2m front end filter

Folks,

(*** see updates here and here)

As you are aware one of the key features of the FCD is its wide frequency range of operation. While this is great, the compromise with wide band receivers are that they can suffer from desense from out of band signals.

Although the FCD has programmable and fixed front end filtering, at VHF the filters are quite wide and if you’re in a strong signal area, popping in a low loss front end filter will help here.

I’ve built three 2m band pass filters based on the design here.

What I’ve discovered is that you really do need to use the PCB layout with the through hole links, and you must use screening around the unit either with PCB material or for example some brass shim. Without these considerations I found the filter lossy and difficult to tune.

In addition, I found the inductors as specified made the filter LF at 116MHz, so a fair bit of stretching was needed. Luckily I have a VNA in the kitchen lab to do this although you could have a go yourself, tuning for maximum signal on a beacon.

I managed 1.2dB loss with about two or three minutes of coil massage. Probably I could tweak it a bit more given time.

The pictures show the filter and its response. The span is from 50MHz to 250MHz. Thus, in the UK, it might not be the best if you are suffering close in pager interference around 138 or 153MHz where notch filters will provide better immunity.

Howard

This entry was posted in Uncategorized. Bookmark the permalink.

21 Responses to 2m front end filter

  1. Alan Melia says:

    Hi Howard I have a small number of punched “filter boxes” that might be usable for this purpose. They are punched for BNC but I also have some unpunched ones. I could probably get some punched for SMA but I guess most people could drill a couple of holes and save 30p :-))
    The is a picture part way down the web page quoted above.

    Great Work….Thanks
    Alan G3NYK

  2. Hi Howard,
    I was thinking to use some surplus filters ; I have them from long time and I have used them with elder tech rxs in the past. One is a very small helicoidal filter, 5 cells, thae was used in Autophone mobile riogs for police etc. I do not know its insertion loss, but it’s very good to use it for listening if you have strong signals in the band.
    For weak signals usages, I used these times a surplus silvered medium size single cavity filter, very sharp, on the rx cable before the antenna in of the RX but after the power RF switches & eme pre that are on the antenna itself.
    When I’ll get the FCD, I’ll notify you; hope that the address error reported by paypal will not affect the arrival time , should not .
    73 de Augusto

  3. Ernest says:

    Not sure as to what type of front end filtering
    Is required hereā€¦
    Check here under Amateur radio. Myself use the
    Satellite units which work very well.

    http://www.dci.ca

    Ern

    • admin says:

      Hello Ern.

      Thank you for your note. Absolutely, these work, I have a set myself for 144 and 435MHz. But they are twice as expensive and about 100 times the volume of the FCD though!

      Howard

  4. admin says:

    Update: I spent some time understanding this filter a little more in an effort to narrow the filter’s passband.

    I changed the inductors to 97, 101 and 97nH respectively. The 10pF caps are changed to 8.1pF. The 3.3pF caps are changed to 1.8pF. All caps were 0603 size.

    I found that if you wind the 1mm diameter copper wire coils on a 5mm drill bit as a former you end up with an outside diameter of 7mm as it springs out a bit. This works with the above inductor values, but to tune I had to pull them out to about 11 or 12mm.

    This gives a slighty increased 1.4dB insertion loss, but with the benefit of being 10dB down at 138MHz and 153MHz. Not as much as I’d like, but a step in the right direction. The circuit simulation suggested 16 to 17dB down, I guess I need to figure out some higher Q coils.

    Howard

  5. admin says:

    I have been experimenting with using preamps. Although we don’t really need much gain, it seems that one means of getting a narrow bandpass filter without excessive losses might be to use an active method. I’ve been experimenting with a BF998 design, it has 138MHz at -30dB and 153MHz at -48dB. Elated as I was, I then discovered that the noise figure was dire, so I need to do a bit more fiddling. See pics 4 & 5 in the gallery at the top.

    By the way, the next FCD PCB spin will probably have software switchable 5V phantom power for such a preamp.

    Howard

  6. jimmie reeves says:

    Receiving my funcube , After purchasing a sma connector to 239 I found that as shipped (not as pictured ) Wher I live the smb and smc. dont know which is the new connector is next to inposible to get . any have any suggestions? is it a c or b?

    • admin says:

      Jimmie

      I can assure you that it is an SMA. I test every FCD it with the same setup and that has an SMA plug on the end of it.

      Are you sure your adapter isn’t an RP (reverse polarity) SMA?

      Howard

      • jimmie reeves says:

        As I stated in my email , to Howard , There was a dead short between the floor and the key board. I accepteed Radioshack ‘s markings a being accurate . The connector is a sma ,not smb or smc as stated . by the way the sma to 239 cables can be purchased on ebay for $3.00 not $19 as radioshack charges . You still have to buy a coupling. sorry jim

  7. Pizarro says:

    What kind of pre-amp we can use? The bias currente is available? What is LNA enhance? How did it work?

    73

    • admin says:

      Hello Pizarro

      Within the FUNcube team we are working on a preamp: the FCD itself is very sensitive, but at VHF in particular the front end preselection filter is quite wide, so it is subject to out of band strong signals causing intermodulation and blocking. There is some information on some filters I’ve been experimenting with here:

      http://www.funcubedongle.com/?p=456
      http://www.funcubedongle.com/?p=611

      You can safely take up to about 50mA at 5V. LNA Enhance only works when you are runnning at full LNA gain, and even then I have been unable to really see it make much difference. It is a function available on the silicon, and I have provided access to it from the front end for experimentation.

      Many thanks, Howard

  8. omidksh says:

    schematic diagram filter band pass 2m ?

  9. reza says:

    I dont know what kind of filter is it, (butterworth,elliptic,chebychev). but it is so interesting for me. What is the capacitor on beside 10pf and 3.3pf because in the pdf file (schematic) there are capacitor series with ground. What is the value?

    Thank you very much. Rezade

  10. reza says:

    Oh, i’ve been reading. i simulated with filter software. the result is great.
    I want to try to build it. What kind of pcb do you use? fr4 or something.
    How width of the line in pcb (mm)?
    How about the size of box?

  11. reza says:

    Do you have a air coil calculator?
    because in google, there are many type of them.
    I dont know which one is true.
    What is your reference for making the coil.

    Thank you very much

  12. reza says:

    excuse me sir, if i want to build 101nH and 97nH. How much turn and diameter to build it?

    • admin says:

      Hello Reza,

      Make them the same as the original design, and then hand massage them for the best response. If you have a network analyser or tracking generator plus spectrum analyser, it will make the job very easy.

      It’s not really possible to make these parts accurately at home without having to do some adjustments in situ.

      Many thanks, Howard

Leave a Reply

Your email address will not be published. Required fields are marked *


− six = 3

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>