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
Output impedance of RF sinewave generator (Tiny Ultra)
|
Today, as I was going to measure the frequency (sensitivity) plot of a 500 MHz Keysight oscilloscope, I found that the measured amplitude was going to increase by approx. 6 dB from approx. 150 to 300 MHz, and then drops down somewhat. Despite a direct connection using SMA-BNC adapter and direct coupling without coax cable and a 50 ohms scope input. At any case: the behaviour got (expectedly) a bit worse when I selected a high impedance input while using a BNC T-connector for termination with a BNC 50 ohms termination resistor. After doing some experiments, I discovered this: with the latter configuration I also switched the internal 50 ohms to active state (resulting termination: 25 ohms). It was a nice surprise that the amplitude shown on the screen was really constant over the whole range that was used (up to 500 MHz), while the amplitude under 150 MHz was fairly the same as in the case of a single 50 ohms termination. The question: what about the output impedance of the generator?? Rob |
|
That means that the output is low impedance, i.e. not 50 ohms? Could I insert a resistor at an appropriate place behind the mux, in the signal path from the generator? On Thu, 20 Mar 2025 21:43:59 +0100, Erik Kaashoek via groups.io <erik@...> wrote:
-- Erstellt mit Operas E-Mail-Modul: |
|
Please do not modify the tinySA HW, you will make it worse.
The output impedance is roughly 50 ohm but for best impedance matching you can consider adding a 6 dB external attenuator.
I never do this myself as the output impedance without external attenuator is good enough for me.
Keep in mind the output level is not adjusted when doing calibration, by default is uses the factory settings. There are several settings you can adjust to improve the output level accuracy if needed.
Have a look at the wiki.
?
--
Designer of the tinySA For more info go to |
|
Of course, adding a resistor in an RF-path with impedance target is quite dangerous (only 0402 would constitute a limited risk). Regarding the 'roughly' 50 ohms output impedance: I forgot to mention that an only very minor drop in amplitude below 150 MHz when the load termination is halved (related to the first approach with one termination resistor should be an indication of an output impedance lower than - say - 20 ohms. Before doing final measurements, I have updated the firmware (for the first time) - the firmware could have improved things, you never know. I'll order a 50 ohms BNC attenuator (yet I see that they are simple series resistors and not a black box with voltage divider - that fits exactly). A good opportunity to also order the defective LNA. -- An infrequent user of tinySA4 ? On Fri, 21 Mar 2025 09:59:19 +0100, Erik Kaashoek via groups.io <erik@...> wrote:
|
|
Using an external attenuator is the preferred method with any RF generator.
?
Some measure the output and use it as feedback to control the RF level, so that it is consistent over a large frequency range.
?
This can be messed up when feeding highly reactive loads, so by adding an attenuator into the path, it helps ensure the generator always sees something close to 50 ohms, regardless of the actual load.
?
Don't consider modifying the Ultra, you will just add a whole load of additional problems.
?
Regards,
?
Martin
?
?
On Fri, Mar 21, 2025 at 10:24 AM, R. Maris wrote:
I'll order a 50 ohms BNC attenuator |
to navigate to use esc to dismiss