Keyboard Shortcuts
Likes
- Tinysa
- Messages
Search
Re: How to get a quick EMV approval with tinySA Ultra?
Using TRACE/CALC/TABLE TRACE you can enter a 8 point linear approximation of the antenna factor over frequency , save the table to the SD for future loading, and use the trace table to correct the measured data for the antenna factor by subtracting the trace defined by the trace table from the measured trace.
--
Designer of the tinySA For more info go to |
Re: How to get a quick EMV approval with tinySA Ultra?
¿ªÔÆÌåÓýHello Rainer,The problem is you would need to know the Antenna Factor of your test antennas at all test frequencies to make any meaningful measurements. Then there is a whole lot of math that needs to be applied to your results to get "accurate" field strengths of the emissions. Also, the TinySA does not include the required 9KHz RBW. It has 10KHz which is close enough for comparative testing but not for certification. That is why you need to pay for a pre-compliance test so that you have something to reference your measurements to. The TinySA has a nice feature that allows you to specify a Limit Line to the display which helps with visualising your results. As for using a certified power adapter to power your DUT, that is OK, but there is no guarantee your device won't "drive" the power adapter and hence get back into the mains supply. The other steps you suggest, Faraday room, etc are all good ideas and will enhance the repeatability of your test results, but you need something to reference against. Without an external reference, your results will, unfortunately, be meaningless. Hope this helps. Cheers...Bob VK2ZRE On 4/02/2025 12:01 am, Rainer Hantsch
wrote:
|
Re: Which to believe
When measuring a signal you need to ask yourself what you are measuring. Voltage or power? And when measuring voltage you can distinguish between RMS and amplitude.
A square wave is good starting point as it is easy to make an accurate level signal. The relationship between the amplitude of a square versus the amplitude of the fundamental is 4/pi (2.1dB). To convert a sine wave amplitude to RMS -3dB is taken into account. To convert the RMS voltage to power +13dB is used. ?
In your example you have 5mV amplitude (into 50 Ohm):
db20(5mV)+2.1-3=-46.9dBV. The tinySA can report in mV (dBmV), so +13.1dBmV If you measure power in dBm you should see -33.9dBm. You seem to measure -37.8dBm. That's outside the +/-2dB of the tinySA. |
Re: How to get a quick EMV approval with tinySA Ultra?
Since you know nothing about this system, you are biting off something WAY OVER YOUR HEAD.? You might be able to do a pre-compliance assessment, but that's all.? And.....you need to learn the ropes of making radiated and conducted emission measurements in a responsible and best -practices manner.? If you are also looking at the EU, there are a whole bunch of additional required tests addressing the susceptibility side of things.? A crash course from one of the few consultants giving classes on the subjects would be HIGHLY RECOMMENDED!!!? You are dealing with legally stipulated limits by the FCC and the EU (and other countries and markets).? Also you need to become up-close-and-personal with the regulations, testing methods, and the various world-wide regulatory agencies. ? Please walk before you attempt to run.? I've done EMC/RFI professionally for at least 40-years.? The discipline is not trivial.? Again, crawl, then walk, and finally run. Dave - W?LEV? On Mon, Feb 3, 2025 at 1:01?PM Rainer Hantsch via <office=[email protected]> wrote:
--
Dave - W?LEV |
Re: How to get a quick EMV approval with tinySA Ultra?
Hello, Bob.
Thanks for your explaining. I know that this is a complicated thing, but I bought? "tinySA Ultra"? to be able to find out if my developments are in safe distance to allowed limits of electromagnetic noise, or if (and where) my DUT excesses this limits.? This saves a lot visits for pre-compliance testing and therefore lots of money (-> the reason why I bought the ULTRA for). As far as I know, it is legal to do a self-declaration that a device is complying, but then it - of course - indeed has to be compliant. But knowing that my DUT is far below allowed limits before even doing a first pre-compliance test is always a good start. So I would do the testing by either moving to a location some kilometers away from buildings and mobile phone antennas, etc., move to a deep parking house, or similar, ? and do two scans: One without my DUT turned on (recording the environmental noise), and another one with my DUT turned on (recording environmental noise + DUT's noise). Or I build a Faraday cage using a fine and grounded metal mesh and do the above inside the laboratory. (will also reduce external influence) before. The difference should be my DUT then, right?? As my developments usually are small devices for automating or recording something, they are often battery operated. If not, a certified power adapter with cable length shorter than 1m makes life also lots easier.
?
