really tiny transistor

80GHz fT, but 2.25V max. Whaddya gonna do with it?

--
 Thanks, 
    - Win

Me? Probably nothing. I don't do (very) small-signal RF.

I was impressed by some of the numbers.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

I wonder if they could have made the dimension text any smaller, gee

1/3 of the page to dimensions. It's like reading a 0603 resistor with out a magnifier.

Cheers

Would be really nice if it was more like 3.3 Vceo. Not sure what I'd use it for either but....

It seems to be usable as a transmitter "power" amplifier of 0 dBm up to 10 GHz and even a few dBm at frequencies below 5 GHz. That should be enough feeding an individual patch antenna e.g. in a MIMO array.

Cascode the drain of a 12-GHz pHEMT, if you can keep it from oscillating. I use a fair number of its 45-GHz brethren like that.

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

It has a damn long p/n: BFP840FESDH6327XTSA1

--
 Thanks, 
    - Win

What about the NXP SiGe ON5088 with Vcbo=10 volts? Use to 40mA, 55GHz at 25mA, Ccb=70fF at 2V, $0.23, in a sensible 4-pin SOT343F package. Two emitter pins on opposite corners, run your big fat copper emitter ground between the base and collector. Too bad I don't do much of anything above 3GHz.

--
 Thanks, 
    - Win

That pHEMT + SiGe:C NPN cascode of mine from several years back is essentially the perfect gain device, apart from having a high 1/f corner. Flatband noise of 0.3 nV in 1 Hz, sub-picofarad C_in, practically infinite collector resistance, so you can get a lot of voltage gain with excellent linearity. (pHEMTs have sucky collector resistances.)

It does need a 5-ohm bead (BLM15BB050SN1D) in the base to keep it from oscillating at 12 GHz or so, but the pHEMTs are surprisingly stable, even running as followers.

I used it to make a nanoamp TIA for a biochip project at Samsung's research lab (SAIT) some years back. The spec was "shot-noise limited at 1 nA in 100 MHz bandwidth". 100 MHz one-sided bandwidth is equivalent to 5 ns in the time domain, and 1 nA for 5 ns is 31 electrons. I thought that was completely impossible, but eventually I got within 6 dB by running the first stage completely barefoot except for a DC restore loop, and fixing up the frequency response in the second stage. Worked like the bomb.

The 10-MHz 1/f corner frequency didn't worry me much because the design was limited by e_N*C noise, which suppresses the 1/f contribution. So you really can use these sorts of parts well below a gigahertz.

BTW the Mini Circuits SAV-551+ is even better than the ATF55143.

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

I'm about to try a PHEMT diff pair, driven from ECL, as a differential digital output stage.

Very nice part to replace dead Avago parts. Rds-on is 2 ohms!

The SAV541 has better DC curves, for higher current switching.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Many things. They'd be useful as a transmit stage for pulse-echo such as TDR. Also as a preamp to listen to very faint echoes because of their low noise figure.

The higher the ft the lower the allowed Vce. I wish they'd also spec reverse Vbe because that is important for just about anything connected to the outside world. For diode protection you need to know where to set the negative rail. Less voltage, more capacitance. More voltage ... PHUT.

The device we can buy these days are fantastic. The datasheets, not so fantastic.

--
Regards, Joerg 

http://www.analogconsultants.com/

I have an interesting bootstrap design for high-capacitance, highish speed photoreceivers such as tiled MPPC arrays. It uses a SAV-551+, and initial indications are that it's much quieter than the ATF38143 of the previous version.

All of the 12-GHz f_max pHEMTs that I've used are surprisingly stable--as long as the layout is tight, you can use them as followers, bootstraps, diff pairs, Schmitt triggers, and so on with no birdies at all. The SiGe:C bipolars are a different story, but of course they're much faster.

Fun stuff.

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

Fun, yes. Phemts and GaN have shocking low Miller capacitance, so behave surprisingly well.

I'm hoping my diff pair behaves itself. I have gate resistors just in case, but I don't want to slow things down.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

I did a SKY65050 source-coupled Schmitt trigger, which worked fine but wasn't super fast. That was probably on account of the low drain impedance, which reduced the voltage gain pretty badly. (The ATF devices were much much worse--followers with gains of about 0.6.)

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

The SAV541 data sheet has curves. I eyeball a follower, with a constant-current source sink, as having gain well above 0.9.

Source impedance should be pretty low too, 2 ohms or so.

I might try it.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Not awful--that's almost in normal JFET territory.

Yeah, that's what makes them good bootstraps for applications where you don't need a huge bandwidth increase.

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

You call that a really tiny transistor? That's not a really tiny transistor.

That's a really tiny transistor!

--

  Rick C. 

  - Get 1,000 miles of free Supercharging 
  - Tesla referral code - https://ts.la/richard11209

Web request id: 10136324568529214447:YBA01-CPVNPPRY18:1576255437:0

Report this error

403

--
Johnny B Good

Thanks, that led me to a datasheet. :-)

datasheet-43351533.pdf

--
Johnny B Good