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
QMX (80-20m) Rev 2 Harmonics
|
ReefKeeper2K23
Good evening,
I just finished assembling my QMX (80m-2m) rev 2 kit a couple of days ago and I'm in the process of testing it out. This is my first HF transciever as well as my first kit build so please bear with me. I hooked it up to a dummy load and then to my Tiny SA Ultra spectrum analyzer to see how clean the signal was on the various bands after calibrating the analyzer. I am seeing elevated harmonics on all bands that are suppressed to some extent but if I understand correctly not enough to be in compliance here in the US with FCC rules. Off the top of my head the level of the nearest harmonic was on average 27-32db below the fundamental frequency. Runing the latest firmware (1.00.014) and I've performed the suggested diode modification already. Here's a screenshot off of the Tiny SA Ultra doing a harmonics test at 14.074Mhz. If I'm reading this correctly the 2nd harmonic is 28.6db below the fundamental frequency. Please correct me if I'm wrong. Has anyone else tested their QMX and seen simillar results with regards to harmonics? If so how did you address it? I'm wondering if it's something I perhaps did wrong in assembly, configuration or if it's common to this particular kit.? Suggestions? Thanks, -Bryan |
|
Hi Bryan You didn't say how you connected the TinySA to the QMX. QMX would need a 50-ohm dummy load and proper attenuators. Just putting the TinySA nearby doesn't provide good enough measurements. So please check that as you need to find out if this is the problem or if there is an actual problem with your QMX. 73 Hans G0UPL On Sat, Jan 13, 2024, 8:05?AM ReefKeeper2K23 <bgraf2311@...> wrote: Good evening, |
|
ReefKeeper2K23
For testing I had it connected via a short low-loss coax patch cable (BNC to BNC), that connects to a pigtail (BNC to SMA), then the 25w attenuator, and another short pigtail (SMA) into the Tiny SA Ultra.
Overhead of the general layout. One short pigtail could be removed if I had the appropriate connector but this is what I had on hand: |
|
Bryan, How much attenuation does that 25W load provide? JZ KJ4A? On Sat, Jan 13, 2024, 12:32?AM Mike McCarty <mmccarty@...> wrote: That is showing that the 2nd harmonic (28.12 MHz) is 16 dB below the fundamental frequency.? It's showing the level of the fundamental in dBm.? The others are in dBc, relative to the fundamental.? I'd suspect there is something wrong with the measurement. |
|
That is an excellent result, Bojan. Bryan, Is it possible that your " attenuator " is not that at all, wherein the output BNC is in fact a peak voltage rectifier port for measuring power? JZ On Sat, Jan 13, 2024, 6:27?AM Bojan Naglic <bojan.naglic@...> wrote: Hi, |
|
Hello Bryan,
Welcome to the building community! From the pictures you posted, the TinySA Ultra is overloaded.? Note the HARM message on the left of the display.? That is telling you the input is of a high enough level that you may be causing damage to the TinySA input.? It would be best to run the self-tests to verify that you have not damaged the TinySA. The maximum input should be below 0 dBm or 1 milliwatt.? That is for protection.? The TinySA series works best (is most accurate) when the input signal is below -25DBm.? A 5 watt signal is 37dBm.? To get below -25dBm, you need 37+25=62dBm of total attenuation.?? I also suggest you build or purchase a step attenuator between the output of your higher-power fixed attenuator and the TinySA.? As with all test equipment, verify that the power ratings match the measurement setup. I use one like this, rated for 2 watts: There is a kit available that is created for 5 watts with fewer total attenuation steps: I have used both. Here is a link to the TinySA Ultra specifications: There are many examples and YouTube videos in the WiKi.? I would suggest that you scan them to see if the measurement you want to do is on one of them: I test all the kits I build (QCX, QDX, QMX, 50w PA).? I can say that none of the QRP-Labs kits I built have ever been out of FCC specifications as long as there were no build errors or damaged parts.? That is not true for all the non-QRP-Labs kits that I have built.? Some are very much out of spec.? I am glad that you are testing your kits for compliance. 73 Evan AC9TU |
|
A very common measurement problem is overdriving a spectrum analyzer.
The general rule of thumb is to not apply more than 0 dBm into the spectrum analyzer lest one see far more harmonic content than is actually present--exactly what your photo suggests. With a 5W transmitter, a 40 dB attenuator is a good, off-the-shelf choice for meaningful spectrum analyzer measurements. By the way, you are in good company. I've had to deal with reports of non-compliant harmonic content measurements made by a "very famous magazine's" reviewer using $100k+ equipment. It was this problem: reduce the input with attenuators and those horrible "harmonics" disappeared. -- 73 de Bob, K6XX -- 73 de Bob, K6XX |
|
ReefKeeper2K23
The green 'HARM' text on the left of the display in the screenshot actually appears on my Tiny SA when the unit is placed in harmonics testing mode: Menu>Measure>Harmonics. You then input the fundamental frequency and the desired display range if any:
It sounds like I need another attenuator. I'll see what they have locally and report back when I get a chance to retest. |
|
Did you run the self-test?
I am sorry for my misinterpretation of the Harm message.? I would still do the self-test. Looking at your picture again, two points come to mind: 1 - Are there male and female connectors on the attenuator?? The male connector goes to the transmitter, and the female to the test instrument.? The better attenuators are marked.? For some, the direction does not matter. 2 - Be careful when testing the QMX without a case supporting the boards.? I fried one while trying to isolate a control board issue.? That one is now spare parts. At some point, you might want to purchase a nanoVNA.? If I get a questionable reading, I use my nanoVNA to validate the cables, attenuators, and connectors. 73 Evan AC9TU |
|
ReefKeeper2K23
Just realized that replies here are not in-line and get stacked on the end. Sorry for the concurrent messages.?
I did run the self-test on Tiny SA Ultra (all passed). Last night I also updated the firmware to the latest build, returned it to default settings, and ran the calibration again which passed. |
|
From a little Googling it appears to me the 10 in the part number means 10dB and the 25 means 25W. You'd need much more attenuation than 10dB to avoid totally overloading the TinySA. 73 Hans G0UPL On Sat, Jan 13, 2024 at 11:09?PM ReefKeeper2K23 <bgraf2311@...> wrote: The attenuator does have an 'in' marked on one end and an 'out' marked on the other. Photo below: |
|
That one should work, though I would connect the first one in a series to get 50dB of attenuation (37dBm-50dB = -13dBm).? This is under the maximum that the TinySA Ultra can handle but still not quite enough to meet the -25dBm accuracy threshold.? Since the 10dB attenuator is rated for 25 watts, it would be first in line (closest to the transmitter).
A step attenuator is recommended for the last amount needed.? This also allows you to verify that the TinySA is not overloaded by switching in and out a fixed attenuation and noting how the scan changes.? 73 Evan AC9TU |
|
I got tired of worrying about power ratings of attenuators and accidentally putting too much power into a device so I built one of these:?
After testing with a NanoVNA, I know the worst case out of the tapped side.? Works well for my purposes and I can run 100 watts through it into a dummy load, with low power attenuation devices on the tapped side as necessary. 73, Brock VA7AV |
|
Hi there,
A quite common mistake at using the broad band instrument, as the spectrum analyzer is, also is when the dynamic range of the instrument is not sufficient to measure weak signals whereas the main signal would be too strong for the instrument front-end. In this case the measured result could be quite wild if not even catastrophic for the instrument input circuits. In such situation after the power attenuator an additional narrow band filter (a band stop filter) is needed to eliminate the strong main signal. -- 73 Bojan S53DZ |
Bob Wolbert, K6XX
to navigate to use esc to dismiss