>But you can also apply local resistive feedback from the >collector to the baseSure. If you apply that to the bias divider fine, it will further stabilise the stage but will lower input Z. It may even clip off the top of the output if you really want the stability. The lower the Z of the bias the better Re works. I could put you together a stage that has near infinite voltage gain, but has one ohm input Z and 32 megohn output Z. Pretty much useless. (PRETTY MUCH, THERE ARE TIMES...) In anything near that you don't have much real gain. (if any) It can be useful but if you get anywhere near that then you want to use a common base stage. In fact a real old time cascode stage would really do it. How much real gain does a cascode have ? It has voltage gain coming out of the cracks in the sky, but real gain is a whole different thing. Conversely take the typical audio amp. At the output stages with just the drivers and outputs of course there is gain. But it is current gain, in fact it inherently has negative voltage gain. But there is a stage before that which does the voltage amplification and the output stage which is only current amplification serves to allow the voltage amplifier stage to run into a lighter (higher Z) load. I know I responded to you and you probably know all about this, but I think you forgot to mention a thing or two. Most of this is for the OP. What I said about what you said is to find out if you agree and perhaps start a much better argument. (there is an idea I aim to contact you with, but I gotta write it)
Amplification factor for common emitter amplifier
Started by ●August 8, 2019
Reply by ●August 14, 20192019-08-14
Reply by ●August 14, 20192019-08-14
"Steve Wilson" wrote in message news:XnsAAAB3EB324456idtokenpost@69.16.179.22... John Larkin <jlarkin@highlandSNIPMEtechnology.com> wrote:> On Tue, 13 Aug 2019 20:14:05 GMT, Steve Wilson <no@spam.com> wrote: > >>John Larkin <jlarkin@highlandSNIPMEtechnology.com> wrote: >> >>> The transconductance Gm is 1/(.025/Ic) = 40 * Ic, and the voltage gain >>> is Gm * Rload. That magically collapses to Vgain = 40 times the DC >>> voltage drop in the collector resistor. >> >>> Transistors are actually pretty simple. >> >>Here are two circuits with identical operating points. Your formula sets >>the gain at 40 * 5 = 40. > > You might ask someone to check that multiply for you.Already posted the fix. Also included Legg's formula Vg = Zc/Ze, which you ignored. Here's my post: Steve Wilson <no@spam.com> wrote:> Here are two circuits with identical operating points. Your formula sets > the gain at 40 * 5 = 40.40 * 5 = 200 Legg is right. Vg = Zc / Ze>>Circuit A has a gain of 0.535. Circuit B has a gain of 176.7.>>Your equation is wrong.> George specified> (assuming no/small resistance from emitter to ground.)> Why did you snip that?You have to include the impedance in the emitter.> 176 is close to 200. A little Early effect will drop the gain some > from an ideal transistor. The 0.025 factor isn't exact either. And the > estimate is of course small-signal.>Your formula is useless and misleading.Ahmmm....>You normally never run a transistor >with the emitter grounded. You need some way to stabilize the bias.Sure, the bias needs stabalised...many designs use an emitter resister bypassed with a big cap....taking the resister out of the game. Some designs use a grounded emitter with a potential divider from collector to base and ground, fixing the output voltage referred to as a Vbe multiplier>Also, >there is usually some load on the collector, which you ignore. You could >also have transformer coupling or a resonant circuit in the collector, >which would produce a low dc resistance and consequently low voltage drop. >This would produce low gain in your formula.Sure... What's your point?>Legg's formula includes the impedance on both the collector and emitter, >and includes loading due to the following stage and bypassing on the >emitter.>Legg's formula can also consider the DC case by omitting the load impedance >and emitter bypass. So you can get both the DC gain and high frequency >gain.>Your equation is only for DC gain, and ignores the resistance in the >emitter. It does not include the load impedance on the collector or >degeneration in the emitter. It is useless for real-world circuits.It's a very well known, useful approximation for the maximum gain with a resistive load. That is, the actual gain will always be less. One can easily add in the effect of emitter resistance by computing Re' = re + Re,, re=1/(40.IC) One can also include the effect of base resistance by adding rbb'/hafe to Re' With a current source load it becomes Va/vt as the maximum voltage, where Va is the early voltage. I have a tutorial on this here: http://www.kevinaylward.co.uk/ee/bipolardesign1/bipolardesign1.xht Pretty much all hand calculations are of limited value in the real world. However, today, Spice is the only realistic way to design reliable multi transistor circuits, well for asics anyway... I'm designing fairly complex analog asics with 10k transistor system level blocks, and I would say out of millions of simulations, I do hand calculations as often as number as the fingers on that hand. Its just not worth the agro to piss about when there are computers. Even a diode resister circuit requires some effort to solve... Id = Vt/R.W(is.R/vt.exp(Vs/Vt).... Where W is the Lambert W function... http://www.kevinaylward.co.uk/ee/widlarlambert/widlarlambert.xht -- Kevin Aylward http://www.anasoft.co.uk - SuperSpice http://www.kevinaylward.co.uk/ee/index.html
