#606 RF Bandpass Filter with a Surprise

Episode 606
The filter also contained a directional coupler
Part Number 6927E92, was made by Varian for the Minuteman Missile system
original video about the filter:
Be a Patron:

[subramanianr]

Oh. What a fantastic architecture.! An analog BPF with 0.9 dB loss and rock steady construction. The PCB trace coupling with 50 ohm termination at one end and the other end for observing/ and tuning the filter at the factory as well for future maintenance adjustment if needed.
And the small wires from the studs on the underside of the PCB are probably "gimmick capacitors" to handle the leakage inductance of the long trace.
there.
Thanks for the presentation 😊

[radiolens]

Another nice video from a an inquisitive technical mind. I remember Varian, they used to make high-power amplifiers (HPA) for satellite earth stations, a very respected name in the satcom industry.

[petechongy]

Make a nice dab pass filter and thanks for the vids you have taught me quite a few things, good work.

[germanjohn]

lol looking at the coupled part without terminating the filter? Wow, just wow.

[fredmitchel]

Thanks IMSAI Guy...



The little cavity is to allow space for the wires to be soldered

First tiny glance your coupler looks like a transformer...to me....

With all the poles in the BP filter..

Does it then have 5 poles?

Surprised at the frequency beyond the filter has
So maybe it is Mr. Chebyshev...nice and flat in the passband...outside

Does the rolloff steepness match a 5 pole..I think perhaps so....

Or does 7 caps and inductors constitute 3 poles...since the two caps on the front and end form the input and output...?

I am glad you didn't tweak...but measured first with my future envious tool....the Nano VNA...

There was probably lots of effort the test person tweaking the adjustable caps to get that nice flat passband response..

Curiosity drove me to MiniCircuits...Couplers..

Since your insertion loss in passband 0.9 dB...then I guess the coupler couples 8 dB...

They listed a nice table of theoretical minimum thru mainline insertion loss in dB...

Such as a 6 dB coupler will have 1.2 dB minimum loss...no active components

10 dB coupler = 0.46 dB IL
15 dB coupler = 0.14 dB

And 30 dB coupler 0.004 dB IL

That makes sense...for if your sucking a lot off power from the main line to your coupler...then less power comes out of the main end...

So what we might have folks is a bandpass filter Fc 240 Mz...with a 8 dB ish coupler ..with a bandpass of 80 MHz wide...

I did not quite hear that reader input of the program or equipment...

A "Soulmate"...?

It's all good...

Nice construction design and I am glad the curiosity cat with IMASI Guy won out and then he showed us...

So I can see how that coupled port could be fed back to level out...perhaps in like a system of AGC...

As a function of power into the filter...I wonder if there in like the upper noise floor do frequencies power like at 400 MHz ever smooth out....like there is a sweet spot of like a range of millawatts in or what is the range of power....do the inductors saturate....?

And Fred you can go read too...

240 MHz...militaryish UHF...

Or maybe Soulmate is just part of a spectrum analyzer instrument?

Thanks IMASI Guy...once Again...

I am looking foward to our very wise readers feedback...and their comments....

May no one consider me a babbling pest

Kind regards

Fred

[AECRADIO]

VHF filter, LC.
The PC trace is either a coupled input, or output.
Time to make use of those piston caps.

[subramanianr]

I am sorry, IMSAI guy, now, only now I understand correctly. Before the signal even enters into the filter section it passes through the long "U" like PCB trace; a parallel trace with a 50 ohm on one end (termination) comes out as the coupled port for the purpose of observing (like you did on the nanoVNA) and for fine tuning of the capacitors at the factory.
The hollow region of the whole block is for taking the tappings out; the input and the output of the filter are let through plastic (teflon or something) insulator; so is the observation coupled port. Someone down here asked if the coupling is on the input or on the output.
The sample for observation is taken out (by the PCB trace coupling) on the input of the filter even before it enters the filter section.
Thanks 👍

[danboy]

Any chance those horizontal inductors are resistors wound with an inductor coil?

[subramanianr]

Oh, I now understand that the small wires are the ones capacitively picking up the signal.

[ruhnet]

Very neat filter. Just out of curiosity does the insertion loss change when a load is on the coupled port, vs when it is left open? Looks like the coupled port is intended to be approximately 10dB down.