Feedback for version 0.9a here

Information and discussions on Spectrum Analysis
SteveAndrew
Posts: 213
Joined: Thu Nov 16, 2017 11:06 pm
Location: Gold Coast, QLD, Australia

Re: Feedback for version 0.9a here

Post by SteveAndrew » Wed Aug 01, 2018 10:59 pm

Hi everyone

Sorry for the late response, I've had some problems with my PC, hopefully things are back to normal now.

Dave, your AS9851 looks like a very sick puppy to me. Good to know it turned out to be a clock problem, and not related to code. Interestingly, your AD9850 module is the same physical format as my AD9851, and they both work when plugged in with my original code. Perhaps Don will be able to confirm which type of module he is using once he gets back from his camping trip.

I'm still waiting for my AS9850 module to arrive, it's the same type as you have Dave, so I'm not expecting any problems with it.

In the meantime, anybody using the AD9850 or AD9851 that is not the same format as Dave's AD9850 might want to check that on their own version, D2 is connected to ground, and D0 and D1 are connected to +5V via pull-up resistors. This is done to ensure that the IC starts up in serial load mode. Perhaps some modules do not include this and they may need to be modified. I'll look into this a bit more, there may be a software solution. The only reason I haven't looked at this before is because the module I used, worked first time, and I had no reason to investigate further.

We'll have to wait for Don to get back and let us know what type of module he's using, and for my AD9850 to arrive.

Reason: No reason

K9AQ
Posts: 37
Joined: Sat Jan 14, 2017 1:20 am

Re: Feedback for version 0.9a here

Post by K9AQ » Thu Aug 02, 2018 11:14 am

Steve,

I am going to be home today but only overnight, then we are leaving again for another two weeks. I will try to send you a photo of my module.

To make life much easier for you, maybe we can identify one module, from on source that reliably works, and then just support that module. This seems to be taking a lot of your time, which could be better spend on the spectrum analyzer software.

I really need the ability to scan the entire HF spectrum much more than I need the tracking generator, as I can to what I need to do with a noise generator.

Just my thoughts.

Don

Reason: No reason

DaveR
Posts: 25
Joined: Fri Jul 27, 2018 1:07 am

Re: Feedback for version 0.9a here

Post by DaveR » Thu Aug 02, 2018 12:51 pm

maybe we can identify one module, from on source that reliably works
How do you propose that we do that, Don? Buy a dozen over a few months from each of several suppliers and see who has the highest "working" rate? Steve's software already works, and the modules all appear to be the same with just minor layout differences, so they don't need individual support - it's just down to the user's luck whether they receive a good one or a bad one (as with any cheap gizmo from China). I'm pretty sure that most of the module sellers on ebay, Alibaba, Banggood or wherever (the Chinese "Wild West") are the same bunch of people, and most of them sell under multiple accounts and only act as agents, passing on orders to one or two central suppliers for dispatch. Finding a "reliable" source on which to base the development will be next to impossible.

Even though I've got a 60MHz signal generator with which I can do sweeps, I love the idea of Steve's integrated tracking generator for doing RF filter design and testing, so I've ordered another AD9851 module - this time a more expensive one, on a newer board with SMA connectors - but I'm still aware that it may come with a dodgy chip fitted, in which case I'll just have to get a refund and try another one. In the worst case, I might have to resort to buying a genuine chip from Mouser or Digikey and replacing the one on the module myself :)

Regards,
Dave

Reason: No reason

K9AQ
Posts: 37
Joined: Sat Jan 14, 2017 1:20 am

Re: Feedback for version 0.9a here

Post by K9AQ » Thu Aug 02, 2018 10:59 pm

I got home from a month long trip to find that I received a package from China. It was supposed to contain an AD9850 and an AD9851 DDS, but they only shipped me an AD9851. It looks just like the one that I already have. I contacted the seller and hopefully in another month I will get the AD9851. I don't have time before I leave again for another couple week trip to test it. I will test both of them again when I get back.

Don

Reason: No reason

SteveAndrew
Posts: 213
Joined: Thu Nov 16, 2017 11:06 pm
Location: Gold Coast, QLD, Australia

Re: Feedback for version 0.9a here

Post by SteveAndrew » Thu Aug 02, 2018 11:51 pm

K9AQ wrote:I am going to be home today but only overnight, then we are leaving again for another two weeks. I will try to send you a photo of my module.
To make life much easier for you, maybe we can identify one module, from on source that reliably works, and then just support that module. This seems to be taking a lot of your time, which could be better spend on the spectrum analyzer software.
I really need the ability to scan the entire HF spectrum much more than I need the tracking generator, as I can to what I need to do with a noise generator.
The two modules that Dave is using seem to be the types most commonly available on Ebay. I managed to track down schematics for both of them. Both of them have D0 and D1 tied to VCC via a resistor as required for serial mode programming. In the case of the board with jumpers on it, jumper J1 must be in place to enable serial mode. As to trying to identify a single supplier - They seem to come and go, and so many of the suppliers seem to get their stock from the same source. Don't worry about the time I'm spending on this. Research is never wasted, and I'm determined to get to the bottom of what appears to be a rather odd problem.

