QCX-Mini doesn't Auto-set time with GPS #gps #qcx #troubleshoot


David (W5CWT)
 

I am trying to set the clock on the QCX-mini with the QLG1 gps unit. The GPS unit seems to work (all LEDS operate as expected) but the radio never sets the clock or the grid square. Calibration menu items 8.11 and 8.12 both seem to work without issue. 

When I scope the output of the QLG1 I see the 1pps on the tip and what appears to be serial data on the rig BUT it is not very clean. I am a noob with a scope, so I could very be at fault for the traces. I can't seem to get a reliable (or valid) decode with the RS232 decode function on the scope.

I've added photos. Any suggestions?

Thanks,
David W5CWT


G3SPL
 

Well, the good news is it seems the 1 pulse per second is working as  you are able to calibrate the crystal oscillators (8.11 and 8.12).  Those serial data traces look a tad noisy, but as you say, that could be due to things like earth loops on the scope input.  But I would concentrate on looking at the serial data - does it look clean directly on the output of the GPS, for example?  You may have a grounding issue with your lead or perhaps the serial data is not the full 5 volt swing?

Check that a good 5V swing of serial data is available on pin 19 of the microcontroller (IC2).  If the keying paddle works OK it would pretty much rule out any poor connections from the keying jack to the controller.

Interestingn one, this.  Hope you find out what's going on!
--
Peter Lee
G3SPL


Mont Pierce KM6WT
 

On Tue, Aug 31, 2021 at 10:40 PM, David Chenault wrote:
I've added photos. Any suggestions?
Very Nice Scope.

This is 5v logic, so you need to set your probe to 1x.  You should see 5V or 2.5V per division.

Your pic shows it's displaying 200mv per division.  That's what mine looks like if my probe is on 10x instead of 1x.


73
km6wt


Mont Pierce KM6WT
 

On Wed, Sep 1, 2021 at 11:21 AM, Mont Pierce KM6WT wrote:
This is 5v logic, so you need to set your probe to 1x.  You should see 5V or 2.5V per division.
See pics on my previous post #68528 (click here).

My scope is showing 2V per division, and the pick signal is at 5V.

(wish my scope had serial data decoder...  Very Nice).


73
km6wt


David (W5CWT)
 

On Wed, Sep 1, 2021 at 01:26 PM, Mont Pierce KM6WT wrote:
On Wed, Sep 1, 2021 at 11:21 AM, Mont Pierce KM6WT wrote:
This is 5v logic, so you need to set your probe to 1x.  You should see 5V or 2.5V per division.
See pics on my previous post #68528 (click here).

My scope is showing 2V per division, and the pick signal is at 5V.

(wish my scope had serial data decoder...  Very Nice).


73
km6wt

 Thanks! So, my probe was on 10x. Changed it to 1x and modified the scale. New photo included below. Still, very noisy and no solid decode by the scope. Haven't cracked open the QCXmini to check IC2 yet.


David (W5CWT)
 

Ok, I may have an issue with my cable. It's not apparently shielded very well. I scoped at the output of the board and got a slightly cleaner signal and a good decode. Still has a number of spikes when the signal switches. I'll try to solder a new cable and see if that helps. 


john@...
 

David,
that part of the noise which looks like overshoot on the waveform - the spikes at each transition - look like scope probe calibration is needed. There will be a facility on the scope to rectify this. The underlying waveform may be quite clean. 

John G3ZTU


David (W5CWT)
 

Ok, finally got a chance to crack open the rig and pin 19 in IC2 shows the same 5V trace as what's coming off the GPS unit. Even got a decent decode of a GPGSV statement. Any other ideas?


Mont Pierce KM6WT
 

On Wed, Sep 1, 2021 at 08:27 PM, David Chenault wrote:
It's not apparently shielded very well. I scoped at the output of the board and got a slightly cleaner signal and a good decode.
I don't know about the spikes at the top/bottom of the signals.  I have them too, but not as apparent.  It could be as John stated it could be scope probe calibration, or artifacts from probe cable, etc.  I don't know...

