Now available over in the Downloads area!
This has the following changes:
o Set sensible gain and filter defaults at plug in time (does the same as the “Defaults” button).
o Initialises the tuner and codec _after_ USB enumeration to keep within USB current draw specs.
o Implements different (hopefully better) VHF filtering.
o Implements command 101 – SetFrequencyHz with more granularity using a 32 bit unsigned with the frequency specified in Hz rather than kHz. Also returns actual frequency set.
o Removed sigma delta software imposed limitation on the fractional-N PLL. Need to see what effects this has.
o Recoded the vendor supplied PLL and filter setting code to use constant tables rather than a ton of if/then/else statements.
o Recoded PLL calculation code to use floating point.
Any pointers on how to go about the loading? When I run the bootloader program, it shows this:
HIDOpen of Vid_04d8&Pid_FB56 returns 1
USB VID/PID found.
HIDQuery returns 0
Error, query device failed
The only button I can select is “Open file”. When I select the new firmware, no other buttons become active…I’m missing something here…
Okay, fixed my own problem.
First step–use the full interface (FCHid.exe) and hit the “reset to bootloader” button first, then exit.
Second step–use bootloader program (FCHDBL.exe), then select new firmware file, and then “write firmware”
Appears to have gone very well! Verified just fine. Testing now…
There’s a new section in the Downloads section about how to use the bootloader. Not sure if you’ve seen that or not.
Shame on me! I didn’t see it…hadn’t scrolled down far enough to see it. The directions are good and clear. And the default values are better, in my hands. I swear my signal:noise is better copying local stuff on the discone. I hooked the dongle up to my normal preamp/array and AO-51 sounded great (using WRplus).
Maybe the new changes can cover the low end of the 6m band?
Sorry, this isn’t going to happen, the VCO keels over I’m afraid.
Great – thanks.
New filters and lower LNA gains plus no doubt other improvements – is command 101 active?
I think we still need to consider defaulting to lower LNA/Mixer gain at VHF. Here is rural Oxfordshire I have strong signals either side of the 2m band within 10MHz and if these (pagers?) are in any way typical, a simple duplexer will not be sufficient to reject them sufficiently. Mr Murphy will ensure that in all public demonstrations, the mixing products are on 145.800MHz.
Yes Howard, could you please make the Mixer gain default to 4dB,
I’ am using a LNA gain of +7.5dB on 2M, giving the best results with Signal/Noise ratio.
Yesterday I received a German repeater (DO0MG) on 145.612,5 Mhz on a distance of +160KM with my Diamond U5000 vertical, also received Belgium repeater with French language stations!
Also I see that the Frequency correction value is reset to 999885 every time the application starts again! It would be better that this value will be saved after changing it!
73 and a good newyears eve,
My findings remain the same with the new firmware and echo those of Mike & Frank. Best results on 2m barefoot to a vertical antenna are with LNA at 10dB or less and Mixer at 4dB. Adding an SSB SP2000 pre-amp on the bench in front of the FCD largely cures the out of band signal problems and allows the LNA gain to be reduced further.
As an extreme test, and with the SP2000 in line, the gains of the LNA and Mixer were set to their defaults of 20dB and 12dB respectively. There was little evidence of the out of band problems thus indicating the effectiveness of the SP2000’s internal filter. Clearly these settings with the pre-amp were for test purposes only and it is not suggested they are viable for actual use.
I wonder if there are any plots / specs for the various filter settings?
I’ll chime in. With the 18c firmware, the default Mixer gain of +6 is OK, if not a tad high still. And the LNA gain setting I keep going to is closer to +5 or +10 range, often lower. Improvements in the firmware are noted and appreciated!!
Silly me—I meant IF gain of +6dB (not mixer). Mixer is still default 12 dB *(thanks for getting me straight, David!)
How are you measuring best SNR? I am using a test set and the “default” button tuned for maximum sensitivity that I found from empirical measurements. If you have additional LNAs in front then the figures will be meaningless for raw sensitivity.
Apologies for the unannounced firmware release 18c. It came as a result of some discussions offline yesterday and I pushed it out hence I consider it both beta and interim, and thus has wrapped up a number of other things that I was working on at the time.
On that basis function 101 is implemented but not properly tested as yet. What I can tell you is that the old 100 function is still provided for backwards compatibility and uses the same functionality that 101 uses, and that seems OK.
Meanwhile I am indebted to you for these contributions. I guess I should start a proper forum where we can add pictures etc.
I have implemented 101 (1Hz resolution) in a new C++ builder application that listens to Kenwood style frequency settings from PowerSDR etc. – it works, compensates for the crystal error and IF offsets etc., but not quite ready yet as:
a) I am having trouble saving the com port settings to the ini file.
b) PowerSDR seems to have the spectrum inverted when set to SDR1000 mode – which is needed to enable the frequency tracking.
I am wondering how to set some of the other parameters. Specifically the gains of the amplifier stages. Will there be an API?
Congratulations on your work!
Yes there will be a documented API for the gain and filter settings too, that job is rapidly coming to the top of the pile, I would say give me a week if that’s OK.
It is fine.
Finally I got the controller to work with Power SDR 2 on Windows 7. The earlier version of PowerSDR does not seem to work properly on the new OS. I can tune up and down the bands and set an offset for the IF and Crystal drift. The only thing I have not quite got the hang of is saving com port settings.
Here it is http://www.flickr.com/photos/ad6xy/5328031979/ looking at the GB3RAL 60.05MHz beacon through a 20dB PAD. The QRM in the image is intermods from local signals which overload the FCD even with a 20dB pad in front.
I’m not sure that absolute sensitivity is the target for those of us in ‘noisy’ locations probably more relevant – selectivity. My object so far has not been to measure sensitivity or SNR but to find combinations that mitigate the severe blocking effects of nearby out of band signals which otherwise make reception of weaker signals impossible. 2m is badly hit in this area with portions of 70cm equally compromised.
With a well sited 2m repeater very nearby(145.725) it is almost impossible not to be aware of its presence when listening around 145.800Mhz up, the displayed noise floor of the FCD rising when the repeater keys up.
The reason for optimising for sensitivity is that without it, the link budget for the FUNcube satellite can’t add up. Remember that for educational outreach for which the FCD was originally designed, we are talking about using this with small ground mounted antennas looking into the sky, not chimney mounted colinears.
You are of course welcome to optimise the gain settings for your own environment, and although originally the full front end was designed to be used for internal use and finding sweet spots, I know radio amateurs love tweaking things so I released it as it was.
With the 16 bit ADCs, we have 90dB plus of dynamic range to play with. Innevitably if the sum of the incoming signals will have an effect on the noise floor. To mitigate against strong in band signals, directional antennas would seem to be a way forward.
It is also possible to look into running the ADC at 24 bits, but that will be a significant piece of work, possibly also reducing the bandwidth to 48kHz in the process as there is only so much CPU bandwidth.
By the way, I have the same in band problem here in Central London on 70cm with the TS-2000 when GB3LW comes up.