Re: QMX and FT8CN audio issue - strange issue
Using VOX is a work around, entirely in character with all the other things in QMX.
I'm not suggesting it's a professional solution.
I was able to have SWR protection enabled but was operating into a dummy load, things may be different with a real antenna.
My guess is that something happens when the Tx finishes that pushes the SWR reading over the limit.? It might not be a real effect, reading the reflected power with TX on and the forward power with it off could give a high SWR.
Chris, G5CTH
|
Re: Hans , I can replace IC 203, STM32F446VET6 ,do I need the boot loader?
Oh forgot one thing..
?I received the new STM32F446VET6 from Mouser.? $12 bucks.? I installed it on the bad QMX and it does not boot up.
?No shorts though. So I suppose it must need the boot software to boot up.? No current draw with 12 volts inserted. I press the power button and hold it and get .123 ma current draw. Let off the power button and it goes to zero ma. But the processor won't boot.? So it does need the intial boot firmware as previously discussed.
?So that is what I figured out so far. .
|
Re: Hans , I can replace IC 203, STM32F446VET6 ,do I need the boot loader?
For those whom may need to replace the,IC 203, STM32F446VET6.?
I built a new QMX and installed, my original, "I blew it up myself Chip". It kind of worked.? It is only a cW transmitter.?? But so audio out of anything. no computer audio or speaker audio.
No Sound card found and no computer connection with the usb.?
?So I figure the sound card and usb are controlled through the IC 203.. So the Blown chip , is still screwed up.
I tried inject audio through the antenna port and trace it but no luck. I even yelled at it and I think I heard it laugh at me.
I then removed the blown chip and re-installed the one the came with the board. Takes me about 15 mins to repace the chip. It isn't that easy though.? All works now.
So hopefully we can get just the new
IC 203, STM32F446VET6? chip? someday.?
?
|
Re: Troubleshooting NO FLASH DRIVE
Check continuity on encoder pins when pressed on the power button.
When I put my QMX in a case I seemingly broke the solder joint on one of those and it took me a while to figure out why pressing the power button wouldn't do anything after putting it in the case.
|
Re: QMX VSWR indication
#qmx
Steve, if dissipation is an issue, did you consider changing the winding ratio? That'll come at the expense of bandwidth and possibly other parameters but at 1.5kW dissipation is a major issue. But otherwise looks good.
|
Re: Your opinion please.
#qmx
Another important parameter is the load. As you see the load is via the RFC, which is on 43 ferrite. It has pretty decent Q at 100kHz up to 2 or 3 MHz.
|
Re: Your opinion please.
#qmx
Interesting. What's the open loop gain and how does the open loop Bode plot look like?
Bringing the pole at the npn-pmos is not a good idea. The pole at the drain of the pmos is load-dependent and it can move, so it's wiser to keep the npn-pmos pole as high as possible and bring the pmos load pole as low as practicable. I increased my cap to 12uF and it seems happy. I just went to a POTA park and made more than 10 contacts on 15 and 17m alone, and made more than 30 on 20m with 7.5V battery. Several DX in the mix on all those bands. I also went to 10m and I see mine picked up by RBN but no contacts. This thing is coming together.
|
I'm troubleshooting the same sort of problem... :-(
Do you have any idea what the difference is between the cable that works, and those that don't?
Might be helpful for the rest of us to know! ;-)
Thanks,
Paul -- AI7JR
|
Re: QMX PC Connect, Anyone?
Nick, try TX and you will get there in no time :)
|
Re: QMX PC Connect, Anyone?
On 04/10/2023 17:18, Cliff wrote: Jim,
The QMX works FB with Fldigi with firmware 006. Later firmware versions broke the compatibility with fldigi. Hans knows about it and hopefully some where along the line it will get a firmware fix.
73,
Cliff, AE5ZA Here, in .pt land - there is a QMX that appears to work fine with FLDIGI & JS8CALL. It connects, does the CAT thing & Rxs fine. OK not tried Tx yet but ... Regards all Nick CT7/M0HGU -- 73 Nick M0HGU
|
Re: Troubleshooting NO FLASH DRIVE
I'll try that, though I have the same problem on Linux, too.
On 10/3/23 21:10, Compton wrote:
toggle quoted message
Show quoted text
On Wed, Oct 4, 2023 at 01:23 PM, Paul - AI7JR wrote:
What would be your next step?
Happened to me as well. Shut down and restart of Windows 10 and it
found it.
?
--
Compton
VK2HRX
Sydney, Australia
|
I absolutely agree Gyula! My antenna is a 3-el yagi about 15 metres above ground. That's the most I could get away with in this residential QTH.
As someone said, "If you can't hear them you can't work them!" With QRP? FT8 you can add "If you can't hear them at -12 dB or better you likely can't work them."
--
73, Rick
VE7TK
Website:
|
Re: QMX and FT8CN audio issue - strange issue
I tried this VOX workaround. With SWR protection disabled it works but it is not an elegant solution really.
