Impressive but even with a 17 inch screen I cannot fit all stations on screen and have to scroll back before the next slot is decoded. Speed reading preferred. About 20 callsigns per screen and often there are 50 calls decoded per slot.
35 to 40 calls current decoding on 10 m at 0454 UTC.
A work in progress.
I think I’ll stick to my tablet.
Does that mean there is a good case for using a non-standard frequency, to limit the callers to SOTA chasers. Or would that be the proverbial shooting oneself in the foot. No contacts at all…
Peter’s demo of 10m was interesting. But disappointing that he got no contacts on 10m despite the strong incoming signals. I guess being 13 db behind the 100w level does take the signal down a lot if the 100w signals are already at -10. -23 should be ok though?
The software seemed to run pretty well in auto mode, surprising even Peter.
He makes good videos these days.
73 Andrew VK1DA/VK2DA
I suspect, as with all digital modes, adding an actual (ideally CAT control and audio - but at least audio) interface between the phone and the radio would improve operational reliability- a 4 pole 3.5mm jack to ACC socket cable works wonders.
Agreed. The advantage of Peter’s “universal interface”, ie. acoustic coupling, is that it is equally compatible (and equally appalling) with all radios and all types of phone or tablet. I might even try it myself. I will blame him for getting me into ft8 if that works.
I used the word appalling because that is the very descriptive word Peter used in his review of the tiny cw transceiver kit popularised a few years ago, $8-10 from your nearest Aliexpress website.
I would guess that the extraneous noise from the surrounds would degrade the system capability quite a bit. But only a test would confirm that.
Terminating TRRS connectors and microscopic wires is not for those of us with poor close-up eyesight (anyone over about 45, but at over 70 I definitely qualify). I generally use 2.5 mag glasses for computer and books, but have some 3.0 for soldering work, might even need to upscale that soon. I recently had to repair a 3.5mm cable from my keyer and found it very difficult as my gas powered soldering iron was just too hot even turned down to minimum flame. On a field location that’s all you can do.
73 Andrew VK1DA/VK2DA.
If you can’t solder up connectors, just use a “Gosport Tube” instead.
I simply use old earphone leads - (I seem to go through them fairly quickly) - cut the earbuds and button off just leaving the TRRS plug and cable. Use a multimeter to find which of the wires go which of the 4 connectors on the plug and wire those to an appropriate plug for the ACC socket of the radio in use. Some would say, isolation through a transistor or transformer is needed but with my portable rigs (Xiegu G90 and G106) this has not been necessary.
By the way, there has been an alternative program called FT8 Radio (in the Play store) that has enabled basic FT8 from a Smartphone to a radio for some time. I think that one costs $5 or something but it works as well either with a cable or direct via audio in the air. Correct timing on the phone is everything for both of these apps.
Well dxpeditioners like FT8WW use non standard frequencies and get tens of thousands of contacts. However it would be poor form to do this as a matter of course. The RTTY gang are not best pleased about having 40 intruders on their patch for instance. Using the FT4 sub band for FT8 might be OK from time to time if there is no digital contest.
I think it would be possible to develop the phone app to allow having split screens that could be flipped from one to the other. So one could see screens spilt horizontally and vertically.
Once you start calling someone or they start calling you the other stations are just noise which you don’t need to see.
If the programmers are not regular FT8 users their product will always be a bit flaky.
Still give it a try. You might then want to use the real deal with a tablet, etc.
And another thing Andrew. Peter does a good job however I have worked 1,400 km on that appalling pixie.
Don’t expect the dB to add up as in a lab experiment
I average -8 dB on my reports received compared to the reports sent. I run 70 W in the suburbs. Whenever I suggest generous kW outputs are being used I get howled down with shouts of " Nooo. Its their noise level". I think I am on a Porcine air route as they are often heard overhead these days.
Conversely I have completed contacts with some stations who give me -23 dB. They are typically 10 to 20 dB stronger with me. I have seen 30 dB in favour of the other guy often. Occasionally I get 10 dB In my favour. Propagation or QRP?.
These days finding a frequency on 10 m that is clear in your segment is a triumph.
It’s certainly not set and forget.
After my activation on cw and ssb on Mt Gillamatong today I checked 21074 out of curiosity and found a lot of signals, adding up to about s7. The app loaded on my Samsung logging tablet, but failed to reveal any waterfall display let alone a list of callsigns heard. Using a tablet in bright sunlight (it was about 29C) is not a good formula for success.
I looked at the app permissions and it had access to microphone so presumably should have been working. But with no waterfall display at all, I clearly had not waved the wand in just the right way.
73 Andrew VK1DA/VK2DA
I set up with mic and qth accesses allowed when app was in use. Entered my call and home grid square ref. Set up for different Tx/Rx frequencies.