This will at least give me information on noise generated by the DUT.?? Doing this math without an automated script is a horrible job, because tinySA (Ultra) by itself is not really easy to use (at least for me).
If additional tests are required (i.e. the Conduction Emitted Tests), this is another part I will have to do separately. But most important is the direct emission to air, because I often use plastic enclosures and digital electronics. Well, I already try to always use Microchip PIC controllers with internal oscillators, this drastically reduces radiation, but it is still digital, with lots of square waves therefore. ? |
Re: How to get a quick EMV approval with tinySA Ultra?
thank you for your help ? Mike ? ------------------------- Hello Mike and Rainer, |
Re: How to get a quick EMV approval with tinySA Ultra?
¿ªÔÆÌåÓýHello Mike and Rainer,Gentlemen, I suggest you search this group with terms like "EMI testing" as this has been discussed at great length previously. But the long and the short of this is that EMI/EMC Testing is an incredibly complicated subject. If your product is powered by the mains, there are also Conducted Emission tests that require additional specialised equipment such as a LISN. Look it up:-) There have been some very good links to some excellent tutorials and references on the subject also in this group. The best thing is to get your product pre-compliance tested by a registered Test Lab and use their report as your reference. Pre-compliance testing costs a lot less than a full Compliance Test. Note where your product fails and by how much. Then set up your DUT (Device Under Test) in your own lab and do your own testing. Note the levels you measure at the failure points noted in the Test Lab report and these become YOUR test reference levels. Then modify your design and aim to reduce your failure points by the amount the Test Lab result notes plus at least a 6dB margin, preferably a 10dB margin. When you think your product is right, take it back to the Test Lab and have another pre-compliance test done. Repeat this cycle until the pre-compliance test results all pass. Then you can get your Test Lab to do a full Certification Test. NOTE: You can NOT certify your own products unless you are a registered Test Lab and you would not be asking these questions if you were:-) It is vital that you understand the complexities of ensuring you duplicate your test setup on every subsequent test you do. Photographs of your test setup are your best friend. You need to ensure your DUT, every piece of test equipment and every cable is in exactly the same position each time and that every piece of electronic equipment in or near your lab is in the same on or off condition. This includes any type of lighting other than basic filament lamps. Preferably turn off and unplug any piece of electronic equipment not required for the testing. As a minimum, unplug any unnecessary electronic stuff, microwave ovens, printers, desktop computers, TVs, etc that are on the same power circuit as your lab as even when these things are "off", they can often have some degree of electronics running that can contaminate your test results. And mobile phones are a definite no-no! The TinySA is a great tool and can save you a lot of money if used correctly with appropriate accessories like antennas and E and H field probes to develop your product. But as noted earlier, you can not certify your own equipment. So welcome to the wonderful world of EMI/EMC Testing. It is a fascinating and at times, very frustrating, area of product development. (Been there, done that!) Enjoy your research! HTH...Bob VK2ZRE On 3/02/2025 10:26 pm, Mike Ward wrote:
|
Re: How to get a quick EMV approval with tinySA Ultra?
I also would love some instructions and maybe a script to generate a report on? self-certifications of developed hardware and/or to find out if my hardware will meet requirements for approval. ? I only run under Windows ? Mike ? -------------------------
|
How to get a quick EMV approval with tinySA Ultra?
¿ªÔÆÌåÓýHello. ? I bought my tinySA ULTRA in 2023 to do self-certifications of developed hardware and/or to find out if my hardware will meet requirements for approval. Though, I did not have any success yet. The ULTRA is calibrated and tested, but I have no idea on how to use it for this purpose in best way. Is a video tutorial available, showing the usual way/process of doing that? I am working to 100% on Linux (have a Linux workstation running openSUSE 15.5 x64). It? would be great if a program is available, generating a report with needed details by controlling the tinySA ULTRA over USB and finally generating a printable report for archiving. It would be great if somebody can help. Thanks in advance! |
Re: QtTinySA update to v1.0.2
#software
On Sun, Feb 2, 2025 at 11:39 PM, WyoHam wrote:
Well done.? I use kubuntu 24.04LTS on my desktop machine but it works on my Pi4 OS12 (which uses Wayland) and my ideapad with Lubuntu 24.10. It uses PyQt5 not PyQt6.? I will update to PyQt6 eventually. |
Re: How to find this signal with all its harmonics
Well, that got out of hand quickly.?
?