With SWR protection enabled it ends up in S locked state after every transmission as with CAT control.??
|
Re: Your opinion please.
#qmx
Cliff,
I am delighted by your results, and by your kind words as well!
May your radio serve you well for a very long time!
73, JZ KJ4A
toggle quoted message
Show quoted text
On Wed, Oct 4, 2023 at 12:39?PM Cliff <ae5zaham@...> wrote:
John,
I beat my QMX up all yesterday and evening constantly running FT8. Not a blip of issue. Power stayed within 0.1 watt of the set 3.5 watts the whole time. Again this morning all is well. The only really warm area is the bottom of the board in the CPU area. I've put it back into the case and after an hour the case was warm, not hot, though I'd rather it not be so warm. I'm thinking of drilling holes in the sides for ventilation.
I'm very encouraged. If this holds up then it seems like your original call of modulator issues was right on.
Thanks again, I'd never have found that on my own.
73,
Cliff, AE5ZA
On Oct 3, 2023, at 17:58, John Z <jdzbrozek@...> wrote:
Cliff,
Good! I hope it holds up!
As I looked over pix in the assembly manual I saw that Q507 and L502 are adjacent, very tight. I have no idea what might happen if the drain of Q507 nicked the wire on L502. They are connected anyway but strange things can happen.
Keeping fingers crossed for you!
JZ
On Tue, Oct 3, 2023, 5:51 PM Cliff <ae5zaham@...> wrote:
That's helpful Chris. I'm sure you've seen my last where I now see 1.4v drop. Apparently that is normal. Whew!!!
73,
Cliff, AE5ZA
On Oct 3, 2023, at 16:45, Chris KB1NLW via groups.io <chrisrey1@...> wrote:
The PA voltage is not very close to the supply voltage, I usually see a drop of significantly over a volt. Hans in his documentation leads us to think it would be a fraction of a volt.
I cannot check my PA voltage since yesterday I blew one of my BS170s doing an SWR plot at full power (I didn't realize the SWR protection is disabled for the scan mode).
|
Re: QMX PC Connect, Anyone?
Good grief - volume control issues, computer connect problems - only usable with WSJT-X. This thing has morphed itself into a low band QDX. Looks like time to toss it into a drawer until it's ready for prime time. I give up.
--
Jim / K7TXA
Eagle, ID
SKCC 10447C
BUG 301
|
Re: Your opinion please.
#qmx
John,
I beat my QMX up all yesterday and evening constantly running FT8. Not a blip of issue. Power stayed within 0.1 watt of the set 3.5 watts the whole time. Again this morning all is well. The only really warm area is the bottom of the board in the CPU area. I've put it back into the case and after an hour the case was warm, not hot, though I'd rather it not be so warm. I'm thinking of drilling holes in the sides for ventilation.
I'm very encouraged. If this holds up then it seems like your original call of modulator issues was right on.
Thanks again, I'd never have found that on my own.
toggle quoted message
Show quoted text
Cliff,
Good! I hope it holds up!
As I looked over pix in the assembly manual I saw that Q507 and L502 are adjacent, very tight. I have no idea what might happen if the drain of Q507 nicked the wire on L502. They are connected anyway but strange things can happen.
Keeping fingers crossed for you!
JZ That's helpful Chris. I'm sure you've seen my last where I now see 1.4v drop. Apparently that is normal. Whew!!! On Oct 3, 2023, at 16:45, Chris KB1NLW via ??< chrisrey1@...> wrote:
The PA voltage is not very close to the supply voltage, I usually see a drop of significantly over a volt.? Hans in his documentation leads us to think it would be a fraction of a volt.
I cannot check my PA voltage since yesterday I blew one of my BS170s doing an SWR plot at full power (I didn't realize the SWR protection is disabled for the scan mode).??
|
Re: QMX VSWR indication
#qmx
Ryuji, points taken.? I should have mentioned in my earlier post? that the large toroids stack is also required for power dissipation reasons. Attenuators were put on the log amp board.? It works fine and has a measured directivity of >30dB from 80-10 meters.? ?I also made a "low power" VSWR bridge that can handle up to 500 watts (tested) also with >30dB directivity for 80-10 meters.?Works fine with my custom designed dual log amp board.?I built it into a small case with small graphics display/touch panel? for display of? VSWR and power.? Software for it was originally developed by Loftur Jonasson and extensively updated by Johan Holstein in the Radio Stuff 开云体育 group. Good accuracy for ~10 milliwatts to 500 watts.
- Steve K1RF
------ Original Message ------
Date 10/4/2023 9:00:51 AM
Subject Re: [QRPLabs] QMX VSWR indication #qmx
Steve, the ferrite loss of a parallel transformer is determined by magnetizing conductance (just the real part) in relation to the line conductance. Calculating the reactance and set that to be 10x the line impedance or greater is a common amateur radio practice, and while it is usually ok as a rough guideline, it does not lead to the best design tradeoff. The voltage sensing transformer loss should be relatively small.
