Use this space to cheer the creator along, and talk to your fellow backers.
Have a question?
Yes, Seeed is actively seeking resellers world-wide. They are also selling on Amazon now.
do you know this offer?
If you want to sell something in the European Union, this must have a CE marking. And in Germany a german manual is required. Both have been missing but which in my case as a radio amateur should not be a problem. In order to know this, the ladies and gentlemen had to be well educated. It was not danei technical problems! Only Bureaucrats. :D
Interesting. I was wondering when this might happen. Lack of FCC, CE, RoHS, … compliance is sort of the dark secret of the low-end SDR market (and hundreds of other devices as well). We haven’t checked carefully but wouldn’t be surprised if many, if not most, of the low-end SDRs lack basic compliance. We have plans to obtain our approvals, but it is a slow and expensive process. If done incorrectly we could spend more money than all the profit left over from the Kickstarter and current sales. We’re hoping Seeed can helps us as they do their own in-house pre-compliance testing so we hopefully won’t spend as much time at the expensive test range. We’re a little bit ahead of the game. The BBG has full approvals of course. And the Kiwi is specifically designed for low emissions (in or out) so it doesn’t interfere with itself.
Hi Valent. Thanks for your answer.
The matter has been clarified. For the import, the device requires a CE marking, unless you are a radio amateur. This was just ignored and I had to insert. Bureaucrats! :D
Greets from Germany
Hi Ralph. Please email us (see kiwisdr.com/ship for email address). We'd like to know more about what the problems are. We had another customer is Germany who had customs ask him a few questions. But the unit was eventually allowed entry. All the other units sent by DHL were delivered.
We will get a unit to you. Seeed has a dealer in Germany now and if necessary we will have one shipped to you from them (at our cost of course).
Still no good news for me. I'm still fighting with the german Federal Network Agency and the customs office on importation. Don't know how this ends. It's possible I had so send it back.
Greets from Germany
Thanks for the prompt replies. I had measured at the pin labeled "5V" originally. I just checked at the "5VE" pin and measure 4.8V. I then re-measured the "5V" pin and get 4.72 V this time. I will keep on eye on this if I get unexpected shutdowns or other problems.
I will contact the support address for the admin problem.
4.67 might be marginal. My Jameco measures 4.8 on the Kiwi header pin labeled "5VE", which is the Kiwi input 5V, after filtering, being delivered to the Beagle. The "5V" pin measures 4.75 and is the 5V returned to the Kiwi after going through the on/off switching of the Beagle power controller IC.
Because 5V is only used as input to lower voltage regulators on the Beagle and Kiwi (e.g. 3.3V, 1.8, 1.5) it doesn't matter too much that it's not exactly 5V. The "3.3" pin on my Kiwi measures 3.39V which is fine. But at some point the power controller IC is going to be unhappy and do an under-voltage shutdown. If your Kiwi does random power downs this is most likely the cause. I don't know what the under-voltage threshold is. It is software programmable which makes it difficult to determine. I have a precision hp power supply I could use to characterize this, but it is currently broken unfortunately. Maybe this could be a project for someone..
Hi Chuck. Could you please contact us at firstname.lastname@example.org ? A number of people had trouble getting to the admin page despite being on a local network. We think this problem has to do with IPv6 overlay networks. We've added some debugging prints in the latest release, but everyone who had this problem reports it had "gone away" before being able to tell us the output of the debugging. So we still need more info about this problem. Thank you.
I am using the Jamesco 5V 1amp supply suggested in the quickstart guide. I measure about 4.67V on the Kiwi board. Is this good enough?
I just powered up and am stumbling around. I have not been able to connect to the admin page from the same browser I used to connect to the SDR. The SDR connection works, but an attempt to "browse" to kiwisdr.local:8073/admin produces a "KiwiSDR: software-defined receiver Try again. Password:" and a box for password entry.
I understood that when connecting to admin from my local network, no password is required. I was concerned about shutting down safely so I used the ssh login and the kd command for kiwi down, which hopefully was safe enough. I found that after powering back up I had to use the ku command to get the kiwi back up again. I hope there is some easy way to get access to the admin page login for easy shutdown.
Yes, the board length is a problem, no doubt. We were very worried about self-interference, either from the ADC digital outputs, FPGA or Beagle when we did the first layout. So the analog front-ends were moved a bit further away even though this meant having a non-standard length Beagle cape (1/r^^2 and all that). Same reason why the HF and GPS front-ends are in separate cans. Sure enough, when our prototype was first run the noise was terrible but we couldn't conclusively prove this wasn't our antenna and/or ground system. It wasn't until a beta test unit went out to the fantastically quiet SK3W contest site 150 km north of Stockholm that we saw how good the noise floor could be.
