Date   

Re: QCX Frequency Calibration

Erik
 

Thank you everyone for the replies and help! I was confused with the procedure.
I will try Al's advice using CHU since I do not have a reference radio hooked up at the moment.

Thanks again, OM's

73 de Erik - N9OJR


Re: QRPp Antennas #u3s #antenna

ajparent1/KB1GMX
 

Trueblue..  (IBMer?) ;)

Anyhow.  There are many ways to look at this.
LEts review the three rules:

1. For a bigger signal put the wire higher.  Altitude always helps.
2. If you cannot put the wire higher, put up more wire.  Longer wire,
a half wave to tune easier using a weak ground, add a counterpoise
on the floor for examples.
3.  Any wire you put up is still better than a bucket of wire in storage.
The position there is try something anything see if with all the constraints
if you can make it better.

That is a practical and pragmatic approach.  Minimaize losses where
it applies, get the best match to transfer as much of the power as possible.
Add grounds or counterpoises to insure the coax and radio are not the inefficient 
half of an antenna.

A dipole for 20M is 33ft and some inches out in the wind shorter indoors.
It can be bent in three dimensions as needed (L, V, C, W or Z!).  

Tuners, um rather, antenna matching units of decent design cost little in terms
of power loss unless your trying to load a very short antenna for a given frequency.
A dipole tuned to the right length has no need of one for example.  The best place
to do that tuning is at the antenna (less feedline loss) or better yet experiment
with loops or self contained antennas like tuned Loops.  Experiment, wire is cheap
and can be re-used.

Allison


Re: QRPp Antennas #u3s #antenna

Mark, WD4ELG
 

It will be interesting to see the responses from others.  Answers vary because there are many variables, as you know.

 

And I echo Alan, and add “The best QRO station has…the biggest antenna with low losses, just like the QRPp station”  Remember when Clinton used to say “It’s the economy, Stupid?”  My Elmer used to say “It’s the antenna, Dummy!”

 

In this case, you answered a very specific question…good on ya OM.  I will provide a very specific answer.

 

You asked: Do QRPp devices substantially lose power from antenna tuners and other peripheral whatnot beyond the concerns of a more QRO system?

 

My response: BEYOND the concerns of a more QRO system?  No.  The concerns are the same.  We must look at the entire system to see total efficiency, QRO or QRP or QRPp.  One footnote: as you already know, with QRO the concern is for the ability of the components in the tuner to handle higher voltages from more *power* contained in the mismatch.  But that is the one difference.

 

If I was in your shoes, and I was asking the questions, my next question would be “OK, Mr. Dunning, do I need to be concerned about the losses in my coupler in absolute terms, then, rather than relative to QRO?”

 

My typical answer, because I am an engineer, is “it depends” and “it’s all relative”

 

We need to viewing the entire setup as a complete system, from rig to antenna, we must consider several factors:

 

1.       What is the SWR mismatch?  Coax losses could be higher than the tuner losses depending on length and type of coax.

2.       What are the components used in the tuner at your QTH?  Higher SWR at the tuner will be accompanied by higher loss across the components, but “better” (higher quality, better design) parts will have less loss.

 

Of course, the goal is maximum radiated power. Best to illustrate with my example (feel free to stop reading if this is boring, I apologize but it’s my nature).

 

Also keep in mind these are guesstimate numbers, not measured.  And for the purists on this board, please don’t flame me if you disagree.  I’m 54, not 24, and I never claimed to be a PhD.  The only thing I ever did claim was for the punks in the neighborhood to get off my lawn.  Feel free to comment/add your observations.

 

My setup:

·         I am sending 500 milliwatts of WSPR from my rig to an end-fed wire, 100 foot vertical and 32 foot horizontal.  I am transmitting on 80 meters, that’s a half wave long wire.

·         I have a remote coupler at the base of the wire, a CG-3000 unit, rated for 300 watts.  Because of this, I can pretty much ignore SWR losses on the coax.