Just a remark on the subject of China-bashing: the Chinese are very well capable of producing top-quality products, if you ask them to. But consumers want things cheaper, not better. China just caters for this demand. If you want crappy non-compliant stuff they'll make it for you. 'You want a 10 dollar HT? Here it is. We could make it compliant but than it would cost you 12 dollar. It's up to you.' I own a China-made children's toy HT on PMR. It is on par with and often beats top-brand HT's. Still it was only €6,5 .
?
Back to the topic:
From the images it is 99% clear to me that the receiver is the RTL-SDR dongle. I too love this little RX and it sparked a lot of interest in RF with many people.
But: it isn't the best receiver ever. It has no real front-end, it is a broad-band direct conversion sampler with a single mixer directly connected to your antenna. And worse: you have to connect it to a device that is full of unknown clock sources.?Below is a picture showing a local repeater about 1 km away. Only one signal is real ...?
?
What I am trying to say is that you might see a lot of strange signals on your SDR, but chasing them with the SA might be like chasing ghosts.
?
|
Re: How to find this signal with all its harmonics
This is interesting. There has been some other conversations with similar suggestion I did read elsewhere. But I like to understand something and prevent accidental TinyAS damage. I have a directional antenna as well that I can use. What kind of attenuator is any should I use to prevent accidental TinySA damage from overpowering it specifically when you get closer to the source. I understand that there could be some signal at any given time that is not related to source you are looking for that might happen anywhere you standing. Would community here suggest something like this switchable 2W attenuator
If TinySA, attenuator and directional antenna are setup, could someone explain here in an algorithmic way how he would go safely search for the signal?
?
1. Set band start and end on TinySA.
2. What attenuator setting to start with. 3. What appropriate TinySA settings should be made related to the above attenuator value used.
4. What to pay attention on TinySA and make appropriate adjutements and try again starting from point #2 above.
?
If there are additional steps which should be added to the above, please add as much as you think it is relative to know.
If any of you have some good YouTube links related to this, please share
?
Thank you
? |
Re: QtTinySA update to v1.0.2
#software
Hey, Mikey! It works! ran: pip install --upgrade [name] Against all dependencies listed, a few updates accomplished and tried it again - it launches.? Now to try it with my updated TinySA Ultra connected Dependency?list:?? ? ? What I now have installed here altgraph-0.17.4 numpy-2.2.2 platformdirs-4.3.6 PyOpenGL-3.1.9 PyQt5-5.15.11? PyQt5-Qt5-5.15.16 PyQT5-sip-12.17.0 pyqtgraph 2.2.2 pyserial 3.5 Sorry for the clutter On Sun, Feb 2, 2025 at 4:28?PM WyoHam via <wyoham=[email protected]> wrote:
--
Tate "Success is not Final, Failure is not Fatal. It is the courage to continue that counts." ? -- W. C. |
Re: QtTinySA update to v1.0.2
#software
OK, that was my bad - i'd missed the need to update the Database,? Deleted the old one and tried?again.? Maybe a Fedora specific issue and it gets past the previous error, copies the new database, flashes on the screen and then fails?with: [user]@[machine]:~/QtTinySA/src$ ./QtTinySA.py QSocketNotifier: Can only be used with threads started with QThread QtTSAprefs.db copied from /home/[user]/QtTinySA/src/. to /home/[user]/.config/QtTinySA QtTSAprefs.db open: True ?Connection = "settings" QtTinySA v1.0.2 The Wayland connection experienced a fatal error: Protocol error Just to ensure I've not done something else ignorant, I'm going back through all the prerequisites, again. On Sun, Feb 2, 2025 at 4:19?PM WyoHam via <wyoham=[email protected]> wrote:
--
Tate "Success is not Final, Failure is not Fatal. It is the courage to continue that counts." ? -- W. C. |
Re: QtTinySA update to v1.0.2
#software
I believe you'd also have to use an older version of Linux,?or at least older versions of Python and Qt.? On Fedora 41 with Python 3.13.1 and Qt6 yields this: [user]@[machine]:~/QtTinySA/src$ ./QtTinySA.py QSocketNotifier: Can only be used with threads started with QThread Database QtTSAprefs.db found at /home/[user]/.config/QtTinySA QtTSAprefs.db open: True ?Connection = "settings" QtTinySA v1.0.2 Traceback (most recent call last): ? File "/home/[user]/QtTinySA/src/./QtTinySA.py", line 1805, in <module> ? ? checkboxes.mapWidget('checkboxes') ?# uses mapping table from database ? ? ~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^ ? File "/home/[user]/QtTinySA/src/./QtTinySA.py", line 1191, in mapWidget ? ? self.dwm.addMapping(eval(gui), int(column)) ? ? ? ? ? ? ? ? ? ? ? ? ~~~~^^^^^ ? File "<string>", line 1, in <module> AttributeError: 'Ui_MainWindow' object has no attribute 'marker1' On Sun, Feb 2, 2025 at 3:57?PM Ian Jefferson via <ian.jefferson=[email protected]> wrote:
--
Tate "Success is not Final, Failure is not Fatal. It is the courage to continue that counts." ? -- W. C. |
Re: QtTinySA update to v1.0.2
#software
On Sun, Feb 2, 2025 at 07:59 PM, JCR wrote:
It doesn't work on Windows version < 10 and it's 5 years since Microsoft ended support for Win-7.
