开云体育

Hello Michael,

Thank you for your message on the status of the FA-VA5. I waited some time to see if someone else responded first but I see no response. There was also QRL (which is such a time sink!) and I wanted to do some measures first but unfortunately I didn't see anybody bat me to responding to Michael's message.

This response is going to be long and detailed with the explicit desires to request responses and perhaps a discussion.
I created a new Subject line so that people can use the "mute topic" feature in groups.io to block the discussion w/o unsubscribing.
Please use the "mute topic" feature and not "report spam" because the latter will give a lot of work to the people of groups.io and we will see this reflected in the annual bill from groups.io.

Again, please argue / enhance / agree / disagree...


MARKET POSITION
Frankly, if I'd known that the stock of VA-FA5 was about to be depleted and re-stock not an automatic thing, I would have ordered myself a spare FA-VA5. For me that says how much I like the design we currently have.
I think most of us have read about the cease of production of MFJ, one of the other "antenna analyser manufacturers". I read elsewhere that "stock already run out" and without new MFJ production, I can only see the market share of the FA-VAx grow.
I know that there is also nano-VNA these days but for antenna work and a lot of other work, I much appreciate the VA-FAx.
So, I am looking forward to a FA-VA5+, or FA-VA6, or whatever...


PARTS AND COSTS
I recognize the issues with parts but wonder if a slight "overhaul" is enough to translate the design to parts that are more easily available. Note that before corona there was a huge, huge stock of "obsoleted parts" and Corona cleaned them up. After-corona, not all parts are built again, but I wonder if an overhaul would return the design to the sweet spot of device availability. I am sure I don't have all the details here: the only thing I can do is look at the PCB and "see if I recognize the devices" but that won't tell the whole story. On the other hand, without schematic I'm not sure what more I can say.


PROCESSOR
The current design leans heavy of Microchip devices: the dsPIC33 CPU, the A/D converter, an EEPROM and the clock chip / timekeeper. I think the latter two can be / should be generic devices and I think alternatives should be possible. Having made some designs with microchip devices myself I fully see the erratic supply behavior which frankly, I don't understand. I am in the semiconductor industry myself (and MicSemi is not a customer or I would not be allowed to speak) but while I do see annual market forcasts, MicSemi's behavior does not seem to match the whole market and I know of several other manufacturers who are suffering supply issues. Unfortunately, some alternatives like Atmel no longer exist.
I don't know how deep the software depends on the dsPIC architecture but alternatives exist and if delivery is more reliable there AND the software effort is not too big then I think another processor family with more reliable delivery may not be a bad idea.


DISPLAY
I like the current LCD display: cheap, low-current, good readability in the sun and in the lab (with backlight). For me, I would NOT change to a colour display because of energy, readability and cost. I don't know about future availability of LCD displays with the resolution of the current display though but if availability is not an issue, I would keep as-is.


KIT OR BUILT
I like the current "kit model" and given that this does make production easier, would suggest to keep the current approach as-is. (the cost of a fully-built device would be significantly higher and more complex)

?

BATTERIES
Here is where I would make changes, and I know I probably surprise you. I measured and a FA-VA5 typically uses about 100mA. Because of the power booster, the current goes up when the voltage goes down and I have seen as much as 180mA on 2.4 volts. One would suggest that a set of AA cells (typical capacity 2500mAh) would give 25 hours of operation but for some reason that does not happen and I see "battery low" more often than I like and always carry a second set of AA's when I use the analyser.

I would ask to consider using LiPo because this would allow one to charge the analyser before commencing work tomorrow and because the energy is simply much greater.
But there is a much more important reason to say: "go LiPo" and that is that in my experience, alkalines of infrequently-used devices "always leak". I have too many bitter experiences of some piece of test kit that I put away (too much rain, no antenna weather, recognise?) only to find that when I use the device, get the "Low battery" message, I open up and am greeted with Cauliflowers in my expensive radio because the )*^*)^* battaries have leaked.

For that reason, I always remove the batteries in my test equipment and on the FA-VAx, that is finding a screwdriver, fiddling with 4 small screws (&(& and hoping none gets lost, then putting the batteries in a place I can find them. In contrast, LiPo's can stay in.

I am explicitly not suggesting "deliver with LiPo battary" or "ship with LiPo inside" because I know that this would make shipping logistics very, very complex. Shipping companies hate, hate them and it reflects in shipping possibilities and price. Suggestion would be to ship with a battery holder for an 18650 and have the user supply the battery (which is simple, ask anyone who vapes!). The circuit would need to change to allow for this, include a charge curcuitry, a discharge watcher and the booster circuitry modified. Note that this would make the unit smaller and lighter, and probably with longer battery life.


