Page 7 of 21

Re: Feedback for version 0.9a here

Posted: Sun Jun 17, 2018 1:59 am
by SteveAndrew
Kees_01 wrote:Some additional information. With a still uncoupled, not connected RSP I launched your SAS s.a. app...
Firstly, I suggest you try that test again, with the RSP connected and see if you still get the same results.
Kees_01 wrote:I selected the right Com port and the red SQUARE turned green immediately, the Enable button "enabled" and the message "Trackgen V1.0" appeared in the textbox.
When a comms port is selected, the analyser sends "XXX" and then sends "VERSION". Tracken responds (if present) by sending "Trackgen Vx.x". The anayser checks for the presence of "Trackgen" in the the returned string, and if found, activates the tracking system. This action is repeated every time a comms port is selected.The "XXX" string is initially sent due to a problem with the Microsft comms driver and is ignored by Trackgen. Microsoft mentions that the comms receive function does not always generate an event due to stream buffering. I've had problems with this in the past and found that initially sending some dummy data seems to clear this problem.

The "Enable" button activates the buttons you have mentioned, The "Spot" button tells Trackgen to generate a single spot frequency. The "Start" button becomes active if the "Track" button is pressed, and is used to tell Trackgen to start the generator sweeping function. The "Calibrate" button is not currently used - See release notes.

Using the test software,it seems that the handshaking between the analyser and Trackgen is working. I cannot see why the Trackgen firmware should not respond in the same way. The code I posted is the same as used in the firmware. I suggest you try the firmware again, using the Arduino terminal window. I've downloaded Xloader and used it to load the Trackgen HEX file. Everything works as expected, both with the Arduino IDE, and with the analyser. Just a point, I assume you have uploaded the firmware version that does not include the bootloader ?

The overall comms interaction between the analyser and Trackgen is perhaps not as robust as it could be and may benefit from having the usual ACK/NAK handshaking implemented. Once a link is established, all the comms traffic is one-way, from the analyser to Trackgen, Trackgen sends nothing back. Perhaps the analyser needs to send VERSION several times, I'll look into this.

In the meantime, keep trying, given the above results you have reported, the firmware should work. Let me know how it goes.

Re: Feedback for version 0.9a here

Posted: Sun Jun 17, 2018 6:50 am
by Kees_01
Hi Steve,

With the Arduino IDE (serial monitor) the Arduino UNO returned "Trackgen V1.0" immediately after sending "Version". Done with the ArduinoUnoCommsTest sketch and togetheter with the SPR connected now.

I had uploaded "DDS_tracking.hex with rhe Xloader and not "DDS_tracking.with_bootloader.hex".

With the Xloader I then uploaded "DDS_tracking.hex". During upload I looked to the RX- and TX leds on the UNO boards. It's looks like as usual. Xloaded reported 9550 bytes uploaded.
Xloader closed and Arduino IDE launched. I typed "Version" or "VERSION". Nothing was returned with this hex file. I also repeated sending "VERSION" several times. Nothing was returned.
Closing the Arduino IDE and launching SAS s.a. the SQUARE remains red after selecting the right COM. At that moment I see the RX led of the Arduino UNO with a one time flash.
I hope this help.

Re: Feedback for version 0.9a here

Posted: Sun Jun 17, 2018 7:03 am
by Tech_Support
maccountrypilot wrote:well uno will not load when the spectrum analyzer tries to run .... it says there is no rsp present

You can only run one at a time as each application will reserve the RSP.

Sincerely

Tech_Support

Re: Feedback for version 0.9a here

Posted: Sun Jun 17, 2018 7:47 am
by SteveAndrew
Kees_01 wrote:With the Arduino IDE (serial monitor) the Arduino UNO returned "Trackgen V1.0" immediately after sending "Version". Done with the ArduinoUnoCommsTest sketch and togetheter with the SPR connected now.
Kees_01 wrote:At that moment I see the RX led of the Arduino UNO with a one time flash.
That flash will be the Analyser sending "Version" to the Arduino. The only thing I can suggest, is you let me know what I/O pins you are planning to use on the Uno, and I'll change the defaults in the firmware to match what you plan to use. I can then post the updated HEX file here for you, along with a modified Defaults.ini file for the analyser. It may just be possible that because I'm settings pins that do not exist on the Uno, that it's locking up in some way. I would have thought the Arduino library would take care of that, but you never know. I don't have a Uno, so I have no way of testing this.

