Thanks Chris, that's really helpful, I will get the oscilloscope out this evening and check that the left encoder is working.?
@KD8CGH I seem to have hijacked your thread but hope this information is useful to you.

Simon
G0FCU.

开云体育

On Wed, Sep 6, 2023 at 08:15 AM, Gary W9TD wrote:

At Motorola, we put a known SWR load on the end of a line stretcher and were therefore able to test a PA over the entire fixed VSWR circle of the Smith chart. This was easy to do at UHF, but would not be practical at HF. I would guess that many of the points around the Smith chart VSWR circle could be generated by programmable impedances (LCR networks switched by relays). The quarter wave line mentioned by some only inverts the impedance connected to its far end, i.e. j50 to -j50.
Gary
W9TD

yes at UHF its smaller.

That's the hack using the 1/4 wave line can be used for;
?The idea of a 1/4 wave line is that it reflects the inverted result.
Aka short is an open, open is a short.?? With LCR and r you can
create any composite mismatched load possible. I use it more at
VHf and up but it works even at 1MHZ though the coax will be fairly long.

For many the experiment is easier at 10M as then we are talking about
less than 6ft of rg58.? Put the NanaoVNA or VNA at one end and play with
the terminations using the Smith format see what is permissible for a 2:1 circle
and again for 3:1.? Its educational and informs one what antennas
and their associated feed lines or even antenna matchers do when
the match is anything but pure resistive 50 ohms.?

Its the reactive reflections that can cause havoc as well as the reactive load.? With
switching amplifiers (class C,D,E,F) a reactive load can look very different to the finals
especially when looking though LPF? (hint filters sometimes are like transmission lines).?

--
Allison
------------------
Post online only,?
direct email will go to a bit bucket.

Hi Andy. That's a pretty general statement. Most modern rigs have many software defined functions as part of their architecture but the software defined functions are not obvious to the operator because they are triggered by selections with knobs and buttons. But you're right in that I'm not a fan of software centric rigs where there are so many features than it's difficult to grasp every bell and whistle without a steep learning curve. To me, in addition to raw specs, a good rig is one where someone unfamiliar with the rig can walk up to it and use its basic functions with a minimal learning curve. I've seen many top end rigs which were so complex that on one memorable field day, someone lost an hour just trying to get the rig to transmit because it's owner wasn't there to provide help.

-Steve K1RF

I wish Hans would publish some guidance on the power up process to enable troubleshooting. There's bits and pieces spread across this forum. So far I think I've got:
- the 3.3V linear regulator starts up first to enable the rotary encoder switch to initiate the power on
- after power on the firmware takes control and the 3.3V linear regulator (PCB #2) is switched off and the SMPS (PCB #1) takes over

On the other hand this may be incorrect but without a definitive answer (or significantly more skill in electronic engineering than I possess) it's difficult to troubleshoot.

Simon.

The more options the better - after certain core functions are all working correctly.? I've put my QMX away in favor of a (tr)uSDX until I can regulate the side tone volume.

Steve & Hans

The more options, the better. That is, of course, as long as they do not get in the way of good operating. Detailed menu options are great, as long as you almost never have to look at them after the initial setup. You need to have basic functions easily available from the front panel and the less used ones in the menu.?

I have an Elecraft K3 and they really got it right. Almost everything you need in real time is up on the front panel. The menus rarely get looked at after I set it up first time.
Older Yaesus such as the FT-857 were a nightmare with 3 menus that needed to be accessed constantly for even the simplest operations.

The QCX and QMX are much simpler, but follow the same good principle. Everything you need to operate is available from the front panel with clear markings and logical arrangements. Less used and more detailed settings are tucked away in the menu.

So, I think the best principle is Simple to Use, but with powerful options. Keep up the good work, Hans! And get well quickly.
--
73, Dan? NM3A

Mike, all
Just a round trip ticket to what I found.
It took me a while to get to the bench...... school has started... busy, busy.?
I did change out the BS170s when I found the tuner (MFJ969) bought new 26 years ago has a bad solder connection in the back of the large induction coil.
The roller bar was not making connection. That's why the error was intermittent and all over the place.

I noticed this had effects on the other radios in the shack I didn't catch right away.
I thought it was my imagination or......... age? Nah.... couldn't be that?
I started checking everything.....?
I put on an auto-tuner from another maker and found no problem with any radio including the QDX.?
The change for the BS170s was just a precaution.
The unit from MFJ was repaired after I found the bad connection.
All is well with the world and back in the air.

Thanks for your help and encouragement.
73
James?