POWER BUTTON
If one goes the 18650 route, then the power button would be a temporary pushbutton which switches the meter on and off, allows shutdown in case the battery is empty (SAFETY REQUIREMENT!), and an automatic power-off possibility. The place of the button should stay as-is since in the case I an using it is protected against accidental activation (the front buttons are not, but the power button is).


USB CONNECTOR
This one is easy: should be USB-C and allow for charging (I don't see a need for speed-charge, but who knows what's possible here). Note that I mostly use the USB to update firmware, I don't use it to connect to a PC for results read-out (sorry!) and wonder how that is with others.
And in case the 18650 battery is empty one can run the unit from the USB connector.


BLUETOOTH
I don't have much use for bluetooth but wonder how that is with others. Note that "bluetooth association" problems are rife and for that reason alone I would stay away. Also note that of you were to do bluetooth, the bluetooth would need to "look outside" and the metal case shields.

For me, no complexity, no cost, no "look outside", no bluetooth.


CASE
The case is not rolls-royce but is robust and serves the purpose. I see no reason for a "sleek design" certainly not at an increase of cost.
Note that if we'd go the 18650 route then there is much less need to open the case and the current screws more than suffice.
I see no need to weather-roof the case more than what it currently has (not much)

CONNECTOR
I am very happy with the current BNC that works all the way up to 600 MHz, is robust and easy couple / un-couple. An N connector for me would be awkward because couple and uncouple is hard and would quickly have a N->BNC adapter permanently connected to it. Quality BNC connectors are spec'ed up to several GHz so there is no reason to change that. And, one doesn't use an N connector on the HF bands and that is most often where I see the FA-VAx used.

?

USER INTERFACE
I am happy with the current "few buttons" interface but would suggest to introduce the concept of "bands" and have a way to switch up to "next higher band". The FA-VA4 is a bit awkward (just go from 3.5 to 28 MHz) and while the FA-VA5 has an "autorepeat" function, moving from 144 to 432 MHz is still awkward. In contrast, I don't see a lot of FA-VA5 work between 200-350 MHz and yet we have to work through it when we go to the other band.

By having, oh, 20 user-selectable bands one can put in all the bands that one can use and simply switch to the next "band" instead of having to skip several hundred megaherts. The firmware doesn't know the bands, just knows that there are a few and they should be kept in frequency order. And I think 20 bands would be about right: 137, 472, 1.6, 3.5, 5.3, 7.0, 10, 14, 18, 21, 28, 40, 50, 70, 144, 220, 432 and that would leave slots for other possible future allocations. And it should be possible to "switch bands off" so that a user who only uses the classic HF bands (3.5, 7.0. 14, 21, 28) doesn't need to cycle though any other.

This would bode for perhaps one additional button bit that cost is not too high and if a switch fails, you can replace it. "Doing membrane" is much harder, specially "find a replacement membrane" 15 years from now.


FEATURE REQUEST: generator
Another "small box" that I have is a signal generator that can deliver signals using calibrated signal levels: -107 dBm/1uV, -73dBm/50uV (== S9 < 30 MHz), -33 dBm and 0 dBm. I would add -93dBm because that is S9 above 30 MHz). Having semi-calibrated signal levels helps to align S-meters, do proper I/Q balancing for SDR radios and the like. And I think you already have the generator for it.


Again, this is long and detailed, and I hope to provoke some responses. And I hope I "didn't forget" once I press the send button.

Let the FA-V5 (or it's successor) live!

73, Geert Jan PE1HZG

Hello Geert Jan,

thanks for your valuable reply and all the information. May I comment some of the topics:

PROCESSOR
The dsPIC was choosen due to historic experience. Today, STM32 are valid alternatives but also still suffer from significant supply issues and high price. A change on the processor is possible but is relevant work as the code highly utilizes the DSP features.

BATTERIES
If there is a change towards LiPo, it probably would go exactly the way you describe: 18650 cell with charge controller over USB-C.

USER INTERFACE
Some like it, some hate it! Let's get more thoughts on this from other users. Band selection makes sense. I see it more as a software feature.

FEATURE REQUEST:
As a generator function is already included, is it more about defined signal levels? It would be possible to add a software attenuator to arrive at defined signal levels. But, remember that the signal is square wave. High pass / low pass filters are needed to filter out harmonics for a sine wave.
Doing this for all frequencies would require a large filter bank. Too expensive. Maybe having external filters for thoose who need such a feature?

Thanks again for your input. Let's see what other users think about the discussed topics.

Please feel free to share your thoughts. There is no right or wrong.

73 Michael, DG5MK

Dear all,
thanks for your valuale input. Let me try to bring this together and evaluate what is doable and feasable.
Thanks again for taking the time. If there are more ideas or comments, don't hesitate to post them.
Regards
Michael, DG5MK