Reply by ●August 14, 20192019-08-14
On Wednesday, August 14, 2019 at 2:52:36 PM UTC-4, piglet wrote:> On 14/08/2019 4:24 pm, George Herold wrote: > > On Wednesday, August 14, 2019 at 7:46:32 AM UTC-4, tabb...@gmail.com wrote: > >> On Wednesday, 14 August 2019 11:09:55 UTC+1, Steve Wilson wrote: > >> > >>> Your formula is useless and misleading. You normally never run a transistor > >>> with the emitter grounded. > >> > >> odd thing to say > >> > >>> You need some way to stabilize the bias. Also, > >> > >> There's more than one way to do that. And despite the myths, suicide bias IS used in commercial circuits. > >> > >> > >> NT > > > > What's suicide bias? I went searching for biploar transistor and suicide bias... but got a lot of mental health sites. (maybe lithium doping would help. :^) > > > > GH > > > > If want it to be linear but emitter well and truly grounded one can > always de-suicide the bias by using feedback resistor to base from > collector. For max AC gain decouple it midway with a capacitor to ground. > > pigletpiglet, :^) I've never done a suicide bias... mostly I've got some opamp or digital output driving a transistor. (with neg. feedback somewhere.) So I drew a resistor from collector to base, do I still keep the R from the power rail to base? (or just get rid of that... as long as the load has a resistive part, current is going to flow.) And then split that resistor in the middle with a cap to ground... 'cause that's going to be too much bias (on average). I'm much better with pictures, words confuse me at times. George h.
Reply by ●August 15, 20192019-08-15
On 15/08/2019 1:27 am, George Herold wrote:> On Wednesday, August 14, 2019 at 2:52:36 PM UTC-4, piglet wrote: >> On 14/08/2019 4:24 pm, George Herold wrote: >>> On Wednesday, August 14, 2019 at 7:46:32 AM UTC-4, tabb...@gmail.com wrote: >>>> On Wednesday, 14 August 2019 11:09:55 UTC+1, Steve Wilson wrote: >>>> >>>>> Your formula is useless and misleading. You normally never run a transistor >>>>> with the emitter grounded. >>>> >>>> odd thing to say >>>> >>>>> You need some way to stabilize the bias. Also, >>>> >>>> There's more than one way to do that. And despite the myths, suicide bias IS used in commercial circuits. >>>> >>>> >>>> NT >>> >>> What's suicide bias? I went searching for biploar transistor and suicide bias... but got a lot of mental health sites. (maybe lithium doping would help. :^) >>> >>> GH >>> >> >> If want it to be linear but emitter well and truly grounded one can >> always de-suicide the bias by using feedback resistor to base from >> collector. For max AC gain decouple it midway with a capacitor to ground. >> >> piglet > > piglet, :^) I've never done a suicide bias... mostly I've got > some opamp or digital output driving a transistor. (with neg. > feedback somewhere.) > > So I drew a resistor from collector to base, do I still keep the R > from the power rail to base? (or just get rid of that... > as long as the load has a resistive part, current is going to flow.) > And then split that resistor in the middle with a cap to ground... > 'cause that's going to be too much bias (on average). > > I'm much better with pictures, words confuse me at times. > > George h. > >Here is a picture of simple self bias. It is negative feedback, broken at AC in the lower sketch. <https://www.dropbox.com/s/ze0vo98nut8gwqc/_SelfbiasedQ.jpg?raw=1> piglet
Reply by ●August 15, 20192019-08-15
On 2019-08-14, George Herold <gherold@teachspin.com> wrote:> On Wednesday, August 14, 2019 at 7:46:32 AM UTC-4, tabb...@gmail.com wrote: >> On Wednesday, 14 August 2019 11:09:55 UTC+1, Steve Wilson wrote: >> >> > Your formula is useless and misleading. You normally never run a transistor >> > with the emitter grounded. >> >> odd thing to say >> >> > You need some way to stabilize the bias. Also, >> >> There's more than one way to do that. And despite the myths, suicide bias IS used in commercial circuits. >> >> >> NT > > What's suicide bias? I went searching for biploar transistor and suicide bias... but got a lot of mental health sites. (maybe lithium doping would help. :^)It's been a term of art here (in sci.electronics) for as long as I can remeber. Basically the the base is connected to some sort of current source, and the expression for emitter current is dominted by the beta. --+-+--------+-----+-+-------+-- VCC | | | | | R | | R | R +-+ | R +-+ R | R | R | R | | c | c | c | c +b +b +b +b | e | e e e R | R | | | | | | | ----+--------+-----+-+-----+-+-- GND Q1 Q2 Q3 Q4 Q1, Q2, Q3 suicide bias, Q4 maybe not -- When I tried casting out nines I made a hash of it.