It's possible we might try a PCB shrink at some point. There are a few hardware changes we'd like to try. But this is pretty far down on the priority list.
Just read Project Update #21, and have received my KiwiSDR about a week ago with the "slow" Singapore Post shipping to the USA. S/N 1280 if it helps. I'm in NY state.
Wish the form factor of the KiwiSDR was shorter. I've been looking at extruded aluminum cases, and it's a little too long to fit in a reasonably sized case. Best I've found so far is a
BUD EX-4502 that is 6.48 inches end plate to end plate internally. It is about $24 US though!
Pricey! If I use it, I plan on cutting the length shorter.
If the two shield cans are rotated 90 degrees and one is placed on the bottom of the board, you can shorten the board a bit.
73, Jerry WA2RKN
Thanks for a great project.
My KiwiSDR arrived this week, now running fine. Good instructions and a very simple assembly/software installation process. It runs well on a long wire antenna and a $3.50 active GPS antenna from:
Currently using a switching-mode (iPhone) power supply, but my next step will be to put together a transformer-based supply to get rid of the hash around 100/200kHz.
Thanks! Somehow in reading it through about 4 times I managed to completely miss that bullet point over and over. I found a 5v power supply with the correct barrel plug and it works perfectly, aside from some psu noise.. I did see the section about noise from poor quality PSUs so that isn't a surprise.
Hi Troy. Powering details are on page three of printed instructions included with your unit (also here: http://kiwisdr.com/quickstart/quickstart.pdf) Apply +5V ONLY, center pin positive, >= 1A (preferably 1.5A) to the 2.1mm jack on the Kiwi board. The Kiwi powers the Beagle over the header connectors. Do not power the Beagle via its DC jack / mini-USB (BBB) or micro-USB (BBG). Only power with +5V, not higher.
Hi, silly question here... How do I power the KiwiSDR? There is some advice about powersupplies on the web page, but I can't find a simple 'where do I plug the power supply in' statement anywhere. Can I power it from one of the two USB ports on the beaglebone or do I need to use the coaxial power socket on the cape itself? Is there an on-board regulator of any sort? If I plug into that coax power port with a power supply >5V will it destroy the cape? I.e. 8v, 9v, etc? I have a handy 9v wallwart here that will fit it, but I'm reluctant to try.. Also, is it centre post positive?
The bug / wish list is now on kiwisdr.com/bugs
Please send us your additions and corrections. Thanks!
Hi Charlie. I sent you a PM with another link to the tracking info. Your unit has arrived in the US. I don't think it's a problem Seeed begins selling units before you receive your KS unit. I mean, I've been selling pre-order units for months. It's a question what order you receive them. I didn't ship any pre-orders before the KS units went out. I wouldn't expect any Seeed customers to receive units before what we've already sent. We will be glad to be out of the shipping business. It has stopped development dead in its tracks except for some little stuff we've managed to get out in the last month. Not enough hours in the day, that's for sure. :)
Will I receive my stuff before they start selling it openly? I thought that was part of the incentive to back it early.. how can I find the status of my stuff? Trying to be patient but eager to play!
Got my radio today. Promised for November 2016, delivered at the beginning of October. First Kickstarter I ever got early (and I've backed a lot of projects). Congratulations, and thanks for running your project so competently. Got it all set up (I'll second the point about the rivets) and will be adding it to sdr.hu once I'm sure I've shaken any bugs out of my setup.
Hi Philip. Yes, apologies for the enclosure choice. It was a compromise. It was the only off-the-shelf Beagle enclosure we found that fit the extra length of the KiwiSDR PCB. A custom enclosure would have added many months to the schedule and driven up the cost. We tried to get Seeed to pre-assemble the enclosures, but they were smart enough to see the folly in that, lol. So a little pain spread all around I'm afraid. Why is the PCB longer than a standard Beagle cape? That's a separate, interesting story.
Got my KiwiSDR Kit today. Assembled the boards into the supplied case. Will give it a go once I have set up the aerial and power supply. Also - I don't want to see any of those little black plastic rivets ever again LOL! You'll know what I mean when you assemble the case...
Hi Brent. Yes, there are some recommendation about antennas, power supplies and other topics at: http://kiwisdr.com/quickstart
Any recommendations for an antenna?
Thanks for mentioning this extremely important issue. Yes, fixing the audio lag is at the top of the list of "difficult problems that must be solved". It is an interesting technical story and I want to write about it in detail (including explaining why WebSDR works so well). But I got absolutely crushed today with emails and problems after the shipping surveys went out. So please give me a few days to recover from that so I can give this topic the attention it deserves. Thank you.
