In a receiver, for wideband (10-150MHz) biasing of a MMIC amplifier using
inductors between the RF trace and the power rails, I'm finding that a single
inductor tends not to work so well due to (1) wanting a largish inductor (say,
one with 500 ohms of reactance at 10MHz in a 50 ohm system -- 8uH) but (2)
not going beyond the self-resonant frequency of the inductor, which of course
is smaller the larger the inductance (that 8uH inductor might typically have
an SRF of 50MHz, noticeably below the 150MHz I'm trying to achieve RF blocking
to!).

Is there a better means of providing wideband DC biasing/RF blocking than just
placing a large and small inductor in series? I've run SPICE simulations of
this, and -- just as when you stack multiple capacitors in parallel for wider
RF coupling -- there are significant anti-resonances that drop the overall
reactance of the pair of inductors to much less than 500 ohm (even less than
50 ohms!) unless you're very careful in the choice of individual inductor
SRFs, inductances, etc.

Thanks,
---Joel Kolstad

"Joel Kolstad" bravely wrote to "All" (23 Oct 05 20:26:37)
--- on the heady topic of "Wideband RF blocking"


Add series or/and parallel resistance to dampen the resonances.

A*s*i*m*o*v


JK From: "Joel Kolstad"
JK Xref: core-easynews rec.radio.amateur.homebrew:88454

JK In a receiver, for wideband (10-150MHz) biasing of a MMIC amplifier
JK using inductors between the RF trace and the power rails, I'm finding
JK that a single inductor tends not to work so well due to (1) wanting a
JK largish inductor (say, one with 500 ohms of reactance at 10MHz in a 50
JK ohm system -- 8uH) but (2) not going beyond the self-resonant
JK frequency of the inductor, which of course is smaller the larger the
JK inductance (that 8uH inductor might typically have an SRF of 50MHz,
JK noticeably below the 150MHz I'm trying to achieve RF blocking to!).

JK Is there a better means of providing wideband DC biasing/RF blocking
JK than just placing a large and small inductor in series? I've run
JK SPICE simulations of this, and -- just as when you stack multiple
JK capacitors in parallel for wider RF coupling -- there are significant
JK anti-resonances that drop the overall reactance of the pair of
JK inductors to much less than 500 ohm (even less than 50 ohms!) unless
JK you're very careful in the choice of individual inductor SRFs,
JK inductances, etc.
JK Thanks,
JK ---Joel Kolstad

.... Why do we park on driveways and drive on parkways?

In article ,
says...

Is there a better means of providing wideband DC biasing/RF blocking than just
placing a large and small inductor in series?

Not really. That's the way it's usually done (along with a resistor).

Have a look at the ADCH-80A from Mini-Circuits -- it's designed
expressly to bias MMICs. Good to several GHz.

-- jm

------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------

Have a look at the ADCH-80A from Mini-Circuits -- it's designed
expressly to bias MMICs. Good to several GHz.

Thanks, that's a pretty nice part!

A resistor alone has also been used. I have used a combination of L & R and
L with ferrite beads as well.
While the 500 ohm (10:1) guideline is a good one, teh impedance can drop
well below that before causing trouble.

ALso, the suggestion of the resistor to "dampen" a resonance is valid. You
can take the choke above self resonance this way (self resonant freq can be
in the upper part of your freq range). In the past, chokes were specially
wound to raise the self resonant freq, but for this range, I suspect there
isn't much help there.
73, Steve, K,9.D;C'I

P.S.
Paralleling caps and series'ing inductors is generally frowned upon. If
needed, parallel caps are usually separated by a bead or resistor. I've not
seen series chokes used.


"Joel Kolstad" wrote in message
...
In a receiver, for wideband (10-150MHz) biasing of a MMIC amplifier using
inductors between the RF trace and the power rails, I'm finding that a
single
inductor tends not to work so well due to (1) wanting a largish inductor
(say,
one with 500 ohms of reactance at 10MHz in a 50 ohm system -- 8uH) but
(2)
not going beyond the self-resonant frequency of the inductor, which of
course
is smaller the larger the inductance (that 8uH inductor might typically
have
an SRF of 50MHz, noticeably below the 150MHz I'm trying to achieve RF
blocking
to!).

Is there a better means of providing wideband DC biasing/RF blocking than
just
placing a large and small inductor in series? I've run SPICE simulations
of
this, and -- just as when you stack multiple capacitors in parallel for
wider
RF coupling -- there are significant anti-resonances that drop the overall
reactance of the pair of inductors to much less than 500 ohm (even less
than
50 ohms!) unless you're very careful in the choice of individual inductor
SRFs, inductances, etc.

Thanks,
---Joel Kolstad