Reply by ●August 15, 20192019-08-15
On Thursday, August 15, 2019 at 6:31:22 AM UTC-4, Jasen Betts wrote:> On 2019-08-14, George Herold <gherold@teachspin.com> wrote: > > On Wednesday, August 14, 2019 at 7:46:32 AM UTC-4, tabb...@gmail.com wrote: > >> On Wednesday, 14 August 2019 11:09:55 UTC+1, Steve Wilson wrote: > >> > >> > Your formula is useless and misleading. You normally never run a transistor > >> > with the emitter grounded. > >> > >> odd thing to say > >> > >> > You need some way to stabilize the bias. Also, > >> > >> There's more than one way to do that. And despite the myths, suicide bias IS used in commercial circuits. > >> > >> > >> NT > > > > What's suicide bias? I went searching for biploar transistor and suicide bias... but got a lot of mental health sites. (maybe lithium doping would help. :^) > > It's been a term of art here (in sci.electronics) for as long as I can remeber. > > Basically the the base is connected to some sort of current source, and the > expression for emitter current is dominted by the beta. > > --+-+--------+-----+-+-------+-- VCC > | | | | | R > | | R | R +-+ > | R +-+ R | R | > R | R | | c | c > | c | c +b +b > +b +b | e | e > e e R | R | > | | | | | | > ----+--------+-----+-+-----+-+-- GND > Q1 Q2 Q3 Q4 > > Q1, Q2, Q3 suicide bias, Q4 maybe not > > -- > When I tried casting out nines I made a hash of it.Thanks, Isn't Q3 the 'first circuit' one learns to bias a common emitter? Well, it was the first circuit I learned. GH
Reply by ●August 15, 20192019-08-15
On Thursday, August 15, 2019 at 4:49:11 AM UTC-4, piglet wrote:> On 15/08/2019 1:27 am, George Herold wrote: > > On Wednesday, August 14, 2019 at 2:52:36 PM UTC-4, piglet wrote: > >> On 14/08/2019 4:24 pm, George Herold wrote: > >>> On Wednesday, August 14, 2019 at 7:46:32 AM UTC-4, tabb...@gmail.com wrote: > >>>> On Wednesday, 14 August 2019 11:09:55 UTC+1, Steve Wilson wrote: > >>>> > >>>>> Your formula is useless and misleading. You normally never run a transistor > >>>>> with the emitter grounded. > >>>> > >>>> odd thing to say > >>>> > >>>>> You need some way to stabilize the bias. Also, > >>>> > >>>> There's more than one way to do that. And despite the myths, suicide bias IS used in commercial circuits. > >>>> > >>>> > >>>> NT > >>> > >>> What's suicide bias? I went searching for biploar transistor and suicide bias... but got a lot of mental health sites. (maybe lithium doping would help. :^) > >>> > >>> GH > >>> > >> > >> If want it to be linear but emitter well and truly grounded one can > >> always de-suicide the bias by using feedback resistor to base from > >> collector. For max AC gain decouple it midway with a capacitor to ground. > >> > >> piglet > > > > piglet, :^) I've never done a suicide bias... mostly I've got > > some opamp or digital output driving a transistor. (with neg. > > feedback somewhere.) > > > > So I drew a resistor from collector to base, do I still keep the R > > from the power rail to base? (or just get rid of that... > > as long as the load has a resistive part, current is going to flow.) > > And then split that resistor in the middle with a cap to ground... > > 'cause that's going to be too much bias (on average). > > > > I'm much better with pictures, words confuse me at times. > > > > George h. > > > > > > Here is a picture of simple self bias. It is negative feedback, broken > at AC in the lower sketch. > > <https://www.dropbox.com/s/ze0vo98nut8gwqc/_SelfbiasedQ.jpg?raw=1>Thanks. GH> > piglet
Reply by ●August 15, 20192019-08-15