I'm glad to see KiwiSDR and software progressing, new featured being added, however I would like to know if there are plans to address the delays between UI and audio. The audio lags 3-5 seconds behind of what is shown on the waterfall. The lag is always present, regardless of the server, client computer, or location, so it's not a problem with single node, client hardware, or network connection. The lag between UI actions and result in audio is also in that range. Compared to WebSDR, use of OpenWebSDR is a frustrating exercise in patience.
It's too early for a framework spec. For now we're just making this up as we go. Writing the first set of extensions should help determine what's required. I'm sure it will be an iterative process. We picked WSPR because an open source C version of the decoder was available and it was going to be a moderately complex extension that would expose a lot of design issues.
The WSPR decoder and plug-in extension framework are very interesting developments.
Where can we find the decoder source code and (draft) extension framework specification?
Sorry for the delayed reply. I've been unbelievably busy. Lots of important news coming soon.
Well, technically the Nyquist limit given the current 66.6666 MHz ADC clock is 33.3333 MHz. But in practice you might run into trouble trying to push reception to this limit. We're already over-clocking the ADC slightly (2.6%) from its rated 65 MHz (we used the 66.6666 XO clock part because it was a standard value available in quantity with good characteristics and a low price). The other issue is the front-end "30 MHz" LPF (7-pole Chebyshev). The design has been modeled but the actual circuit never swept (yet another item on the "todo" list). So it is unknown exactly what the attenuation from 30 - 32 MHz would be.
But I have changed the upper limit to 32 MHz on the New Zealand receiver (http://kiwisdr.com:8073) and the software handles it at least.
Hi, would it be possible to extend the reception with a software setting to let's say 32 MHz ? I plan to use one KiwiSDR with 2 UHF converters and it would help me if there was a setting so I can shift 6 MHz from UHF to 25-31 MHz. Any chance for that ? Thanks, Ben
The biggest problem is finding open-source decoding software. I got WSPR to work mostly because Steve, K9AN, had recently done a C language port of the original WSPR code, some of which was still written in Fortran! There are some existing free (as in no-fee) decoding programs, but virtually none of them publish their source code.
would be interested to see the software roadmap for adding decode options for other modes - eg, tune to the digital portion of a band, and select PSK31 for example and see the decodes right there without having to use a virtual audio cable to push it to another application.
Would love to see WSPR, JT9, PSK 31/63. RTTY and even have the same for CW decodes as well!
Jerry: Reception of ELF/SLF/ULF (3 Hz - 3 kHz) is a problem for lots of reasons. Let's look at a practical example. Here's an early afternoon screenshot of the NZ Kiwi at 0-29 kHz: kiwisdr.com/KiwiSDR/ELF.png Notice how only the strongest lightning strikes (horizontal lines) make it down to 5 kHz. The Kiwi front-end itself is DC coupled but will be limited by the low-end frequency response of the coupling capacitors (470 nF & 1 nF in parallel, see schematics). The bump at around 7 kHz is probably caused by the coupling transformer I use in my active antenna to reduce noise (Mini-Circuits T-622). The transformer response almost certainly takes a dive below this frequency. The next problem is that the Schumann resonances (SR, 7-34 Hz) don't occupy enough waterfall bins (pixels), even at the highest zoom factor, to be visible. In the image the passband (yellow outline on frequency scale) is set to 30 Hz and is way off to the left. The increase in noise below about 1 kHz is probably due to the noise pedestal surrounding the large DC offset at 0 Hz (the first couple of FFT bins are always zeroed for this reason).
So these are some of the problems with the current design. Could they be overcome? Maybe. There are certainly lots of examples of people seeing SR using a sound card and E- and H-field antennas known to have the required frequency response, e.g. http://www.vlf.it/Schumann/schumann.htm Either the existing zoom limit would have to be increased or a separate higher-resolution FFT display operating on just the audio channel spectrum.
I notice the low end coverage is specified as 10 KHz, but the KiwiSDR demo site (University of Victoria, Canada) tunes to 0.01 KHZ. Any reason it can't be made to go to 7 Hz (Schumann resonance 7.83 Hz) on the low end? I've backed the project because this is the first one I've seen that covers VLF.
Louis: The "kit" reward includes an existing enclosure Seeed sells that happens to fit the KiwiSDR / BBG combination (not easy because most BB enclosures can't accommodate the extra length of the KiwiSDR board). However we are looking into other custom solutions. Difficult because of the time and cost involved.
