Bandpass filters

During the last lighthouse weekend in Ouddorp Zeeland we found out the hard way that there is a lot of radio and radar equipment in lighthouses; this resulted in high QRM levels on the HF bands, the noise floor on my FT-857D was S7 to S9 on the 80m band near the lighthouse. Also, radio operators next to one another on different bands interfere with each other. The advise I got was to invest some time in set of a bandpass filters (BPFs). I got a DIY filter kit from DG0SA. His BPFs are rated at 200W, they are traditional 3-kreis filters and you can get them for all amateur bands. I have built the 80m, 40m and 20m BPFs and this is what you get:

The documentation of the filters can be found on Wolfgang Wipperman’s website
In the end this is what they look like, it took me 12 hours to build three filters. Tricky is the L2 torroid which comes with a trap. You can adjust C2 with tuning capacitors.
Performance of the 20m bandpass filter.
Performance of the 40m bandpass filter.
Performance of the 80m filter.

The insertion losses of the BPFs within the amateur bands is less than 1dB, and the crossband suppression of each filter is approximately 30dB, if you want something better than this then you need to build higher-order filters. I thought it was too much trouble for what I wanted. Wolfgang provides tuning capacitors to adjust the C2 capacitor, but in the end I did not use them. I did check the resonance frequencies and the filter characteristics with my VNA, without a VNA you need to find a friend to help you out.

Receiver test with the Yaesu FT-991

I tested the 40m filter on a weak station, an S1 station on the 40m band with the FT-991 directly connected to the g5rvj antenna (DSP filtering was off, no pre-amplifiers: IPO and no noise blanker). The BPF was inserted in the 50Ohm segment before the tuner that I use for the g5rvj antenna. Once you insert the BPF the S1 station becomes somewhat stronger, possibly up to 2S points. Reason is that the FT-991 is challenged by strong out of band signals that enter the receiver. With a BPF you make it easier for the receiver to select only that signal you are interested in, reason is that the BPF suppresses everything outside the 40m band. This may be an explanation why the signal to noise ratio is somewhat improved with the BPF.

Receiver test with the SDR: airspy / spyverter combination

The test is similar, but also more interesting because the SDR comes mostly with digital filtering, nothing else reduces out of band signals and it is expected that the SDR will be more affected by large out-of-band signals. The SDR combination is connected to my favorite active antenna that I describe here. The consequence of inserting a 40m BPF is illustrated in the waterfall plot below, we are centered on the high end of the 40 meter band.


The black bar in the center of the waterfall plot is the BPF switchover point, the lower part shows the waterfall where the BPF is inserted, the upper part sis the waterfall where the SDR is directly connected to the active antenna. All other settings in the SDR# software (gains for HF, mixer, IF and the AGC are the most important parameters) are the same. So here we also see a gain of a couple of dBs, I measured 5 dB, moreover you can see that the intermods caused by strong broadcast stations is significantly reduced. This test works best in the evening when the 40m propagation is enhanced compared to the daylight hours.

Intermodulation products are likely to be reduced by inserting a BPF before the receiver, the plot below is the same as others here, the lower part is without the filter, and upper part is with. The intermodulation product of a broadcast station is indicated by the yellow arrow, and it disappears when you insert the filter.


Last update: 6-sep-2017


4 thoughts on “Bandpass filters

  1. I have had an 57D and the performance was even with small antennas in a rral area in te evening something to make me cry. The 7000 did a better job there. But hey. nothing to complain.

    Regarding the 991 where you state that the SNR got better with the BPF — cool stuff. Were you able to measure the SNR improvements? If you can hear it, it’s already substantial, obviously.

    Now if I look at the airspy, I seem to see an increase of signals/overload and noise. Did you try to measure the SNR changes there? If I only would see the image and not listen at all, it looks bad to me but then again — the eyes can be deceiving. I generally see a decrease of the noise.

    FWIW, I am using two A octave trk presel at the front-end accompanied by switchable digi-sel filters. Thta’s only in the RX path obviously so for multi-multi it doesn’t help. In fact, harmonics still would be received if the TX is not clean
    See for some information. (I do not have a 7700 though)

    1. Hi Roeland, these BPFs work in both ways, TX and RX. Measuring the SNR is a bit tricky, because the SDR test showed that you can get rid of an intermod, so if I would measure it at that part of the spectrum it would be something else than that I would measure the SNR at another part where there is no intermod. Ernst

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s