On Thu, 15 Aug 2019 06:35:17 -0700 (PDT), George Herold <gherold@teachspin.com> wrote:>On Thursday, August 15, 2019 at 6:31:22 AM UTC-4, Jasen Betts wrote: >> On 2019-08-14, George Herold <gherold@teachspin.com> wrote: >> > On Wednesday, August 14, 2019 at 7:46:32 AM UTC-4, tabb...@gmail.com wrote: >> >> On Wednesday, 14 August 2019 11:09:55 UTC+1, Steve Wilson wrote: >> >> >> >> > Your formula is useless and misleading. You normally never run a transistor >> >> > with the emitter grounded. >> >> >> >> odd thing to say >> >> >> >> > You need some way to stabilize the bias. Also, >> >> >> >> There's more than one way to do that. And despite the myths, suicide bias IS used in commercial circuits. >> >> >> >> >> >> NT >> > >> > What's suicide bias? I went searching for biploar transistor and suicide bias... but got a lot of mental health sites. (maybe lithium doping would help. :^) >> >> It's been a term of art here (in sci.electronics) for as long as I can remeber. >> >> Basically the the base is connected to some sort of current source, and the >> expression for emitter current is dominted by the beta. >> >> --+-+--------+-----+-+-------+-- VCC >> | | | | | R >> | | R | R +-+ >> | R +-+ R | R | >> R | R | | c | c >> | c | c +b +b >> +b +b | e | e >> e e R | R | >> | | | | | | >> ----+--------+-----+-+-----+-+-- GND >> Q1 Q2 Q3 Q4 >> >> Q1, Q2, Q3 suicide bias, Q4 maybe notIt's not. There's feedback, essentially a vbe-multiplier. Q2 isn't suicidal because there's feedback there too. The transistor can't saturate no matter how high the beta.>> >> -- >> When I tried casting out nines I made a hash of it. > >Thanks, Isn't Q3 the 'first circuit' one learns to bias a common emitter? >Well, it was the first circuit I learned. > >GHUsually an emitter resistor would be added. Otherwise it's still suicide bias. One nice option is to use the circuit of Q4, but return the lower resistor to a negative supply. That defines the operating point better over temperature.
Reply by ●August 15, 20192019-08-15
On 8/14/19 7:46 AM, tabbypurr@gmail.com wrote:> On Wednesday, 14 August 2019 11:09:55 UTC+1, Steve Wilson wrote: > >> Your formula is useless and misleading. You normally never run a transistor >> with the emitter grounded. > > odd thing to say > >> You need some way to stabilize the bias. Also, > > There's more than one way to do that. And despite the myths, suicide bias IS used in commercial circuits. > > > NT >Sure, e.g. the late lamented BCV61/62 current mirrors, where the output side will run away if its dissipation gets too large. (It's made of two chips, so the thermal coupling is lousy.) There's a spec for that in the datasheet. 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
Reply by ●August 15, 20192019-08-15
On 8/14/19 10:25 AM, jlarkin@highlandsniptechnology.com wrote:> On Wed, 14 Aug 2019 04:46:28 -0700 (PDT), tabbypurr@gmail.com wrote: > >> On Wednesday, 14 August 2019 11:09:55 UTC+1, Steve Wilson wrote: >> >>> Your formula is useless and misleading. You normally never run a transistor >>> with the emitter grounded. >> >> odd thing to say >> >>> You need some way to stabilize the bias. Also, >> >> There's more than one way to do that. And despite the myths, suicide bias IS used in commercial circuits. >> >> >> NT > > The same simple concepts can be appied to a differential amplifier > biased by a current sink. > > Suicide bias has poor control over Ic hence poor control over the > voltage drop in the collector resistor, hence less predictable voltage > gain. > > For max gain and lowest noise in a single transistor amp, ground the > emitter. >Improves stability too. A bit of CB feedback stabilizes the operating point. 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