I still needed to do a bit of experimental button pushing and associated cussing but have succeeded in completing an FT8 QSO with BA7LIP on 12 m using my Android phone and the FT8CN app. It would be nice if the help files were in English, but if I can figure it out using the approach of a 7 yo then most ops will be able to get it going.
The small screen problem can be reduced by having only decodes in a portrait alignment. That means giving up the dual spectrum and decode display.
The spectrum looks nice with the noise reduction on. Appending the callsign on the spectra is nice. The text display is rich in information and the colors do not detract from the readability.
I think the audio bandwidth could be a bit wider, but your RX bandwidth might not support that. I’ve been tuning 250 Hz low and 1 kHz to get some stations in the Passband anyway.
So if I’m really in need of some FT8 QSOs on a summit and didn’t bring my bag of tricks, I can use my Android phone.
Thanks to Geoff VK3SQ for his original post. Also to Andrew VK1DA whose posts caused me to try it.
Now to go do something useful
BTW I overlooked an acknowledgement of your local airline, Porcine Airways. I just hope they are lined up every morning ready to fly, as it says in the system implementation handbook.
73 Andrew VK1DA/VK2DA
You should join their rewards program, Frequent Trotters, but never take an afternoon flight as the planes have been sitting on the tarmac, just bacon.and passengers develop rashes.
Never litter while on board. It upsets the cabin attendants.
Hi Andrew - just a thought - what are you controlling the time on the tablet with? GPS? Internet link to a Network time server? FT8 is critical on accurate computer time (as the crew on Bouvet found out!).
This all looks very interesting.
I have done FT8 from a summit with a tablet using a different app (FT8Radio) and I used a SignalinkUSB for connection between FT817 and tablet.
I’m keen to try this alternate app, maybe next weekend.
For anyone not familiar with FT8 I’d suggest setting up a PC with WSJT-X in the comfort of your shack, get used to the format and methods for QSOs, get very familiar with setting the time on the device, and then go try it in the field with a phone and an app.
Like comedy - timing is . . . . . . . everything!
Also it’s wise to consider the duty cycle, power required for data contacts and the heating effect on your final RF amplifier. FT8 is 15 seconds TX, 15 seconds RX for a minimum of 4 cycles per QSO.
For FT8 I ran my FT817 on the second highest power setting, with the LCD meter function on MOD and audio level adjusted for 5 dots showing. This worked for 5 QSOs and my finals are still good - YMMV.
Yes I wondered how well the time was being controlled. It depends on how well the link via the phone works to align the tablet with phone network time.
But I was nowhere near that problem as I had no waterfall display.
In some chats with Ron I have found how to get it to display a waterfall. So the time issue could be next.
73 Andrew VK1DA/VK2DA
I think the power is well within spec if you use an ft817. They are designed to withstand fm on repeaters. 3 minute overs at full power! Just before repeater timeout!
On the kx3 I would use the power control to keep it under control. Acoustic coupling inevitably makes the waveforms messy so it won’t reach full power.
A TRRS cable becomes even more important if this becomes a habit. I’ll resist that!
I was using the Samsung tablet that I use for vk portalog. It’s with me on nearly every activation so no extra hardware required.
FT8 is 12.64 second key down in every 30 seconds. So at a 70 % power setting you are down to the magic 30% duty cycle manufacturers used to stress. Some rigs such as the KX2 might be lacking in thermal mass and need about more power reduction. The FT817 has enough mass but keep it in the shade and drop to 2.5 W if it feels hot.
Using accoustic coupling for Tx is not ideal but you can with judicious placement of the mic and phone get full power as set. The program has a noise reduction feature that helps combat wind noise etc for received signals
FT8CN is now a properly functional product, made possible by professional CN people. It seems to do a single pass decode so unlike the real WSJTX, no recovery of signals buried under a stronger signal and no deep scans. This can mean up to 20% of stations are missed on a crowded band. I base this on my receiving test comparisons.
It won’t replace WSJTX even if I were to run it on a Tablet for the bigger screen. See my initial comments about the number of viewable decodes.
But for a no extra gear, self contained product I now score it as creditable for portable ops.
There is no problem if you can’t get network time sync on your phone.
The App has the ability to change the clock time. Just listen on 21.074 USB and run the program. If the average time offset exceeds half a second you can make a time change. WSJTX copes with up to 2 seconds error but keeping ime sync to within half a second seems to improve performance. I presume the same applies to all derivatives.
I tried FT8CN yesterday on an activation with my FT817 and it worked well. The only bug/feature I found was that I couldn’t call CQ if others were calling CQ - it automatically replied to their CQ calls instead and I couldn’t find a way to override it.
I used my basic TRRS to DIN cable that I use with my windows laptop or tablet but I couldn’t get my USB CAT able to work on Android so I had to use the mic to control the PTT which was a bit tricky.
Has anyone successfully tried using a Bluetooth CAT control dongle on Android? The FT8CN app gives “bluetooth” as a method for controlling the PTT.