I wonder what the insertion loss and directivity are in actual use situation. If it's working perfectly, I guess it is fine, but I would insert a 10dB attenuator on both FWD and REV ports before letting the lines exit the box. Terminating those lines properly is crucial in keeping the insertion loss low. If properly terminated detectors are always connected just outside of that box, that is probably not an issue, but two 10dB or greater attenuators should limit the worse case scenario to a very decent value. It the detector side is AD8310, that side could use another attenuator as well. Directivity may also suffer from asymmetric leakage flux characteristics with different cores.
|
Re: Your opinion please.
#qmx
Ryuji,
I spent some time this morning simulating the modulator. I found it to
be a pretty well behaved circuit. With nominal device parameters and
running at 12V there was a small amount of ringing in response to a
fast step input, and the frequency response showed a broad peak
between 140 and 170 KHz. Phase margin was 50 degrees at loop gain =1.
A 1nF capacitor between the PMOS gate and drain might tone things down
even further.
I will beat up this circuit sim some more in the next few days. It may
be that some stack-up of values will result in oscillation, but I
could be hunting for a good long while and turn up nothing.
JZ
toggle quoted message
Show quoted text
On Tue, Oct 3, 2023 at 7:42?PM John Zbrozek <jdzbrozek@...> wrote:
Thanks, Ryuji.
I may simulate the modulator sometime soon.
JZ
On Tue, Oct 3, 2023, 7:26 PM Ryuji Suzuki AB1WX <ab1wx@...> wrote:
I looked into the stability of the modulator circuit because my LDMOS finals are getting blown in conditions that are far milder than I expected. I started looking at all directions. But I noticed that the output of that amplifier was reading HIGHER than the supply voltage. At that time I didn't think deeply and linked that to the flyback voltage you were discussing at length at that time. That's why I was thinking that modulator was PWM until yesterday, without taking a close look. When Hans pointed out that was an amplifier, I was embarrassed not looking at it closely earlier before saying anything about it. Earlier today, I just rearranged that circuit diagram and it looks super obvious that it is a differential amplifier, so while at it, I calculated the gain and the pole locations. It's super simple. Those pnp's are emitter followers. One of the BC817's just a diode to match the B-E junction drop of the other. The non-diode BC817 has a decent large gain there (remember the gm of a BJT is always 38 IC at 300K, and the IC can be guessed from the VGS of the MOSFET), and the p-ch MOSFET has some more. The 10k and 1.91k give the negative feedback. You could do a lot more detailed study in SPICE but for my purpose, a sheet of paper was enough.
That amplifier can be stable, marginally stable or unstable based on the component tolerance, the load (final amp) impedance, temperature, and various factors. It may cause issues only in some copies, specific situations or part of the keying ramp, or can be something we haven't looked at yet. That's the thing about the stability issue.
When it oscillates, it can make the emission spectrum really dirty, including frequencies below the carrier frequency, and may also blow the final transistors. I initially suspected some strange problems due to premature breakdown of the final transistors but as I started paying attention I see dirty emission in my spectrum analyzer in some situations.
The easiest and best way to deal with it is to add another multilayer ceramic cap, like 16V 10uF in parallel to C506. But I'm in an early stage of testing this solution. C506 is a SMD but it is near the edge of the board and next to the two resistors.
|
Re: QMX PC Connect, Anyone?
Jim,
The QMX works FB with Fldigi with firmware 006. Later firmware versions broke the compatibility with fldigi. Hans knows about it and hopefully some where along the line it will get a firmware fix.
toggle quoted message
Show quoted text
On Oct 4, 2023, at 10:53, Jim Bennett / K7TXA via < w6jhb@...> wrote:
I posted a question yesterday asking if anyone has successfully connected their QMX to MacLoggerDX (MLDX). No replies so far. So, I'm wondering if anyone has connected their QMX to ANY application other than WSJT-X?
As I mentioned in the other post, QMX connects to WSJT-X easily, but no matter what I try with MLDX, it is a no-go. I've involved Don Agro, the developer of MLDX and he is at a loss as to what's going on. He says that the data returned to MLDX from QMX is garbage. If so, then how does it work with WSJT-X? I've had a bunch of other QRP Labs rigs (QCX+, QCX Mini, QDX) and all were able to connect to MLDX with no issues. But not this QMX.
Has anyone connected their QMX to ANY application besides WSJT-X on any platform, PC or Mac?
--
Jim / K7TXA
Eagle, ID
SKCC 10447C
BUG 301
|
Re: QLG2 GPS unit 1 second slow
Guys,
If you are using the QLG2 with the QMX does it work fully with the QMX Transceiver?
My QMX doesn't go into transmit when using WSPR when the time comes, and I was wondering if there was a bug in the code?
If it does work I am wondering if I have the wrong setting somewhere.
Thank you, Michael
|
Ryuji Suzuki AB1WX
Paul - AI7JR
Rick Williams - VE7TK