New board committed and parts ordered for first batch!

Last night at 4am I committed what I hope will be the last spin of the PCB before production, should arrive this weekend!

I’ve also ordered a full reeled set of parts for at least 100 units.

This entry was posted in Uncategorized. Bookmark the permalink.

19 Responses to New board committed and parts ordered for first batch!

  1. Rob M0TFO says:

    Wow thing are moving on, best get my xmas wish list to my xyl asap.


  2. Rob M0TFO says:

    Just to say a new yahoo user group
    has been created, hope you can join to discuss this great idea by Howard, G6LVB as well as keeping an eye on this Blog.

    Rob M0TFO

  3. Levent WW2L says:


    How can I order one from coming batch?

    Levent – WW2L

    • admin says:

      Hi Levent

      Thanks for your note.

      Once I’ve figured out how we’re going to do the distribution I’ll post it here.

      Bear in mind that I have three hurdles yet! (1) check out the new PCB – hopefully by Monday, (2) get production quatities of PCB on 8 day turnaround made by early November and (3) assembly on 20 working day turnaround takes us to early December.

      By far the biggest cost is assembly. By stretching it out to 20 working days the costs reduce.

      That’s if everything goes right…

      Howard (G6LVB)

  4. Ronen says:

    Is there a chance to get a schematic of the Receiver ?

    Is there a chance to get Higer Bandwith then the sound card sample ?
    I saw a SDR radio that give out its data also to the USB bus but can give a span of 30MHZ …..

    Thanks For any info
    Good Job
    Im waiting to be able to order one
    Ronen – 4Z4ZQ

    • admin says:

      Hi Ronen

      Thanks for your note.

      The FUNcube Dongle block diagram

      To keep costs down, inside the FUNcube Dongle we are using a hjighly integrated device that is subject to a non-disclosure agreement and thus I am afraid that the schematics aren’t publicly available. The key design rationale for the FUNcube Dongle was (and still is) for a unit capable of narrow band satellite reception that is accessible both in terms of cost and in terms of ease of use and setup for educational outreach, so providing publicly available schematics wasn’t top of the agenda I’m afraid. We had discussions internally within the FUNcube team about the NDA. I am pretty sure that if we hadn’t used the device requiring the NDA, there would be no FUNcube Dongle with such a wide appeal.

      Firmly in the development inbox is most definitely a wideband model. However this will mean a number of expensive changes, expensive in terms of unit cost and expensive in terms of development effort. The PIC will need to be changed to an ASIC or FPGA. Application specific cross platform USB drivers will need to be developed (non are required for the current version – thus making it easier to set up). This will not be compatible with existing host software like Rocky, SpectraVue etc so host application development will be needed. There will need to be a faster more expensive and more power hungry ADC with an associated external decimation filter. So, all in all, a lot more work needs to be done to realise a wideband device. I agree though that the application of this will be even more exciting to us techies than the narrowband FCD.

      Back down to Earth with a bump, the priority now is to move to production of the narrow band FCD, and it’s full steam ahead on that.

      Howard (G6LVB)

      • Rick Spanbauer says:

        Is the Funcube based on the Elionics E4000 tuner?

        • admin says:

          Hello Rick

          Thanks for your note. The current FCD does use the E4000, although there is no need for the end user or programmer to be concerned about that as the device configuration abstracted with an API. This allows the possibility of using different or additional devices in possible future versions.

          I will be able to provide a better answer if you let me know what you’re trying to achieve?


  5. Mike says:

    Hi Howard,

    How did testing of the LNAs go?
    Which LNA is the final one? Did you see even better NF numbers with the new addition?

    • admin says:

      I am down to two. A BGA2717 which is cheap as chips and an MGA-62563 which is about six times the price. I have made the PCB such that either will fit. The BGA2717 is simple to setup, but the MGA-62563 performs better in terms of gain, noise figure, OIP3 and P1dB. Problem is, the P1dB is greater than the tuner’s maximum ratings so I have had to put something in to limit the output. More expense… so if the BGA2717 suffices we’ll stick to that. The BGA2717 was coming out at 0.135uV for 12dB SINAD on both 2m and 70cm. The MGA-62563 was 0.122uV.

      Howard (G6LVB)

      • Mike says:

        Thanks for the information.
        Maybe the Pro version can use the MGA-62563?

        • admin says:


          It is possible, but we’ll see how testing goes. My first attempt to limit P1dB was met with disaster. The last thing I want is a bunch of FCDs returned and a QFN rework job on my hands!


          • RD3BCH says:

            0.135uV it is very close to 0.122uV, and “A BGA2717 which is cheap as chips and an MGA-62563 which is about six times the price”. BGA2717 It seems quite comprehensible. IMHO.

      • hi,
        i’ve ordered the funcube-dongle. i hope i’ll get it soon 🙂

        how much input power does the tuner allow? maybe the output of the MGA-62563 can be limited in some way.

        another question: has the funcube-dongle pro a ESD protection on the input? If not, maybe the ESD0P2RF-02LRH from Infineon is a good choice 🙂

        vy 73, OE2AIP Philipp

  6. Dave Bullock says:


    Do you have any plans of open sourcing the design? Our hacker space (Null Space Labs) would love to build some of these. We would need the BOM and Eagle files obviously. If you don’t plan on open sourcing it, how about selling it in kit form?



    • admin says:

      Hi Dave

      Thanks for your note. The primary design criteria of the FUNcube Dongle was to provide an educational outreach tool to allow access to the FUNcube satellite in a simple and reasonably priced way. As part of this we have worked hard to both keep the cost down and make it plug and play. With these axioms, we ended up using parts that have intellectual property that demand the signing of a non-disclosure agreement.

      As such, the FCD has ended up having the added benefit of being a wide band all mode receiver with many more uses than it was originally designed for and even more that we haven’t even dreamt up yet.

      We thought long and hard about using parts with associated NDAs. At the end of the day, the primary requirements were to keep costs down and provide plug and play capability. Providing schematics, firmware and detailed PCB layouts is, I am afraid, not possible under the NDA. Even without an NDA, I am not sure just how many people out there would have workshop capabilities and be confident enough to make boards with QFN parts.

      Believe me, this is scary stuff! I have built 13 so far. Even now, all are met with trepidation, although I’m getting better, the last seven have worked first time. That is not before getting rather blasé yesterday when I positioned and soldered a 0.5mm pitch 32 pin QFN at 90 degrees to true: luckily I noticed during construction of other parts. Rather a schoolboy error, but equally if you are not very careful, and without the right equipment, the whole board would have been destined for the trash.


      • Fred W0FMS says:

        I hope what the chip itself is isn’t supposed to be under the NDA. If so I’d be more careful with the pictures! :O)

        Interesting chip though, that Elonics E4000. Having all of that down to I/Q on one chip sure is a nice situation. For the life of me I don’t understand the NDA though for a tuner chip… nothing in there that isn’t generally known anyway…

        I guess that is why I’m an engineer and not a “businessman” (though I do have a MBA w/honors…)

        Fred W0FMS