But, as far as TTL logic goes, the spikes above and below will not affect the logic.  TTL logic, anything 2V or higher is read as logic 1, and at 0.8V or lower is read as a logic 0.

So I don't think the spikes you see at the beginning of each bit should not affect anything.

Although, I am not an oscilloscope expert, so, I really don't know.


Do you have a USB to TTL Serial convert you could hook up? 
Something like this on on Amazon (click here)?

These work great to power QL kits that run on 5V (QLG1, QLG2-SE, Clock, VFO/SigGen).
I have a lot of USB cables along with several AC/USB chargers.

They can also be used to connect the QLG1 to a computer's USB port.  Then using a terminal program like PuTTY you can observe the GPS data stream.


73,
km6wt


David (W5CWT)
 

I've found a breakout board for USB to TTL in my junk drawer that I'm going to try and see if it will decode with PuTTY. Sidenote: I wondered if my Boafeng programming cable was the same thing and if I could feed the TX and the GRD into the pins and then open the port with PuTTY, but I haven't been successful with it yet. 

David


David (W5CWT)
 

UPDATE: I connected the GPS to a USB-FTDI board and received perfect copy via the serial port, so the GPS is working fine.

$GPVTG,175.02,T,,M,0.02,N,0.04,K,A*3A
$GPGGA,235658.000,3209.8110,N,09447.6680,W,1,7,1.00,168.1,M,-24.9,M,,*6D
$GPGLL,3209.8110,N,09447.6680,W,235658.000,A,A*42
$GPGSA,A,3,31,22,21,10,26,25,32,,,,,,1.34,1.00,0.90*0E
$GPGSV,2,1,08,31,83,235,44,32,48,049,40,10,35,132,38,22,34,315,35*71
$GPGSV,2,2,08,21,33,270,34,25,25,063,28,26,22,168,34,03,13,317,25*70
$GPRMC,235658.000,A,3209.8110,N,09447.6680,W,0.03,208.39,040921,,,A*78
$GPVTG,208.39,T,,M,0.03,N,0.05,K,A*3B
$GPGGA,235659.000,3209.8110,N,09447.6681,W,1,7,1.00,168.2,M,-24.9,M,,*6E
$GPGLL,3209.8110,N,09447.6681,W,235659.000,A,A*42
$GPGSA,A,3,31,22,21,10,26,25,32,,,,,,1.34,1.00,0.90*0E
$GPGSV,2,1,08,31,83,235,44,32,48,049,41,10,35,132,38,22,34,315,35*70
$GPGSV,2,2,08,21,33,270,34,25,25,063,27,26,22,168,34,03,13,317,24*7E
$GPRMC,235659.000,A,3209.8110,N,09447.6681,W,0.01,45.62,040921,,,A*4F

I then used two 3.5mm connectors and alligator clipped the GPS to the QCX-mini (with all the proper connection) to eliminate an issue with the cable. Problem persists. GPS powers up, lights flash as expected, but not auto-set by the radio. 

Still searching for an answer....

David


AK4XO - Jon
 

Full disclosure, I have not read the entire thread in great detail, but, I was also having the same issues with the mini not setting time from gps. 

How are you powering the unit? I am powering the GPS from the mini via the PTT port. I found that while powering the mini from an AC power adapter the time would not set. Once I switched to battery power it worked just fine. I have not investigated further and don't really plan to, especially due to lack of proper test gear.

Not sure if others have had this issue or if it will help here, just sharing my experience 

Jon AK4XO


David (W5CWT)
 

Thanks Jon. ALL ideas are welcome! I thought about the power thing as well, but I've been running my rig on battery from the start and still no joy. I've tried two different batteries even.  I even tried powering the rig and the GPS from separate sources and still no luck. 