Any plans for a case? Perhaps plans for a 3D printed one?
@Creator I'm very glad to see that you managed to reach an agreement with András. Congratulations and good luck with further development of the project!
Hi Mike. Thanks for your considered comments, as always. I'm not unhappy about anything. My attitude is pretty pragmatic. If the venture doesn't make any money then it doesn't matter. If it makes a lot then there's plenty to go around and everyone's happy. If it's somewhere in the middle then I personally don't care -- I'm here because I want to see these SDRs available around the world so I can listen and run some interesting experiments. This is not the place to be fighting the Open Source monetization wars. And especially not now so close to the end of the Kickstarter.
On a positive note: Yesterday I had an email from a guy who wants to take the PCB fabrication documentation we publish on Github and go build 10 boards for himself immediately. He wants to use the KiwiSDR as a platform for teaching student interns about engineering this summer (our Kickstarter obviously couldn't guarantee delivery for his timeframe). Now this is the sort of positive effect I like seeing from Open Source licensing.
Less than two days to go guys! Well done!
And I'm still waiting for the author of OpenWebRX to come to his senses - From the license on OpenWebRX's GitHub repo:
"Developers that use our General Public Licenses protect your rights with two steps: (1) assert copyright on the software, and (2) offer you this License which gives you legal permission to copy, distribute and/or modify the software."
How can he go round granting everybody the irrevocable rights to use your work, then when somebody actually tries to use it as a component in a commercial context post disparaging remarks about them (http://blog.sdr.hu/), and then try to extract a commercial arrangement?
If OpenWebRX isn't Open Source it should be re-licensed immediately (none of this dual-license arrangements, where they get to pick which license applies).
If it is Open Source he should publicly apologize immediately for all the confusion he has caused (not to mention the lost funding), think about taking down the blog posts and move on.
I'm just glad that the license protects us from things like this - as long as you keep to the terms it is irrevocable. If he is worried about the project forking, his best option is to make his fork do all this and more, so the KiwiSDR fork is no longer needed.
If this was my Kickstarter I would be very, very unhappy. I wouldn't have put up with it at all.
Kai: Thanks for your comments. I have answered each and every comment, both here and on other sites, as promptly and completely as possible. If you think there has been some sort of misinformation please let me know your specific concern and I'll answer it. The only thing I'm not going to do is spend time presenting the disagreement from my point of view. That's not productive at this point. András states his position very eloquently on his blog and I have encouraged people to read it and think about the context. I've stated from the beginning that I agree with a lot of it. That's why we're developing an agreement to share profits and more forward as quickly as possible. Remember that this isn't just about Kickstarter. Most of the profits will result from running the business long after the Kickstarter is over. So we have to spend a little time discussing the business strategy to set the expectations properly.
I encourage people to run a small business at least once during their lives (and a Kickstarter too I suppose). You'll have a whole new appreciation for the issues faced by millions everyday.
Everything I have read from you in the past few weeks was was neither binding
nor a clear statement about the disagreement between Andraz and you. Still the
same, we're negotiating ... this won't take soo long time if you really interested
in finding a proper solution and agreement between both parties.
Guess it's finally time to get a clear statement from you now, as this would stop
leaving some of your backers in the dark and gives all a good feeling about this
I'm on the verge of canceling my order due to the misinformation from your side.
hope to get some clear answers.
best and thank you for a timely reaction - Kai
Still working on it. It's only been a little over a week since a verbal agreement was reached. Written agreements and checks by attorneys always take forever. You can be certain we'll let everyone know when we're finished. Thanks.
Did the disagreement get resolved? I did not see a post to by both people involved that indicated this.
Here is a good take on Kenya's approach to the internet - https://en.wikipedia.org/wiki/TEAMS_%28cable_system%29 . In one word : aggressive. You will get pretty decent streams off there, I have friends who did telecoms can and set-up a stations.
Tisham: No worries. Kenya would be a great place for a KiwiSDR. I haven't the slightest idea what the internet access situation is in greater Africa. Maybe someone can comment. I know there is a guy in Namibia who contributes to the blitzortung.org lightning detection network. One of the 9 beta test boards will probably go to a guy in Johannesburg who already runs a really nice 8-band WebSDR site.
Sorry guys had to downgrade my pledge to $20 level due to other commitments. At least you are over the line. Hoping to send an SDR to Kenya. Any interested parties ?
If we don't get to it first, adding IQ output would make an excellent project for someone looking to understand the code and do some modifications.
Makes sense. I figured that was probably the case. Plenty of other SDRs out there for that use case.