The junction voltage is fixed by some
relationships given in quantum physics.  As far as I know (But what do I
know?) there are not a variety of devices that successfully avoid the
parameters that determine the junction potential restrictions.

It would seem to me that any switching application that requires switching
at a very low threshold, say in milivolts or microvolts, could be achieved
by selecting the appropriate gain (Beta) for the device; Often a Darlington
arrangement for small currents or a fet differential to a reference, or
several other design approaches.

The only caveat I see in the extremely low level operation is the noise,
linearity and hysteresis over the range.  These also can be addressed in the
design error budget and essential performance spec.  Why would someone be
looking for a different junction voltage? If indeed that is what is being
sought.

I am at a loss to understand why this is a junction problem.


{Original Message removed}

2006\05\14@031205 by Vasile Surducan

face picon face
On 5/14/06, Rich Graziano <TakeThisOuTrgrazia1EraseMEspamspam_OUTrochester.rr.com> wrote:
> I find this one rather confusing.  The junction voltage is fixed by some
> relationships given in quantum physics.  As far as I know (But what do I
> know?) there are not a variety of devices that successfully avoid the
> parameters that determine the junction potential restrictions.
>
> It would seem to me that any switching application that requires switching
> at a very low threshold, say in milivolts or microvolts,

At this low levels the only way for secure "switching" is the
operational amplifier used as comparator with positive hysteresis
(trigger schmidt).

where "switching" = pure non linear application (fast transient
between blocked and saturated state - and viceversa- of a transistor).

> could be achieved
> by selecting the appropriate gain (Beta) for the device; Often a Darlington
> arrangement for small currents or a fet differential to a reference, or
> several other design approaches.

Sure if you're thinking to drive a load proportionaly (a variation of
Ib or Vbe is causing a variation on load voltage or current).  A
darlinghton is not suited for "switching", (because is incapable of
fast response, it has important leackage current) and that's the ideea
why you can't say almost nothing about beta in switching applications.

On the other hand a standard darlington (2x PNP ot 2x NPN) require 2xVbe so
if he want to sense 100mV there is no way just with one transistor...:)
Sure you can design structures with NPN-PNP pairs where the Vbe is
still 0.6V-0.7V and h21e goes up to 600 or more.


>
> The only caveat I see in the extremely low level operation is the noise,

and temperature dependence which cause instability,

> linearity and hysteresis over the range.

... you can't talk about linearity when switch On-Off, just when drive
proportionally the load



 These also can be addressed in the
> design error budget and essential performance spec.  Why would someone be
> looking for a different junction voltage?

because it doesn't know the way how to solve the problem, so it looks
for a germanium transistor instead of a comparator or something else

> If indeed that is what is being
> sought.
>
> I am at a loss to understand why this is a junction problem.

Of course is not.This is hapenning when the question is asked wrong or
the originator does not now exactly what he intend to do and let the
answerers to guess.

Vasile

>
>
> {Original Message removed}

2006\05\14@082138 by Rich Graziano

picon face
Thank you for your kind reply, Vasile. Yes, of course I agree with what you
say, if the problem is one of a switching application. These are simple
academic facts.  I could not tell from what I had seen, exactly what problem
it was that need to be solved, and why the junction voltage would be an
issue. I could not tell, perhaps, due to my own stupidity, if the question
was one of measurement or one of switching. If it was simply a choice
between silicon or germanium it would seem pointless to bring it up.

Furthermore, in a switching application it is wise to avoid driving the
device into saturation because of the junction capacitance and longer
switching times that result; if speed is a factor in the design requirement.

Thank you again for your kind and somewhat informative reply.  It seems from
your response that the question was one of switching. Can you also satisfy
my curiosity as to what the actual application is?  What is to be switched?
Is there some issue about the junction voltage?  Speed? I can admit to
sometimes being ignorant or even stupid, but I must admit to always to being
curious.
Respectfully,
Rich

{Original Message removed}

2006\05\14@141139 by Vasile Surducan

face picon face
On 5/14/06, Rich Graziano <RemoveMErgrazia1spamTakeThisOuTrochester.rr.com> wrote:
> Thank you for your kind reply, Vasile. Yes, of course I agree with what you
> say, if the problem is one of a switching application. These are simple
> academic facts.  I could not tell from what I had seen, exactly what problem
> it was that need to be solved, and why the junction voltage would be an
> issue. I could not tell, perhaps, due to my own stupidity, if the question
> was one of measurement or one of switching.

Have no ideea. :)


If it was simply a choice
> between silicon or germanium it would seem pointless to bring it up.
>
> Furthermore, in a switching application it is wise to avoid driving the
> device into saturation because of the junction capacitance and longer
> switching times that result; if speed is a factor in the design requirement.

This is a good point indeed. But unfortunately in practice you can't
avoid saturation only by adding a resonant circuit in parallel with
the load (changing from switching to pseudo-linear). This is possible
only with inductive loads and not everywhere. In most of the switching
schematics, as high is pomped the base current, as fast will be the
transient OFF-ON (assuming everything is ok with timings and load).
For faster switching from ON to OFF the base is reversed polarised
(the simplest tehnique is to use a small volatge zenner or two series
diodes in parallel with a capacitor in the base circuit; when driving
ON the capacitor is charged, when driving OFF -usually with 0V- the
voltage across the capacitor is reversed on BE junction allowing a
faster turn off).



>
> Thank you again for your kind and somewhat informative reply.  It seems from
> your response that the question was one of switching. Can you also satisfy
> my curiosity as to what the actual application is?  What is to be switched?
> Is there some issue about the junction voltage?  Speed? I can admit to
> sometimes being ignorant or even stupid, but I must admit to always to being
> curious.

I have no clue, fortunately I didn't ask the question.
:)

greetings,
Vasile

> Respectfully,
> Rich
>
> {Original Message removed}

2006\05\15@002100 by Rich Graziano

picon face
Of course when one speaks of the general case it may not reflect the
specifics in a unique design.  In the general case junction capacitance is
problematic for high speed switching and operation in the linear region
mitigates the worst of that condition. You are correct if you are saying
that the load must also be considered in the design and may indeed alter the
design parameters and the circuit configuration. In fact, I usually begin by
defining the load characteristics.  This is all well and good, but I am
still curious about the original question posted, because I have done some
work in the development of semiconductor devices in the lab.

Regards,
Rich

{Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2006 , 2007 only
- Today
- New search...