Hans,

On page 43 of the assembly instructions the nylon screws are ?stated as 9mm at the top of the page and 10mm at the bottom of the same page.

73
John
G4YTJ

J Zbrozek? and I did some work on this for a? Cz ham (for a commercial appn too)? who now seems to have deleted all the posts . Which is a pri_k of a thing to do. I have done some message searches.

Contact me directly if you want to us the Ac08 driver chip to run into a small RF transformer as in? BN 61-202 preferred .? There are some? electronic gotchas if you are not careful but it can be done and around 45mW out as a sinusoid is easily possible if you have a 3 plus three driver arrangement.? That's far more than you want of course.?

I had some mystery audio as well. Mine was the result of the case pressuring the board via the buch converter screw, since trimmed. I never did locate the PCB issue that the pressure invoked.?

I suspect you have a stray solder bridge or a whisker shorting something out. As unique as your problem is, it has to be construction-related. If you don't have a jeweler's loupe, get one and closely examine the board. Another thought - I overheated my speaker jack during soldering, which also caused some audio issues. I swapped mine with the PTT jack while waiting for some new ones to arrive.

It may be worthwhile to test the power boards independently. See
Chris/G5CTH write up at:

/g/QRPLabs/message/106787

-mike/w1mt

On Thu, Sep 7, 2023 at 7:38?AM mike/w1mt via groups.io
<w1mt.qrp@...> wrote:

Both Vdd and Vcc seem high. You're right about the path of PWR_HOLD to
the 5V supply board, it's a straight shot. Please note that Vin is
right next to PWR_HOLD and PWM_5V on the connector jack. Do not short
Vin to either PWR_HOLD or PWM_5V. I shorted PWR_HOLD on one of my
QMX's (described in another post) and it was salvageable. If you screw
up the PWM_5V port that's not as easy to salvage.

If you have a scope look at PWM_3V3 and PWM_5V. If there is no scope,
simply measure the DC voltage at these points. Reason for this is to
see if the CPU is properly supplying a signal to the power supply
boards. Voltage should be greater than 0v and less than 3.2v...

-mike/w1mt

On Wed, Sep 6, 2023 at 11:50?PM n5ebw <eric.n5ebw@...> wrote:

Hi Mike,

Per your suggestion, I checked the voltages of the power supply boards. The 5v board is giving me about 5.5v on the output (VCC), and the 3.3v is giving me 3.6v (VDD). They're not bang on but I suspect that the 3.3v rail is actually 3.6v because of D109 being a 3.6v zener diode that's strapping that voltage and the 5v supply is PWM controlled by the main processor, but I fully admit this is me not knowing SMPS fundamentals as well as I should. Pictures attached of the power supply connector block schematic and actual view of what I tested (in case anyone is following along and wants to see the method I took). I understand what you may be saying about the short, but after a fairly thorough look around (not above doing it again), I came up short (pun intended, hihi). I went and looked at the schematic again. The net for PWR_HOLD does not appear to take a route that would introduce external components and make it susceptible to a short. It looks to be a home run between the 2x4 header and Pin 53 of the processor so I'm very open to suggestions about where to start poking next. Its exposure on the power supply board also looks to be limited to the route between the 2x4 header and where it injects itself into the gate of Q104 and the bleed resistor R105.

Hans,

I am using Linrad to achieve aural single sideband CW reception (yes a CW signal has two
sidebands and we need only one) in a bandwidth of 5 Hz, good for slow Morse at 10 WPM.

The main application is to unearth very weak beacons at long wave.

73,
Roelof, pa0rdt

On 07/09/2023 12:07, Steven Dick, K1RF wrote:

But I'm a huge fan of "keep it simple"

Then SDR rigs are not for you!

开云体育

thanks for the input. ?moving the power connector is not an option, the way I soldered it in. ?I did spring the body of the encoder a bit when I removed the tag, and squeezing it together helped. ?but I will add something as an insulator, and will try the heat shrink tubing first. ?73/72

Both Vdd and Vcc seem high. You're right about the path of PWR_HOLD to
the 5V supply board, it's a straight shot. Please note that Vin is
right next to PWR_HOLD and PWM_5V on the connector jack. Do not short
Vin to either PWR_HOLD or PWM_5V. I shorted PWR_HOLD on one of my
QMX's (described in another post) and it was salvageable. If you screw
up the PWM_5V port that's not as easy to salvage.

If you have a scope look at PWM_3V3 and PWM_5V. If there is no scope,
simply measure the DC voltage at these points. Reason for this is to
see if the CPU is properly supplying a signal to the power supply
boards. Voltage should be greater than 0v and less than 3.2v...

-mike/w1mt