I've had a look at what other schematics I can find, in some cases there seems to be no provision made to enable serial mode input by tying D0 and D1 to VCC via 10K resistors - something to check on your module ?

These zip files contain schematics and board information I've managed to find - They might be of some help.
DDS_Modules_3.zip
(185.04 KiB) Downloaded 937 times
DDS_Modules_2.zip
(110.06 KiB) Downloaded 944 times
DDS_Modules_1.zip
(298.04 KiB) Downloaded 949 times
I just saw your latest post come in as I was typing this - Just have to love some of the suppliers - I'm still waiting for my AD9850 module to arrive. Now a little something that might help cheer you up. Work continues on the analyser, and wide-band sweeps are now working to beyond 500MHz widths, without DC spikes. No release date yet, work continues.

Hope your trip was a good one, enjoy the rest of it

Reason: No reason

SteveAndrew
Posts: 213
Joined: Thu Nov 16, 2017 11:06 pm
Location: Gold Coast, QLD, Australia

Re: Feedback for version 0.9a here

Post by SteveAndrew » Fri Aug 03, 2018 12:22 am

Something else I came across - Incorrectly labeled pins. D7 is being used as the data input pin, so this would certainly cause a problem.

https://softsolder.com/2017/05/18/gener ... -gnd-pins/

Assumption is the mother of all <insert-suitable- epithet-here> :o

Reason: No reason

WA3RGQ
Posts: 3
Joined: Wed Jul 25, 2018 11:05 am

Re: Feedback for version 0.9a here

Post by WA3RGQ » Fri Aug 03, 2018 12:11 pm

When the overload indicator comes on, my software locks up.
Last edited by WA3RGQ on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

K9AQ
Posts: 37
Joined: Sat Jan 14, 2017 1:20 am

Re: Feedback for version 0.9a here

Post by K9AQ » Fri Aug 03, 2018 1:00 pm

Steve,

Sounds like you are making great progress! Can't wait for the release.

Don
K9AQ

Reason: No reason

SteveAndrew
Posts: 213
Joined: Thu Nov 16, 2017 11:06 pm
Location: Gold Coast, QLD, Australia

Re: Feedback for version 0.9a here

Post by SteveAndrew » Fri Aug 03, 2018 8:49 pm

WA3RGQ wrote:When the overload indicator comes on, my software locks up.
There is a problem with the overload indicator, in that it doesn't always clear once the input level has been reduced, usually by increasing the LNA gain reduction. This is the first time I've heard that it caused the analyser to lock up. Does the software lock up every time you get an overload ?

In the meantime, the overload indicator problem has been addressed for the next release, and now clears once the overload condition is rectified.

Reason: No reason

DaveR
Posts: 25
Joined: Fri Jul 27, 2018 1:07 am

Re: Feedback for version 0.9a here

Post by DaveR » Fri Aug 03, 2018 10:32 pm

In the case of the board with jumpers on it, jumper J1 must be in place to enable serial mode.
Hi Steve - in the case of my AD9851 it doesn't appear to make any difference whether J1 is on or off, as it always starts up in serial mode regardless of the jumper. The jumper does actually short D2 to ground as required for serial mode (according to the datasheet) but the chip acts as if it has an override switch which ignores the jumper setting. It could, of course, be another "undocumented feature" of the fake chip on my board, and may not be the case with "good" boards of the same type.
This is the first time I've heard that it caused the analyser to lock up. Does the software lock up every time you get an overload ?
Actually this was something I found happened sometimes as well, usually in the middle of waiting for a sweep trace to complete, with Trackgen running. The easy way to tell that it had frozen was that the overload indicator had stopped flashing and was permanently on, then I'd have to shut down the program from the taskbar because it would be completely unresponsive to all direct mouse clicks. I didn't pay any attention to the freezes at the time, though, as I was too distracted trying to work out was wrong with the AD9851 (for which I've now received a full refund, by the way!)

Regards,
Dave

Reason: No reason

Post Reply