You could install Linux as a dual-boot system if you do not wish to upgrade Windows.
|
Re: How to find this signal with all its harmonics
Ericd, you are one of the few these days!!? THANK YOU!!! My experience over the past decade (retired) and the last three decades (working the previous 2 to retirement, I have had many.....many instances of China just printing labels.? The last 4.5 + decades, I worked as an EMC/RFI engineer for a number of companies including HP and finally StorageTek (the last one I was actually requested to cheat - which I refused and led to firing).? Not on my watch but at HP, our scanners in Germany big-time failed due to "overlooking" and completely creaming reception of TV channels 2 through 6, including commercial FM due to radiated emissions.? Of course, that put us under the microscope and forbid HP from selling scanners in the whole EU for a year.? Another instance:? We upgraded our test computers to test our storage libraries at StorageTek (we had a good OATS).? They came with no regulatory stickers from SuperMicro.? We contacted their West Coast storefront about the issue.? They immediately ask, "How many do you need?"? A week later we received a business envelope full of "just printed" regulatory stickers. And another:? Requesting reports of products which failed formal testing.? Of all my requests, I received usually around 30% of those requested.? What's even more disgusting, none of the reports applied to the product I had requested.? All.....ALL.....of them were just compilations of pages from other unrelated regulatory reports with appropriate title pages.?? More yet:? I have tested, both officially and pre-compliance, many Chinese produces which miserably failed both radiated and conducted emissions per FCC and/or the EU regulations. ? Appliances:? The ultimate responsible party is the integrator of "components" (SMPS) which sail through customs as components.? Except for the German products we have in our 10-year old kitchen,? everyone fails both conducted and radiated emissions.? I had to install an extremely robust common mode choke on our second frig/freezer, a Kitchen Aid, to reduce both radiated and conducted emissions.? They go together as once the conducted emissions are violated, the whole house wiring lights up.? No surprise..... To make things worse, I'm an amateur radio operator and pursue radio astronomy as yet another hobby.? Get the point???!!! In my experience over the decades, China has taught even some of the good and upstanding companies the risk of cheating isn't that great with an FCC that no longer cares about enforcement.? Have a read of my QRZ page for a bit more.? I could fill anyone's email inbox with stories addressing China and the regulatory world.? And let's not mention S. Korea that uses the regulatory world as a trade barrier. Dave - W?LEV? ? On Sun, Feb 2, 2025 at 7:24?PM Erik Kaashoek via <erik=[email protected]> wrote:
--
Dave - W?LEV |
Re: QtTinySA update to v1.0.2
#software
I have Windows 7 professional 64 bits and the QtTinySa program does not work. Error: api-ms-win-core-path-l1-1-0.dll. Is there a solution? |
Re: How to find this signal with all its harmonics
On Sun, Feb 2, 2025 at 10:38 AM, W0LEV wrote:
China just prints labels and applies them appropriately with absolutely no attention or testing to international RFI/EMC regulatory laws.? Dave,
This may be a bit too general.
Much of the modern electronic equipment is manufactured in China, including from very well respected brands.
It may be better to state that very cheap stuf you can buy on Aliexpress and similar webshops may lack attention to international RFI/EMC.
But even that is not always true.
An example is the tinySA. It is cheap and available on Aliexpress and we took great care it is adhering the international RFI/EMC regulatory laws.
So, as a kind request, please do not generalize too much as it does not help uncertain buyers.
--
Designer of the tinySA For more info go to |
Re: Invitation: How high will your tinySA Ultra go?
Why not just set the "STEP" to the fundamental frequency.? Then, everything you request "STEP" in center frequency, the SA will increment to the next whole number multiple of the fundamental.? Dave - W?LEV On Sun, Feb 2, 2025 at 1:27?AM David Feldman via <wb0gaz=[email protected]> wrote: (pertaining to TinySA Ultra? non-+, type 405): --
Dave - W?LEV |