Why this transistor?

Yet quite irrelevant to this application most likely. The signal source isn't shown.

Graham

30MHz.http://homepage.eircom.net/~ei9gq/stab.html

It's simply AD = Analog Devices OP07 ! = ADOP07

Duh !

Graham

Toff is only relevant to saturated switching.

Probably because he's an idiot. There's no way there's any justification for an OP07 there. OTOH they're not so expensive as not to stick with one.

Graham

Not any of the ones I have.

I imagine that Google might prove that incorrect.

You're from a different part of the world where BC parts are more commonly found.

I'm sure there are. :-)

I wonder if the modern versions of the 2n2369 and similar parts are still gold doped. I have about 33 datasheets of various versions of '2369 transistors stored in my computer, and only the very oldest ones, like Raytheon's datasheet for the bare die, mention the gold doping.

But we do know what the signal source is. It is from a VFO, so it's quite likely that the signal is 1V to 2V peak to peak. With no base resistor, would that make it relevant? Seems like an excessive amount of load on the circuit supplying the signal, given the purpose. I don't know that for sure, I'm just trying to learn.

[snip]

I would presume they are still gold-doped. Otherwise how would they meet the storage time spec?

...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     |
             
         America: Land of the Free, Because of the Brave

How are RF transistors made, anyway? Small-ass dies, dash of spices (like gold) to reduce recombination, etc.? Is it then impossible (with silicon) to maximize all parameters? Like, say I want a transistor that does like...

600Vceo, Ic > 1A, t_stg < 10ns, and hFE > 200. (Nevermind lead inductance on actually putting that much collector current and bandwidth into practice...)

Tim

-- "Librarians are hiding something." - Steven Colbert Website @

There will be trade-offs. Such as the hfe penalty that usually comes with gold-doping. Don't expect much above 50, typically a lot less.

Then there are market forces at work. There isn't a lot of market for high voltage RF transistors. Some of the power horses I grew up with have vanished :-(

--
Regards, Joerg

http://www.analogconsultants.com

Yeah. I mean, HV transistors usually have appalling hFE anyway (typical line output transistor circa Vceo ~ 1.5kV, hFE = 4, t_stg = 3us!), but that's optimized for something else. What's the ultimate maximum between Vceo, hFE, fT, etc. that can be physically produced, you know?

I certainly wouldn't expect to find something like this for sale! Most RF amps are modules at what, 48V or something? Throw together a thousand (plus a nutsoid transformer and rectifier for the kiloamp supply rail) and you've got a massively-redundant FM transmitter...or something like that...

Tim

-- "Librarians are hiding something." - Steven Colbert Website @

No idea, I just take what's on the market (and is cheap...). Guys like Jim would most likely know the trade-off details.

Not quite a thousand but modular combos are popular.

--
Regards, Joerg

http://www.analogconsultants.com

You appear to have noticed a genuine area of concern.

Yes, that input stage is total rubbish. What would the source impedance be ? 75 ohms ?

Graham

Glad that somebody finally got it right. I have noticed that at 50 MHz, the

2N2369 will have at least 6 db more gain than the 2222.

Tam

In which case they're not 2N2222s. Probably PN2222s.

One reason I'm not a fan of 2222s is their relatively low current gain by today's standards.

Graham

Hello Tim,

It seems that we are focusing on a combination of high collector voltage, current and Ft and HFE. You are right, RF transistors have extremely small geometry. To get some current capability, many of them are in parallel (on the same die). To make them more rugged, they add some emitter resistance to get a reasonable current distribution over the complete die.

For an RF amplifier, low HFE means just some more bias current. At HF, ft, Ccb and the base spreading resistance (Rbb') is of importance. At RF, the input signal is just charging (and recharging) Ccb and the BE diffusion capacitance (not confusing with Ceb). This all must go via Rbb'. If that one is high, lots of the input power is lost in it, hence reducing gain.

Some datasheets do mention the Ccb*Rbb' product and you can estimate Rbb'. When you are using SPICE, the bad modeling of Rbb' gives unreliable results with respect to gain and power efficiency.

With regards to saturated behaviour, the TR (reverse transit time) parameter is not or badly extracted. In simulation your circuit behaves perfectly, but when you build it, it doesn't function.

When it comes to amplification (so not saturated behaviour), the maximum oscillation frequency would be nice to know also, however it is mentioned in rare cases only.

Silicon bipolair for RF power is becoming less popular. When you look to base stations for UHF cellular, mostly metal gate lateral mosfets are used (LD MOSFET). Nowadays it is not uncommon to buy a device that outputs 100W at 2.5 GHz (continuously). In the lower frequency range, you also see more and more (polysilicon gate) MOSFETs. IXIS have nice devices

Best Regards,

Wim PA3DJS

He had a bunch of them on his shelf.

John

The really big ones (used to) have multiple emitters. Back when I was young we bought those that had "partially" failed. Some of them emitters must have blown and they'd almost give them to use. They were still good even though for a few dB less.

[...]

--
Regards, Joerg

http://www.analogconsultants.com

"Anthony Fremont" wrote in news: snipped-for-privacy@news.supernews.com:

You probably have seen 2N2222A, which have a TO-92 plastic case. I do have one 2N2222 in a metal can. I don't know why some people would actually use one though, unless they don't know about substitution, or need to take parts lists literally.

Here I use 2SA933 and 2SC1815 , as that is what I mostly pull from Asian electronics.