Is S.E.D actually sci.electronics.DUMMIES ??

There's no damned CMRR to speak of and the transistor current biasing is all wrong with Q1 conducting nearly 2x the current of Q2 etc. I'm pretty sure the distortion performance is more an accident of input signal level than by design too.

no I just find it fascinating that its so difficult to arrange an experiment that measures it in one direction only, that I dont think anyone has done it. everything seems to be extrapolated form round trip experiments.

its effectivly been done with two fixed hydrogen masers in buildings some distance apart. the results seemed to cuase some controversy in some cirlces.

moving the clocks introduces an error too, trying to move them realy slowly may not be enough. the two hydrogen masers were also moving at different speeds ever so slightly.

Colin =^.^=

Its still a round trip, one part of it wich is fixed in length is through cable, the other variable part is through air. The velocity is proportional to about 0.66C for some typical cable, as oposed to 0.99 (?) for air.

Colin =^.^=

Gack! You really know how to over-complicate a simple analysis don't you ?:-)

...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.

Actually, I'm hard-pressed to even rationalize using that compound NPN/PNP at all. What did it buy? Absolutely nothing for this application!

Just another bad case of pasting stuff together from what others have developed, without understanding how it works.

Not to say that compound devices aren't useful... I use compound NMOS/PMOS all the time.

...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.

It's easily done with two cesium clocks; just apply money. Heck, two rubidiums would work.

HP once bought two airline seats around the world, one for an engineer and one for a battery-powered cesium clock. They synched two clocks and then flew one around the world. The resulting time error was consistant with relativity.

John

[snip]

I'm curious. Do you think I'm unemployed? I typically work a 60 hour week.

...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.

"Jim Thompson"

** YOU LIE !!

Being a total asshole takes 168 hours a week , minimum.

....... Phil

--
So you work 8 days a week? ;-)

I guess it depends on how literally you need it to be in one direction. The reason it is difficult, is there is no way to convey the information from the finish line to the start line without doing a round trip...

However, if you consider light traveling through a fiber, or light bouncing off of mirrors not to be round trip, you could do something with that.

1) Coil up a few kilometers of optical fiber on a spool, and pop your laser pulse into one end, and measure it with your biased PIN diode on the other, and let your scope do its thing... 2) Arrange two almost parallel mirrors a few hundred meters apart, and introduce a beam at a slight off orthogonal angle, and let the beam bounce back and forth between the mirrors hundreds, of times. You can arrange the mirrors so the beam enters and leaves on the same vector. (Hint, if the mirrors are slightly off of parallel, the bounce can be made to travel from one side of the mirror plane to the other and return back)

Of course, strictly speaking, either method involves the beam making a return trip.

Terribly sorry about that, but Heisenberg's uncertainty principle doesn't allow you to know position and velocity at the same time.

-Chuck

My view.... I tend to gravitate toward full op amp audio designs. I only took some interest in the 2 transistor pair analysis. I would change my view if someone said that the mic preamp S/N ratio is not overkill. As seen on:

My hobby years ago was discrete component audio design... I got some transistor analysis scars and don't want to repeat the experience :) My interests now are more on studying power conversion electronics.. D from BC

quoted text -

OK, here is how I would look at it....

Each collector has a load of about 2000. Divide by the 22 ohm resistor, you get a gain of 90. Times 2 for the other side = 180. Multiply times the op amp stgae of 6.8 equals 1224 total. Of course, I am omitting Q1, 3 emitters, but who gives a butt-snuff.

so (((2.2k//22k)/22)*2)*(150k/22k)= 1224 gain (with perfect devices)

As far as the CMRR, could you elaborate? Perhaps an online reference to some related info on the web as a reference?

Some people are better at it than others, and they can do it in fewer hours. A really blistering post only takes minutes.

The usual way of gaining, and broadbanding the gain of, the compound follower:

View in a fixed-width font such as Courier.

. . . .-----. . | | . --- | . R2 . | ve -ve . +----+----------+---Rg---- . | | | . | | gm1veb1 | . hie1 /|\\ R4 . vd | \\v/ | . -- >----' | +----> -vo/2 . 2 | | . +-----. | . | | | gm2vbe2 . R3 hie2 /|\\ . | | \\v/ . | | | . '-----+----' . | . --- . . . . . vo . Av= -- = ?????

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Best, at the moment that I can guess.....

Short circuit the gain setting pot/capacitor and throw away the NPN transistors. It's a differential pair. The gain is gmRC. gm is set by the 'tail' resistor which is R2||R6.

Put the NPNs back in again. It's still a differential amplifier but the NPNs 'pin' the PNP currents at Vbe/Rbe which makes gm some different value......, gm' less than before.

However the NPNs 'boost' that value by approximately their Beta....Bnpn [1] So

Av = BnpnRCgm' = BnpnRCIcpnp/25E-3 = BnpnRCVbe/25E-3Rbe

That might be a factor of two out because it's a diferential amplifier.

Open circuit the gain setting network and it's just a pair of 'thingy' amplifiers with

Av = RC/RE

In between........ I couldn't be bothered.

[1] Wildish guess that 'feels' right because there is some feedyback thing going on and I think gm is a measure of re, or something.

I won't push my luck further on 'second' order effects because I probably haven't got the first order ones right.

On the basis of the above analysis..... the gain is going to vary all over the shop as Bnpn (and other stuff) varies. So yes it is crap.

DNA

It was in elektor magazine, long time ago, the output stage of an amp would have negative feedback, the output driver stage would have feed back from the output to its input, and finally the input stage would have feedback from the output too, this adds up to a lot more feedback than would be be stable with just one non nested feedback loop.

I think it was the time when people thought negative feedback was good so more must be better.

Colin =^.^=

Let's hold-off a bit and see if anyone else can get the answer ;-)

(Since I do this all the time in CMOS... one presently in the hopper uses this very thing at a 2.2V supply, but has current sources in appropriate places.)

...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.

Besides, I'm not a "total" asshole, I'm a "PERFECT" asshole ;-)

...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 | |

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

I had a strange fualt with 2 soundcards wich fed into my hifi, the output capacitors would go leaky, when I measured the output it was several volts dc, when I touched a resistor accross the op the voltage went away, and stayed away, only to come back some time later, so eventually I soldered a

2k resistor on the soundcard ops lol.

yeah and if its made in enough quantities this isnt an issue, I gues the number of pots on a large mixer may qualify.

the impedance into the op amp stage will then be more unbalanced, wich I hadnt thought of before, wich will mean unbalanced voltages, so a current source here might be needed to improve the CMRR. ofc with a current mirror this will give you a single ended signal.

ah hadnt thought of a current source here.

starts to look a bit more like one or two op amp schematics ive seen.

Colin =^.^=