·         I am using LMR-400 coax for the 300 feet from my shack to the coupler.  (Yes, that’s extreme, but it’s what I had at the time).  That gives me about 0.5 dB loss at 3.5 MHz, if I remember correctly.

·         The CG-3000 is a fancy design that has been tested and found to be fairly lossless (we can assume 95% efficiency with this high impedance load on 80 meters).

·         I have 24 radials, 16 x 32 feet and 8 x 64 feet.  Although it is a half wave wire, it does still need a counterpoise.  Not as much as a quarter-wave, but it does have to be there.  With high impedance feed-point, we can neglect the losses at ground in this case and assume a radiation efficiency that is pretty high, like 90%. 

·         And no gain at low angles for a vertical wire, so let’s consider that factor of unity

 

So in my example above, the tuner losses can be ignored.

 

But what happens if we cut the wire down to 16 feet, we jump to 20 meters, move it indoors, use a small length of coax, and use a tuner at the rig?

 

1.       The wire will have a low impedance (quarter wave at 20 meters is 16 feet), so the tuner will have a harder time to make a match, and our efficiency will go down.  Let’s say 75%.

2.       The coax losses will be small, let’s assume 8 feet of coax from tuner to wire.

3.       No counterpoise, so the rig and the tuner and the human and nearby objects become the counterpoise.  The radiation efficiency goes WAY down, like to 25% or less.

4.       And what about the attenuation of the structure?  That could be minor (5-10%), or it could be huge (up to 50%).

 

In this second example, the losses from reduced antenna efficiency and the attenuation from the structure are way bigger than the losses of the tuner.  I would be more concerned about those losses and trying to improve those if possible.

 

Hopefully this helps. 

 

Oh, and congrats on that WSPR report.  Well done!  Who says there is no propagation? 

 

 

 

Mark Lunday, WD4ELG

Greensboro, NC  FM06be

wd4elg@...

http://wd4elg.blogspot.com

SKCC #16439  FISTS #17972  QRP ARCI #16497


Re: QRPp Antennas #u3s #antenna

TrueBlue
 

On Thu, Sep 6, 2018 at 02:10 PM, Alan G4ZFQ wrote:
QRO + losses is still more effective than QRP + the same losses.
So, yes.

That was my understanding of the logic.  I thought it must have been a trick question.

(EME people seem to view this efficiency quite fanatically.)


Re: QRPp Antennas #u3s #antenna

Alan G4ZFQ
 

an hour and twenty minutes of my 20m WSPR spots (4) from ZL1RS at a distance of 6569 miles, which is well over twice my U3S distance record.
Bob has a really good RX location and Beverages...

Do QRPp devices substantially lose power from antenna tuners and other peripheral whatnot beyond the concerns of a more QRO system?
The same amount. But QRO + losses is still more effective than QRP + the same losses.
The most effective QRP station has the biggest antenna with low losses. The best QRO station has....

73 Alan G4ZFQ


Re: QCX Frequency Calibration

Alan G4ZFQ
 

through some reading I was under the impression that by utilizing the frequency generator, I am able to to align the output frequency of the radio via Menu 8.5.
Erik,
Yes this is correct.
BUT you cannot use the CQX to check itself. The actual RF output, frequency counter and generator relies on the 8.5 setting. Change the setting on 8.5 and it all moves, the output changes but you do not see the result.
You must use some external standard or reference as described under section 8.5.
Or monitor your transmission on a WSPR receiver against other signals.

Or use GPS but be very careful if you do, keep your dummy load connected in case you do not read the instructions correctly.

73 Alan G4ZFQ


QRPp Antennas #u3s #antenna

TrueBlue
 

OK, so in these dismal periods of propagation, ~0.15W and an "antenna" that's merely a piece of wire strung up in the hallway with staples, what should show up on aprsinfo but an hour and twenty minutes of my 20m WSPR spots (4) from ZL1RS at a distance of 6569 miles, which is well over twice my U3S distance record.