Re: Feedback for version 0.9a here

Posted: Sun Jun 17, 2018 9:47 am
by Kees_01
Hi Steve,

It might be worth to give it a try, if it is not too much effort for you to generate a special hex file.
I have selected the Arduino UNO pins as follows:
pin 9 : Update (FQUP)
pin 10: Reset (RST)
pin 11: Data (D7)
pin 13: Clock (W_Clk)

Re: Feedback for version 0.9a here

Posted: Sun Jun 17, 2018 10:10 am
by SteveAndrew
Kees_01 wrote:I have selected the Arduino UNO pins as follows:
pin 9 : Update (FQUP)
pin 10: Reset (RST)
pin 11: Data (D7)
pin 13: Clock (W_Clk)
I'll see what I can do. I'll get the IDE to compile for an Arduino Genuino/Uno. Check back here an an hour or so and you should be able to get the new HEX file. Fingers crossed... ;)

Re: Feedback for version 0.9a here

Posted: Sun Jun 17, 2018 12:16 pm
by SteveAndrew
Hi Kees_01

Good news, and bad news. The good news is I've changed some of the code an re-compiled for an Arduino Uno. The HEX file is attached.

The bad news is, I have no way of testing the code. The IDE would not upload it to my Mega 2560, as the IDE expects to see a Uno connected to it. I had to upload the HEX file using XLoader. The file uploads, but when I used the IDE terminal to check it, I saw the Mega is outputting long streams of garbage. When I connected the Mega to the analyser, the analyser crashed when I selected the correct comms port. Probably caused by some form of buffer overflow in the Microsoft comms driver ? You can try the HEX file, but try it with the IDE terminal to test it. If it doesn't work with the terminal, it is not going to work with the analyser. I suspect there are differences in a lot of the pin mappings as well as other factors that will not allow a HEX file complied for one platform being run on another.

In the ZIP package is an updated default.ini file. You will need to place this in your documents folder. Also make a copy and rename the copy to CurrentSettings.ini

Documents\RSP-SAS Spectrum Analyser\SpecSettings

Rename the original Default.ini and CurrentSettings.ini files in case you need to re-instate them.

I am going to order a Uno. You can wait until I port the code to that platform but that may be some time yet, or you can get a Mega 2560 from Ebay for around the same price as a Uno. around $12-14. The Mega offers a lot more IO, ADC and timers etc. and is the platform I have always worked with in the past.
DDS_Tracking_Uno.zip
(10.4 KiB) Downloaded 895 times
Good luck, let me know what happens.

Re: Feedback for version 0.9a here

Posted: Sun Jun 17, 2018 2:25 pm
by Kees_01
Hi Steve,

Good news, and good news. The 1st good news is that the IDE terminal presented the good news, "Trackgen V1.0" after sending "Version". Well done!!!
Then I quitted IDE and launched the SAS s.a. app.
The 2nd good news is that after selecting the right Comms port the SQUARE turned GREEN!
Except the Cal-button all other buttons could be selected after I pressed Enable. I pressed the SPOT button.
I connected the Oscilloscope to the analogue positive SINE output and saw a waveform appearing to be a sinewave (I know with this DDS it is not a pure sinewave).
I measured 40 MHz, conform the number in the Tracking Gen control section. With the AD9851 DDS below 10 MHz the amplitude of the signal is about 0.8 ...0.9 Vtt (terminated with 1 M Ohm). Above this frequency it falls down.
Then I pressed the Track button and changed the Frequency to 10 MHz. The scope measured 10 MHz and 0.85 Vtt. Then I pressed the START button to start sweeping.
However the sweep is around 40 MHz (default?). The sweep speed is dependant of the Step choice. When Step-kHz value higher, then Sweepspeed higher.
At this moment the sweep start frequency stays around 40 MHz independent of the number in the Freq-MHz box.
I have to read now your manual in more depth. I preferred to let you know the results of your work for the UNO asap.

Btw. I nearly had taken the same decision to (in my case) buy a MEGA 2560 in case DDS_Tracking_Uno should not work.

Thank you for all your effort helping me to get the TrackGen running with an Arduino UNO!

Re: Feedback for version 0.9a here

Posted: Sun Jun 17, 2018 5:56 pm
by maccountrypilot
Hi … I have the following equipment …. I have the RSP2 and the rsp1a. Although I use the RSP2 as intended, I also have an Icom 8500 that has a 10mhz port. What I want to do is use the rsp1a and the spectrum analyzer as a panadapter. I am having trouble getting both rsps to be recognized on the computer

Re: Feedback for version 0.9a here

Posted: Sun Jun 17, 2018 10:33 pm
by sdjack
maccountrypilot wrote:Hi … I have the following equipment …. I have the RSP2 and the rsp1a. Although I use the RSP2 as intended, I also have an Icom 8500 that has a 10mhz port. What I want to do is use the rsp1a and the spectrum analyzer as a panadapter. I am having trouble getting both rsps to be recognized on the computer
You just got the answer from techsupport:
by Tech_Support » Sun Jun 17, 2018 9:03 am

maccountrypilot wrote:
well uno will not load when the spectrum analyzer tries to run .... it says there is no rsp present



You can only run one at a time as each application will reserve the RSP.

Sincerely

Tech_Support