wideband inductors

Can't do that. The PSPL parts have roughly 1500 pF DC feedthrus, which would resonate with an external inductor. When they call a port "DC", they really mean it!

Um, they *are* ebay parts.

John

The Vishay axial is 47 uH with a measured SRF of 28 MHz, which is only

0.69 pF, a fraction of that of a typical surface-mount part. Interesting that an old thru-hole part would be so much better.

John

....

I have seen those compartments before. Doing a design with gain blocks (Minicircuits, Agilent MGA-xxxx, RFMD RF24XX) at 3 GHz at the moment, I wonder where these compartments walls should be:

- over the line connecting two gain stages or

- over the gain block

In most designs I have seen, the wall are over the connecting lines. However placing the walls over the gain blocks would better break the "line of sight" between input and output of the gain block. Do I get something wrong?

Thanks for the help so long!

--
Uwe Bonnes                bon@elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

The digikey page with the Panasonic inductors gives a good picture. The smaller the size, the higher the self resonance frequency. There is also some influence of the construction, but the size is the most important.

Bye

-- Uwe Bonnes snipped-for-privacy@elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt

--------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- e mount inductors.)

-- Uwe Bonnes snipped-for-privacy@elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt

--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

Hello Uwe,

With filters this is often the best approach. You have to place walls whenever coupling between two sections is otherwise hard to achieve or would require too much circuit spread (real estate). Predominantly the unwanted coupling is between inductors.

That's what I do especially with transistor stages. It sometimes does require the source or emitter to be connected to the wall, meaning there has to be some kind of control loop for the quiescent current and gain. Another area where walls across active devices work great are fast log amps when their dynamic range exceeds 60dB. These can otherwise become pretty 'hot'.

No, it's just that this is often the cheaper solution. Any notch that has to be stamped out adds cost. Also, when the achieved margins are good enough there may not be a need for a better approach.

You are welcome. That's what newsgroups are for.

Regards, Joerg

Hi John,

Well, to a large extent it's the SMT pads. A through-hole part is longer and has wires. Which is why I often place regular beads on a wire even when the rest of the board is all SMT. Come to think of it I believe I never designed a through-hole board. All of them were SMT but filter stuff sometimes had to be non-SMT.

Regards, Joerg