Mini-Circuits has been introducing new phemts, which is great because other people (NEC, Broadcom) are killing theirs off.
Wideband bipolars are still fading fast.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Cool that they've got higher-power ones.
Cheers
Phil Hobbs
I measured the SAV551 as a dc switch, Rds-on and such. The RF part specs are mainly useless.
How about 2 ohms Ron and 0.4 pF off. 800 fs time constant.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Horrid. I love it.
Cheers
Phil Hobbs
What does that mean? 0.0008ns?
--
Thanks,
- Win
2 ohms times 0.4 pF is 0.8 ps = 800 fs, But I don't understand where the 2 ohms came from.
George H. (not a high speed guy.)
Rds(on), likely.
--
Thanks,
- Win
Right, but the 0.4pF isn't a constant, and neither is 2 ohms, so the actual TC will be the integral. I wonder how different that is?
Clifford Heath
Aren't those number pretty meaningless? John is looking at it as a switching transistor. As a linear circuit amplifier, best f_T = gm / 2pi C, where C would include wiring, load, etc.
--
Thanks,
- Win
Yes. That takes about +0.6 volts on the gate.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
R*C is referred to as a "time constant" in our world.
The engineering units of R*C is seconds.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
Whoosh!
What's the off resistance? What shape is the curve to the on resistance?
What's the on capacitance? What shape is the curve to the off capacitance?
The TC is the integral of these R*C curves.
Clifford Heath.
Are you interested in electronics or are you just here to be obnoxious, like most of the other posters?
Phemts, like other fets, are more constant-current than ohmic at higher drain voltages. I'm using them as switches and I define and care about the low voltage value of Rds-on. Most mosfet data sheets specify Rds-on, but the RF parts never do, because they assume that people will not use them at low drain voltages.
Cd*Rds-on is a serious figure of merit in my businness. The product is seconds. I call that a time constant. Cdg is important too, and it's crazy low for phemts and GaN.
I can barely define that, much less measure it. It certainly doesn't matter to me.
I only took as much data as I needed for my current application. It looks great.
A full set of drain curves would be interesting, but I ran out of time.
It would also be fun to use them as diodes, but the voltage rating is low enough that one may as well use a schottky. Some bigger phemts do things that no diode can do.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
No, just pointing out that you didn't understand my question.
Duh, please stop repeating the obvious.
Ok no worries. I recall that you were using pHEMTs to discharge fast ramp generators, because of the effect of non-constant C (dropping from
1pF to 0.44pF with drain voltage) on ramp linearity. The transition speed matters less there of course, but it made me think about the effect it must have on TC and charge/discharge curves.The values you listed show a very sharp transition indeed, but probably not enough accuracy to deduce a discharge curve TC. If you have no more data, no worries, it was just curiosity.
Clifford Heath.
We have decided that the best current source for a ramp is a resistor. The exponential nonlinearity, and any effects from the drain nonlinear capacitance (and various ESD diodes and things) can be corrected with a polynomial somewhere.
People are EOLing fast PNP transistors. They will always sell us resistors.
All I care about on the discharge is that it be fast, and go close to zero. Both the rise and fall of a fast capacitive ramp tend to ring too; it's messy.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.