I naturally assume he's on vacation in Las Vegas or somewhere, which would be about a typical signal, and has just failed to update his coordinates.  I asked him just in case and he informs me that, no, these are real New Zealand catches.  Well.

So,  it was enough to get me thinking about getting a proper antenna up again for a few minutes.

Frankly, I had just given up on ever getting an actual antenna design sorted out as it didn't look like anyone really knew what he was talking about.  The books are full of imaginary science and contradictions from amateur hobbyists ("Mr. Dunning, meet Mr. Kruger!").

Here's the latest physical argument that never reached a point of resolution:  Do QRPp devices substantially lose power from antenna tuners and other peripheral whatnot beyond the concerns of a more QRO system?  At least one position in this low-power efficiency debate is wrong.

This is but a small sample of what I'm up against.


Re: QCX 40

Paul Harrison
 

Thanks for explaining that Jim, it's very helpful. Everything seems to point to T1 and a dodgy solder joint. I will report my results. I was not able to complete the fault finding tonight due to other obligations, tomorrow hopefully i will have more time 

Paul DJ0CU.

On Thu, Sep 6, 2018, 22:24 James Daldry W4JED <jim@...> wrote:

Hi, Paul

Your radio has 2 detectors working "in parallel", either one of which will do a good job of receiving signals. There are 2 detectors because by adding the detected signals 90 degrees out of phase one sideband is canceled out - signals are all on one side of zero beat. Like a "real" radio. The 3 controls are to adjust the phase of the audio signals to exactly 90 degrees. The fact that they do nothing suggests that one of the signals is missing. Since T1 has been removed the probable cause is that one of the "little" coils has a bad connection on it. IC4 should show the same voltage on pins 7 and 9, and should show equal resistance to ground with an ohmmeter.

Jim W4JED


On 09/06/2018 06:06 AM, Paul Harrison wrote:
Hi,

I have a strange situation, after removing 5 turns from the large T1 Toriod to optimise the BPF output, I can now adjust the BPF to maximum without the Variable C being fully open. However when trying to adjust the 3 variable resistors nothing happens (cannot obtain minimums). No change whatsoever.

However, the RX seems perfectly sensitive and just the same as when I first constructed the kit. I am able to merrily make contacts no problems and having great fun.

Are the 3 variable resistors basically superfluous? I am loath to desolder the SMD
to try another IC.

Thanks in advance for any suggestions.

Paul DJ0CU.

 


Re: LCD top row is blocks, bottom row is blank

Jeff Whisler W9KW
 

Piotr,

My 40M QCX had the same symptoms.  I did carefully check my solder work as recommended and also performed a reset.

None of these steps made any improvement.  I asked Hans for a new EPROM which he promptly supplied.  All is well now.

Best wishes and good luck.

Jeff
WV9X



Re: QCX 40

James Daldry W4JED
 

Hi, Paul

Your radio has 2 detectors working "in parallel", either one of which will do a good job of receiving signals. There are 2 detectors because by adding the detected signals 90 degrees out of phase one sideband is canceled out - signals are all on one side of zero beat. Like a "real" radio. The 3 controls are to adjust the phase of the audio signals to exactly 90 degrees. The fact that they do nothing suggests that one of the signals is missing. Since T1 has been removed the probable cause is that one of the "little" coils has a bad connection on it. IC4 should show the same voltage on pins 7 and 9, and should show equal resistance to ground with an ohmmeter.

Jim W4JED


On 09/06/2018 06:06 AM, Paul Harrison wrote:
Hi,

I have a strange situation, after removing 5 turns from the large T1 Toriod to optimise the BPF output, I can now adjust the BPF to maximum without the Variable C being fully open. However when trying to adjust the 3 variable resistors nothing happens (cannot obtain minimums). No change whatsoever.

However, the RX seems perfectly sensitive and just the same as when I first constructed the kit. I am able to merrily make contacts no problems and having great fun.

