Are small signal npn transistors really so different from one another?

[John Popelish]

I'll try to do so later, but am off up the pub in a minute. Does
anyone know if it's possible to post an LTSpice schematic to a
newsgroup in a way that's legible to everyone?

Display the schematic on your monitor, hit the [print screen] key,
then edit paste into any graphic program. Then crop and save as a GIF
or PNG in two colors (or 16 if the schematic is in color). This
captures screen images from any Windows program.

 

[John Woodgate]

I read in sci.electronics.design that Winfield Hill
about 'Are small signal npn transistors really so different from one
another?', on Sat, 20 Dec 2003:

Paul Burridge wrote...

Indeed. In general for high-frequency amplifiers you want
to use Vce = 2V or more to reduce Ccb. Furthermore the
last stage needs more room for output-voltage swing.

I'll fight the big guy!

Win, I'm very surprised you didn't point out that wasting nearly half
the Vcc in drop across the emitter resistor can't be right in anybody's
book, especially yours.

Ve = 1 to 1.5 V is ample to make Ic reasonably independent of beta,
especially since in an RF amplifier, the base bias chain can be quite
low resistance.

It's also an interesting trick to set the first stage current at several
milliamps, because that stage is fed from a low impedance, so a low
input impedance isn't significant. For the succeeding stages, you need a
balancing act between input impedance (RL is low compared with the
collector resistor to get minimum shunting effect) and stage gain (RL
not so low that gmRL is reduced too much).

 

[mike]

Hi guys,

I just designed (in Spice) this chain of three c/e untuned amplifying
stages to get from 50uV in to 60mV out (at 40Mhz) using what I thought
were 2N3904 transistors. Optimised the biasing arrangements over
several hours to get maximum Vgain. Now I've realised I was using the
model for the 2N5769 instead by mistake. Upon 'plugging back in' the
2n3904 models into the schematic however, a fresh simulation showed
the ouput was now only 4uV! I can't believe this substitution would
result in such a deterioration of the amp's efficiency. I'm
particularly pissed off as I don't have any 2N5769s lying around to
use for the actual circuit; I'd been planning on using some of my
extensive stock of '3904s.
Can anyone verify that such a huge disparity in gain is indeed
possilbe? And the quickest way around the problem to get my 60mV back?

thanks.

p.

30 years ago, the 5769 was a damn fine transistor. Had excellent
switching performance and was the transistor of choice for low/medium
performance scope amplifier design. Remember, this was 30 years ago
when scopes still had transistors and medium performance was under 50
MHz system bandwidth.
It will run rings around a 3904 in most cases, but I am surprised that
your simulations show such MAJOR differences. Something is wrong in
the simulation.

mike

--
Return address is VALID.
Bunch of stuff For Sale and Wanted at the link below.
Toshiba & Compaq LiIon Batteries, Test Equipment
Honda CB-125S $800 in PDX
TEK Sampling Sweep Plugin and RM564
Tek 2465 $800, ham radio, 30pS pulser
Tektronix Concept Books, spot welding head...
http://www.geocities.com/SiliconValley/Monitor/4710/

 

[John Woodgate]

Something is wrong in
the simulation.

No, his d.c. conditions are for the birds.

 

[R.Legg]

I'll try to do so later, but am off up the pub in a minute. Does
anyone know if it's possible to post an LTSpice schematic to a
newsgroup in a way that's legible to everyone?

Select the LT schematic view by clicking it's window.

Select Tools - 'copy bitmap to clipboard'

open Paint, Irfanview, Labview or other imaging program.

Select Edit - 'paste'

Select File - 'save as'

chose jpg ir gif format to get a file size less than 30K.

RL

 

[mike]

No, his d.c. conditions are for the birds.

Whaddayamean NO???
Birds for dc conditions means, "something is wrong in the
simulation...specifically...maybe...the dc conditions...in the
simulation...which might be what is wrong in the simulation...!!!!!"
and !
and ;-)
mike

--
Return address is VALID.
Bunch of stuff For Sale and Wanted at the link below.
Toshiba & Compaq LiIon Batteries, Test Equipment
Honda CB-125S $800 in PDX
TEK Sampling Sweep Plugin and RM564
Tek 2465 $800, ham radio, 30pS pulser
Tektronix Concept Books, spot welding head...
http://www.geocities.com/SiliconValley/Monitor/4710/

 

[Paul Burridge]

Select the LT schematic view by clicking it's window.

Select Tools - 'copy bitmap to clipboard'

open Paint, Irfanview, Labview or other imaging program.

Select Edit - 'paste'

Select File - 'save as'

chose jpg ir gif format to get a file size less than 30K.

Oh bollocks. Thanks all the same to you and John P. for your advice,
but I've just won a bottle of Scotch and I'm half-way thru it. Hence
you're going to have to wait until 2morrow for me to post a considered
response to your excellent replies (Jeez., I can't believe I managed
to write all that perfectly in the state I'm in).

 

[Paul Burridge]

For a broadband amplifier at 40 MHz built by trial and error, I am
still amazed at an average stage voltage gain of 10, unless he got it
by finishing with an output stage that drives no external load.

