You'll have a hard time finding that. Logically, one might conclude your numbers are wrong... :)
What, exactly, are you doing?
Tim
-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website:
I read somewhere that radio hams making ultra-low noise oscillators found they got best results using that feedback amplifier technique (cool name Kanner Kap) with a low value resistor (4r7 or 3r9 ?) between input and output.
piglet
Should that be "inverting" instead of "noninverting"?
piglet
Quite right, thanks--I was thinking of negative impedance converters. (My n ext product uses one to cancel out the input capacitance of an ADA4817 over a modest frequency range.) The Kanner idea is to move the cold end of a BF C to make it look like a monster BFC.
Cheers
Phil Hobbs
Don't tell him, Syd. He works for the Secret Service. Probably. ;-)
By BFC do you mean Buck Fed Converter? In my simple mind I visualize the Kanner cap as a kind of integrator.
piglet
Rather than calculating exactly a value that is just high enough, a BFC will do the trick.
Big F***-off Capacitor! :)
(Or some variant of the above)
-- John Devereux
I can neither confirm nor deny such accusations.
(The 'S' in "NSA" stands for "Seven"?... ...I don't know where to go with that, though.)
Tim
-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: http://seventransistorlabs.com
Roight, big effing capacitor. The limitations are the output resistance and slew rate of the amp, and of course the output swing is N times the residual ripple. It can be a bit of a puzzle to bias as well, which can limit its low-frequency performance.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
That's what R2 is for--it turns the stage into a V_BE multiplier. Because V_BE doesn't change much, R2 draws a roughly constant current, which puts a roughly constant voltage across R1.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
You need to run it off a separate supply, or maybe off the rail ahead of the series resistor. In Syd's case, where the rail is HV, a lower supply voltage would save a lot of power and probably several headaches.
Of course you have to do something to protect the amp's input as well.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
It's not nearly as good as a regulating cap multiplier, because the amp will be fairly noisy, and all of that noise will wind up on the quiet rail. The 1-Hz noise of a cap multiplier can be below 1 nV.
You can make cap multipliers regulate in a few different ways, e.g. hanging a TLV431 on the base of the last transistor via a single RC, or (my fave) using split feedback with an adjustable regulator upstream. It's like the usual split feedback for driving capacitive loads--you take DC feedback from the cap multiplier output and AC feedback from the regulator output. That way you don't get the noise of the regulator on the quiet rail.
You have to watch out for the sneak path through the feedback resistor--it's liable to limit the noise rejection. Cap multipliers are the bomb--I couldn't build very many of my gizmos without them.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
It's power line comms, stopping the signal from being shorted by the supply. The 8 Amps is specified, the 10mH could go to 5mH without too much signal loss.
Cheers
-- Syd
So, just a regular bias tee?
Why does the RF have to be >300 ohms impedance? Why not 50, or even, say,
10 ohms?Why does the comm have to go down to 10kHz? Why not, say, 300kHz, or
13.56MHz? Why 90kHz (~300% of center) bandwidth and not more or less (using more efficient, but more noise sensitive, modulation schemes)?Why not use a multi-stage (i.e., more than 2 poles) filter, to get higher isolation for less energy storage? (i.e., what you're really doing is isolating the RF and DC ports, since the DC's signal is, well, zero.)
For example, the same frequency difference at 300kHz uses only 13%BW. At
13.56MHz, with maybe 16QAM, the bandwidth can even be Part 15 ISM compliant, over the air (so it doesn't matter if it's leaking out of the wires, which means even easier wiring than 10BASE-T).Tim
-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: http://seventransistorlabs.com
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.