Are the 3 variable resistors basically superfluous? I am loath to desolder the SMD
to try another IC.

Thanks in advance for any suggestions.

Paul DJ0CU.

 


Re: QCX Frequency Calibration

Erik
 

Alan, 
Thank you for the reply.
Maybe I'm not understanding the process correctly, but through some reading I was under the impression that by utilizing the frequency generator, I am able to to align the output frequency of the radio via Menu 8.5.
Not having a reference radio setup to verify my QCX output frequency against, I am assuming that with the lack of WSPR contacts, my output frequency is off slightly.(I have a measured RF output of 4W with 12-13VDC)
I read that the process to align the final RF frequency was to use the signal generator (9.6) and then verify the output on (9.5). Depending on 9.5's output, you can adjust 8.5 to bring the output frequency in line with the generated frequency.
This procedure has been discussed before. What I am experiencing is no change in my output frequency based off corrections made to 8.5.
If this is not the correct way to "tune/calibrate" the radio for output frequency correctness, do you know where I can find information on how to perform this?

73 de Erik - N9OJR


LCD top row is blocks, bottom row is blank

Piotr Skarpetowski
 

Hello,

I've already assembled recently received 40m band QCX TRX(T1.00e). When I power it up I am getting "LCD top row is blocks, bottom row is blank" as it is described in https://qrp-labs.com/qcx/qcxtrouble.html

I do have exactly the same behaviour as Karlis(https://groups.io/g/QRPLabs/topic/23353220?p=,,,50,0,0,0::relevance,,%23qcx,50,2,50,23353220,ct=1&ct=1) who described his problem.

When I am trying to listen about 20Mhz crystal, there is nothing, It is not oscillating.

Is there a way to check that AVR is not empty without using programmer? If not I will buy one.

--

73 de SQ5JUP,
Piotr


Re: Homebrew from scratch #ubitx

Curt wb8yyy
 

OM

no problem.  very nice work.  Perhaps you need a buffer amplifier in-between the PLL IC and the rest of the transmitter to provide more isolation.  Also - I am wondering if circuitry may perform better with ground planes (larger metal conductor) near the parts. 

enjoy your fine construction work. 

73 Curt WB8YYY


Re: QCX 40

Paul Harrison
 

Thanks Al and Hans for the tips, I did Increase the Audio to about midway but it had no effect, I was most surprised that R27 did nothing as it balances the output. I will attack the problem later this evening. I also wonder if the problem is related to the fact that WSPR transmissions are not being heard.

Paul. DJ0CU.


Re: Homebrew from scratch #ubitx

Dimitar Pavlov
 

Please, ask admin to delete/remove this my post in wrong place/group. Thank You in advance.


Re: QCX Frequency Calibration

Guido ON6RL
 

Hello Eric,

 

Maybe you can try with smaller steps in menu 8.5.

Start again with 27.004 and move up or down in steps of 1 kHz, or even smaller and see what happens.

If you have the opportunity, check your QCX tx freq on another receiver.

 

73, Guido, ON6RL.

 

 

 

 

Van: QRPLabs@groups.io [mailto:QRPLabs@groups.io] Namens E
Verzonden: donderdag 6 september 2018 02:43
Aan: QRPLabs@groups.io
Onderwerp: [QRPLabs] QCX Frequency Calibration

 

Hello all,
I just finished building my first 40m QCX.
The build was very fun and powered up great at first power. I setup WSPR and let it run for a few hours without any hits. I started to do some reading and troubleshooting and wondered if my frequency calibration wasn't a bit off. Reading the manual and searching the forums, I tried to calibrate the frequency using Menu 8.5. I am noticing that regardless of what I change this value to, the frequency on Menu 9.5 stays the same. For example, when setting Menu 9.6 to 6.000.000MHz and Menu 8.5 to 27.000.000 - 28.000.000, I always get a reading of 5.99821 on Menu 9.5. I have searched the forums and the Googles, but am lost.
Any ideas?