It's not broadband. The first stage is tuned *before* the input.
BTW, the output stage drives a sub-minature moving coil meter with a
380 ohm static DC resistance.

 

[Genome]

Hi guys,

"I expect history will be kind to me, since I intend to write it."
-

Winston Churchill

Unfortunately you also appear to be dead. Ain't life a ****?
-. DNA

 

[Paul Burridge]

Winston Churchill

Unfortunately you also appear to be dead. Ain't life a ****?

Er, no. He came up with the quote shortly before he wrote the Nobel
Prize-winning 6 volume epic "The History of the English Speaking
Peoples"
HTH (dear boy).

 

[Winfield Hill]

Paul Burridge wrote...

Maybe transistors give best gain when driven close to saturation,
I don't know, but it would certainly account for how my optimising
the biasing has rendered a model change a disaster.

As we explain in AoE page 83, a common-emitter-stage gain is
g_m times the load resistance. The transconductance g_m is
proportional to current (page 81), and is therefore highest
when the stage load resistor is pulled all the way down to the
emitter. Hence the gain is maximum near saturation (page 83).

Thanks,
- Win

whill_at_picovolt-dot-com

 

[Genome]

Er, no. He came up with the quote shortly before he wrote the Nobel
Prize-winning 6 volume epic "The History of the English Speaking
Peoples"
HTH (dear boy).

Winston Churchill

So, he's still dead then?

DNA

 

[YD]

I'll try to do so later, but am off up the pub in a minute. Does
anyone know if it's possible to post an LTSpice schematic to a
newsgroup in a way that's legible to everyone?

Display the schematic on your monitor, hit the [print screen] key,
then edit paste into any graphic program. Then crop and save as a GIF
or PNG in two colors (or 16 if the schematic is in color). This
captures screen images from any Windows program.

That would make a binary and I'm not getting them since astraweb
terminated the free accounts. OTOH LTSpice's .asc files are pure text,
open it in notepad or somesuch, copy and paste into the post. Make
sure to include any models not in the standard library.

By the way, anyone know of any free binaries servers?

- YD.

 

[John Woodgate]

I read in sci.electronics.design that Winfield Hill
about 'Are small signal npn transistors really so different from one
another?', on Sat, 20 Dec 2003:

Paul Burridge wrote...

As we explain in AoE page 83, a common-emitter-stage gain is
g_m times the load resistance. The transconductance g_m is
proportional to current (page 81), and is therefore highest
when the stage load resistor is pulled all the way down to the
emitter. Hence the gain is maximum near saturation (page 83).

That's the highest *gain* condition, but it gives almost zero signal-
handling capacity, so it's not usually a lot of use. And it doesn't
really work (the gain hardly depends on the collector resistor value) if
the collector resistor value so determined is much higher than the input
impedance of the next stage.

Besides, you have to bias very carefully so as not to tip high-beta
devices into saturation, which may be why the 2N3904 version failed. The
report of electrode voltages suggests that.

 

[Paul Burridge]

As we explain in AoE page 83, a common-emitter-stage gain is
g_m times the load resistance. The transconductance g_m is
proportional to current (page 81), and is therefore highest
when the stage load resistor is pulled all the way down to the
emitter. Hence the gain is maximum near saturation (page 83).

Thanks, Win. That certainly figures. I'll check out the reference in a
jiffy.
Incidentally, is there a single identifiable parameter in the Spice
model of a transistor that would show up its susceptibility of going
into saturation and the criticality of correct biasing? Sorry if that
sounds like a load of bollocks but I'm still recovering from a
hangover.

 

[Paul Burridge]

Winston Churchill

So, he's still dead then?

At the time of writing, yes - if you insist on being pedantic. But his
spirit, humour and inspiration to us all live on...

 

[John Woodgate]

(in <[email protected]>) about 'Are
small signal npn transistors really so different from one another?', on
Sun, 21 Dec 2003:

So, he's still dead then?

One is, but the other one is still alive, AFAIK.

 

[Winfield Hill]

Paul Burridge wrote...

It's not broadband. The first stage is tuned *before* the
input. BTW, the output stage drives a sub-minature moving
coil meter with a 380 ohm static DC resistance.

And a rectifying diode?

Thanks,
- Win

whill_at_picovolt-dot-com

 

[Jim Thompson]

Thanks, Win. That certainly figures. I'll check out the reference in a
jiffy.
Incidentally, is there a single identifiable parameter in the Spice
model of a transistor that would show up its susceptibility of going
into saturation and the criticality of correct biasing? Sorry if that
sounds like a load of bollocks but I'm still recovering from a
hangover.

Bullocks ;-)

Please post your schematic and I'll write up an analysis, plus fix
your bias.

...Jim Thompson

 

[Jim Thompson]

Paul Burridge wrote...

As we explain in AoE page 83, a common-emitter-stage gain is
g_m times the load resistance. The transconductance g_m is
proportional to current (page 81), and is therefore highest
when the stage load resistor is pulled all the way down to the
emitter. Hence the gain is maximum near saturation (page 83).

Thanks,
- Win

whill_at_picovolt-dot-com

Voltage gain into a purely resistive load (no other transistors) is
simply the voltage drop of the load divided by (kT/q)... in other
words about 40X per volt.

...Jim Thompson