David


Rob Giuliano
 

I had a similar problem with clock and calibration on the U3S.
I see you have VTG enabled

  Turned ON both ZDA and VTG -> clock did not update, no GPS info on second line.
     Let it run for only a few minutes, but should have updated the information in that time.
  Turned OFF ZDA -> clock updated immediately, and GPS info showed on second line.
     Clock is < 1 second ahead of network time -> dead on with time.is
     Let it run through a few TX cycles to see how the corrections go.
  Turned ON ZDA and VTG OFF -> clock did not update, no GPS info on second line.
     Let it run for only a few minutes, but should have updated the information in that time.

To check out the thread for more info: https://groups.io/g/QRPLabs/message/70420


--
Rob KB8RCO


David (W5CWT)
 

Rob,

 Thanks for the info.  My ZDA is off but yes, the VTG is on. I tried to disable it but was not successful. I was able to send a query command ($PMTK414*33) and got a valid response but my setting command was not successful ($PMTK 314 ,1,1,0,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0*29). At any rate, from your message, it looks like what worked was the ZDA being off and the VTG on. 

Still searching for a solution.

David W5CWT


Mont Pierce KM6WT
 

On Thu, Sep 2, 2021 at 08:45 PM, David (W5CWT) wrote:
Ok, finally got a chance to crack open the rig and pin 19 in IC2 shows the same 5V trace as what's coming off the GPS unit. Even got a decent decode of a GPGSV statement. Any other ideas?
This is puzzling why this isn't working...  The only things left that could be the problem is either a bad processor, or, bad connection to pin 19.

Does the mini work in general?  I.e., with paddles installed, you can generate CW signal into a dummy load? 

When you scoped pin 19 on IC2, did you scope it from the bottom of the board?  Or, right at the processor pin where it plugs into the socket?
If from the bottom of the board, can you scope it again at the processor pin?

If you plug in a paddle into the paddle socket, does it generate the Dits/Dahs correctly?


When you connect the QLG1 to the mini, does the the letter "G" show up on the mini's display?


Do you happen to have another QCX (any model) that you can temporarily swap in it's processor into the mini?



When I connect my QLG1 to my mini, the "G" is displayed immediately on the LCD as pictured:




73
km6wt


David (W5CWT)
 

UPDATE:  Yippee!

Mont, you 'da man! I'd never seen the "G" on my screen before. I had scoped IC2 pin 19 from top and bottom with same results. My QCX-mini works generally, paddles produce nice CW into a dummy load. 

BUT....the radio has been in "Practice Mode" this whole time. I put into Practice mode per the instructions in 4.14 of the assembly manual (see below). But I turned off the practice mode and the "G" appeared (again, the manual says an "A" will appear). I was also afraid to hot plug/unplug the GPS but then I realized that advice (from the QLG1 documentation) was for the previous QCX+ design, with the PTT on the tip instead of the ring of the PTT jack. 

So, with Practice mode off, the "G" appeared and as soon as I got a fix,---BOOM--- the clock updated! 

Thanks for all your help everyone!!

David W5CWT

(Page 102, QCX-Mini Assembly Doc)
4.14
.....
The GPS is therefore wired directly in parallel with the paddle. The GPS may not be used at the same time as the paddle. This would cause the transmitter to be keyed by the 1pps and serial data signals. Always put the kit into Practice Mode (see menu 4.7) before connecting a GPS, to avoid accidentally frying the PA.

The ATmega328P has a USART serial data peripheral. However, the pin assignments were already made, so far down the design process, at the time the GPS receiver option was included. Furthermore, the I/O pin availability was already scarce. So, the two GPS signals share the same I/O as the paddle inputs. A software-emulated USART peripheral module now reads the GPS serial data, at 9600 baud.

If you connect a GPS and forget to put the kit in practice mode, it will automatically recognize serial data at the paddle input and temporarily enable practice mode (an ‘A’ appears in the display) to product the PA circuit. assumed, from reading the instructions, that that was automatic due to inserting the GPD plug.