Electronics > RF, Microwave, Ham Radio

Termination impedance of 9MHz IF filter

(1/2) > >>

I am going to change IF in a HF receiver from 8.9875 to 9.0 MHz. The wide monolithic filter XF01 in the schematic below (Yaesu FT-77) will be replaced with a homebrew narrow crystal filter for CW reception only.

My first question is if I need to build the new filter with specific in and out termination impedances, and if so - how can I determine the impedances for the existing filter from the filter itself or/and the circuitry before and after it.

Nothing is known about the filter except from "8F-20A/S, Monolithic Filter, RX IF Filter" - see page 17 in:


The full schematic for the RF Unit is on page 22. From other Yaesu filter terminology I guess the original filter is a 20 kHz AM (/SSB?) filter.

As a "wild guess":
Most filers have the same input and output impedances.
The dual gate amplifier following the filter would have a very high input impedance.
It is shunted by the 3.3K resistor, so I would guess about 3K for the filter impedance.

My wild guess would be that it could be an odd filter with a low Z 50R input and it wants to terminate into ~3k? on the output in parallel with a few pF.

So maybe 50R input and 3k? // -5pF  output?

Also, you need to provide a DC path across the input pins to allow the correct bias control of D41 etc. This would normally be done via the windings of an input transformer inside the xtal filter.

I have searched a bit on the net and found that most IF filters have termination impedance of some 50 to 1000 ohm. The impedance can be transformed with an inductor or transformer before and/or after the filter. Software exists for the design calculations.

I agree with Fank1 regarding the output side. C62 0.01uF and C63 0.022uF (by default) should represent little impedance at 9MHz, so the termination load is 3300 ohm. But on the input side the filter is in a low impedance loop formed by C59 0.022uF in series with with secondary of T20, D41 PIN diode and C61 0.022uF. And the fact that the so called 'monolithic' filter has 4 terminals could mean it has different termination impedance in and out.

I do not know but I have decided I will learn  :-DMM. But of course, in the end I can just ignore the impedance question and try to measure or listen to the power transfer / loss by injecting 1-50uV at the antenna input. All I will experience (I think) is some ripple and asymmetry in the pass band.

G0HSU, you posted while I was writing. I think you are right. R57 470 ohm is RX 13.5 volt through the filter, the PIN diode, the secondary of T20 and to GND via R55 470 ohm.

Is there a way to measure anything to confirm or discard the two-impedance hypothesis?


[0] Message Index

[#] Next page

There was an error while thanking
Go to full version