Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
QCX Troubleshooting - Please can you help?
|
My QCX 40worked perfectly first time, aligned easily and I made a dozen or so QSOs around Europe - chuffed to bits.
But it did have the start-up problem. I did the mod and messed it up and blew the Atmega microcontroller chip (a single row of white blocks, re-flashing the chip didn't cure it). A new chip arrived today and I installed it. Bingo! Nice QCX message and what band do you want? But the rotary encoder hardly worked at all, turning it slowly or quickly it only responded randomly and not necessarily in the right direction, also the push button switch part of it wasn't working reliably. So it was difficult to select the 40 metre band. So that's problem one. Then I thought I'd listen in to see if the receiver was working. I plugged in the headphones, but plugged them into the key jack socket by mistake - Duh! this switched on the TX with no aerial or dummy load and there was a nice smell of burning plastic, and some smoke. I've replaced the 3 output transistors. Now the rig works fine on 7.020 it receives well and transmits well, but there is no sidetone, just a buzz instead. So my two problems are the rotary encoder problem and no sidetone. Please can anyone with more up-to-date understanding than I have, suggest a trouble shooting route? I have a good digital scope, and a multimeter, I just don't know where to start (it was valves and class C when I passed the exams). Any help would be very gratefully received (I've ordered another QCX kit out of pessimism). Don't want to be without one. Best wishes, Julian Forsey G4ETS |
|
Hi Julian, I have done some thinking on this, and if I were you I would go back to the original configuration (remove the modification) and start from there. The reason I say this is that your problems seem to have surfaced after you did the modification. Although it fixed your start-up issue you ended up with more problems. The sidetone is generated by the processor so with a scope you could check that on the? pin of the processor. There is quite of bit of good documentation in the assembly manual, along with circuit descriptions and diagrams. The Rotary encoder is very dependent on a good 5 vdc supply and I believe that is part of the modification that you had done, depending on which one that you did. Back to the sidetone issue you have a newly programed chip installed and I would check the parameters in the software for the generation of the sidetone level and frequency. Check and make sure you have a good clean 5vdc coming from your 5 volt regulator. Thats about all I can say but, wish you well. Bill KN8DMK
|
|
The encoder part sounds a bit strange, it has dedicated pins to the micro and does not operate in analog mode. Put the scope on those two pins and while rotating the encoder see that it looks like a clean square wave with 5V amplitude. The chip has built in pullup (activated by the program) so the positive going flank should have more of a rounded edge than the negative. Nothing wrong with the socket? like pins not sitting properly ?Perhaps doing a factory reset, woult not hurt and eliminates any possible strange settings messing something up. |
|
Hello Daniel and Bill, and thanks for taking the time to help.
The rotary encoder problem: The 5 volt supply looks good. If I pull the encoder knob upwards as I'm turning it, it works almost perfectly, which makes me think the encoder is faulty, or there's a dry joint/cracked track. Chasing the sidetone, I put the scope on pin 15, sidetone output. There's a rather dirty square wave, at about 35hz - which explains the buzzing sound. So, how am I getting a 35hz square wave instead of a 700hz one? Hans' notes say it's all done in the software, but I have updated software, so what can be going on? I'll attach a picture of the 'scope display from pin 15. Any ideas gratefully received. Julian G4ETS |
|
I've checked menu 4.9, it's set to 700hz as it should be. Now I'm thinking that the 20Mhz clock might be the problem - if there's no 20Mhz, perhaps that stops the sidetone from being generated? And also stops the optical encoder output from being sampled?
So I've put my scope X10 probe on pins 9 and 10 of the chip - no signal at all. No signal on the 20mhz crystal pins either. No DC either. The crystal is connected to the chip OK, no dry joints or cracked tracks. So I've replaced the 20 Mhz crystal. No change. tested the old crystal, perfect. Baffled again. Please, can someone with a functioning QCX and a ' scope let me know, do you get a 20mhz signal on the chip? and what DC voltage? As ever, any help gratefully appreciated. Julian G4ETS |
|
Problem with encoder - try to check voltage at pin AVCC of ATmega. It should be close to 5V.
After "Unreliable microcontroller start-up" modification, I had only 4,6V on AVCC that caused bad decoding of buttons inputs (3 buttons + resistors make a different voltages on A/D input, software compares it to reference AVCC). Inductors 100uH has quite high resistance of 7 ohm that means voltage drop on ATmega pins. |
|
Arv Evans
AVCC should be very close to the Vcc supply voltage.? If that is at 4.6 then either your Vcc is at 4.6 or something is dragging AVCC down.? Normally this pin is unconnected, but if you want best repeat measurements from the ADC you can add a filter to the AVCC lead. ADC reference for analog-to-digital voltage conversions is the built in 1.2 volt AREF potential. If the AREF is not correct then the ADC measurements will be inaccurate. If using ADC inputs to detect specific voltages you can use a high impedance DC meter to measure the actual voltage to see if they are within a range specified by the software. Arv _._ On Sat, Dec 9, 2017 at 10:23 AM, <ondrej.briatka@...> wrote: Problem with encoder - try to check voltage at pin AVCC of ATmega. It should be close to 5V. |
|
Arv This is not correct.? The AVcc pin provides power to the internal analogue voltage circuits in the ATmega328.? The QCX uses the ATmega328 ADC with the reference set to the supply voltage (AVcc). Not the internal reference.? 73 Hans G0UPL On Sat, Dec 9, 2017 at 9:45 PM, Arv Evans <arvid.evans@...> wrote:
|
|
Arv Evans
Hans Okay, I stand corrected.? I guess I have never worried about AVCC before. I always relied on the internal 1.2 volt reference for measuring analog voltages with ADC.?? I went back and read over the comments on AVCC and AVCC1 so maybe I now understand things a bit better.? Is it necessary to have both AVCC leads tied together or both filtered/bypassed to facilitate startup? Arv _._ On Sat, Dec 9, 2017 at 10:17 PM, Hans Summers <hans.summers@...> wrote:
|
Can you check and let me know what voltage you are getting at Pins 9 and 10 please? I get no DC voltage at all, and no 20mhz signal either.Julian I've avoided jumping in because I cannot really help. But it seems to me if the display is showing something the 20MHz must be working. I've no idea about voltages but a scope put on the crystal pins will most likely stop oscillation. The way to confirm is with a RX not connected to the device. 73 Alan G4ZFQ |
I've now "listened" for a 20mhz signal and there isn't one. As it's a new chip, and a replaced (tested) crystal, I can't think what else it can be?Julian Just checked my U3 has 2.4/2.6 on 9/10. I see no reason the CQX should be different. It is a genuine replacement from Hans? I do not know but if you have no voltage then maybe the fuse is wrong. I do not think you will have the micro working if it is not clocked at all but something may well happen if it is clocked at the wrong speed. 73 Alan G4ZFQ |
Daniel Ekman SA2KNG
to navigate to use esc to dismiss