73 de Erik Morse
N9OJR


Re: ProgRock register corruption

Chris Smith <me@...>
 

I had this and found the problem was a slightly less than conservative regulator design on the Si5351A board. The drop out voltage is way too low for the operating point on the LM317LZ so it's borderline losing regulation all the time. Also if the source supply is 5V such as from USB it may actually vary from 4.4-5.4V or so depending on all sorts of parameters. Worst case dropout voltage could be 0.9V which isn't going to work. Plus the divider on the LM317 pushes the output to 3.48V, 180mV closer to the supply voltage than need be.

Ergo, I swapped the LM317LZ on the Si5351A board out for an MCP1700-3302E with a 1uF tantalum cap on input and output and mine has been reliable since. This is not a pin for pin replacement so you will have to bodge it.

I suspect this affects a percentage of units shipped. It's one of those engineering issues that doesn't become apparently until you ship several thousand units and get 10 complaints from those people who have the tetchy part of the LM317LZ batch.

I've added the worst case operating point to the datasheet here which is a cool English day with not much going on in the Si5351A, plugged into a naff USB supply:


Bad juju afoot. That's well under the safe zone.

Best regards,

Chris Smith M0XTE


On Thu, 6 Sep 2018, at 11:38, Mike W wrote:
Hi Folks,
I have a few Progrocks doing various duties here.
One of them has a strange problem that I can't fathom and could use some ideas
on the probable cause.

This fault has started recently, the progrock in question had previously
operated faultlessly for some months driving my 6m WSPR reporter receiver.

Its fitted with a FOX tcxo and usually stays very close to calibration, but
occasionally register two changes and calibration is destroyed.

I'm  controlling it over serial comms from an OrangePi Zero running Minicom.
There is no traffic or noise on the serial interface, other than what should
be there - monitored with another OrangePi and a scope on the serial input to
the progrock- its only register two that's being corrupted, and it happens in
an unrepeatable, irregular manner.

ideas on a ten bob note please :-)
thanks, Mike G8NXD





Re: QCX 40

Hans Summers
 

Hi Al, all

FYI the presets are 24-turn types.

Another thing: make sure gain control is not at minimum. The audio for the unwanted sideband adjustment is taken AFTER the gain control. So if you have gain at minimum this will also show the same symptoms (trimmers have no effect).

73 Hans G0UPL 

On Thu, Sep 6, 2018, 12:25 Al Holt <grovekid2@...> wrote:
Just a suggestion, but you might want to try getting all three back to about the center of each of their ranges and go through the procedure again. Since the resistors are multi-turn without real 'stops' it can be a little tricky. They have a subtle click to them when they are at the end of their range. I'm not sure if they are 10, 15 or 20 turn types. 

R27 is the I-Q balance control and should have the greatest effect on the null. The other two resistors help balance out high and low sides of the audio bandpass. 

I hope that helps!

--Al
WD4AH


Re: QCX Frequency Calibration

Al Holt
 

Since you have a 40m QCX you can use CHU on 7850 kHz for calibration if you're short of other test equipment. When you're 'spot on' the QCX will produce a 700 Hz tone. A smart phone with an audio spectrum analyzer app (usually free) can give you an idea how far off you are. Adjust and test; I'm never sure which way to go, but after a few trials you should get the hang of it. 

Once you're in the ballpark, transmitting WSPR, I'd suggest using K9AN, Steve Franke, for a frequency standard. He seems to cover most bands and his frequency reports are very accurate.

--Al
WD4AH


Re: ProgRock register corruption

Alan G4ZFQ
 

its only register two that's being corrupted, and it happens in
an unrepeatable, irregular manner.
Mike,

It seems unlikely the chip could be reset by a random COM input and less likely that it should always have the same effect.
There is a pin that needs grounding if you are not using GPS.
Otherwise I'd suspect a faulty memory.

73 Alan G4ZFQ