Electronics > Projects, Designs, and Technical Stuff

stable-amplitude sine in the range 100KHz – 10Mhz

(1/2) > >>

Hi everyone! My question is rather practical than theoretical.

I am planning the measurement which should last about one year under room temperature, and within it, I need a source of the stable-amplitude sine signal in the range 100KHz – 10Mhz.
The most important for me is the stability of the amplitude value. For example, I wouldn’t care if instead of 5 Volts it outputs 5.05V, if this value does not change more than 0.2% over the year every time I switch on the device.

The first idea would be to take the common arbitrary signal generator, but all of them (Keysights, Tektronix, R&S) have an amplitude accuracy of +-1%, which is quite a bit too high for me. However, the datasheets do not really mention how stable is this error, which gives me hope that the error may be stable. There is a term "stability" in some of the datasheets (usually 1ppm), but normally it refers to time-axis, isn't it?

Lords of the measurements, I summon you. Maybe there is someone really experienced with the stability of AWGs, or someone who knows another off-the-shelf solution.     

Thanks to everyone who read it! PS. have attached a few datasheets of AWGs

David Hess:
Check out page 63 of Linear Technology application note 47 where Jim Williams discusses two examples of RF leveling loops which will correct for variations of input, time, and temperature.

Interesting requirement, and likely not shown in typical AWG data sheets as mentioned.

Would think a quality AWG might be able to maintain this level over time if properly operated and environment maintained. One could perform a long term experiment using a high quality SA to verify the AWG performance over time.


I would think a good DDS-based synthesizer in that frequency range would be amplitude-stable over time.
There are cheap modules available that use the AD9850 DDS: https://www.analog.com/media/en/technical-documentation/data-sheets/AD9850.pdf
The ADI spec sheet does not quote long-term stability, since most users are more interested in harmonics, spurious outputs, and phase noise.
My guess is that running a 125 MHz clock rate to obtain a 10 MHz or lower output would be safe for your purpose, with a stable power supply.

Have you tried asking the AWG manufacturers direct? This sort of non-standard pre-sales support question is one of the things that distinguish the 'A' brands from the others


[0] Message Index

[#] Next page

There was an error while thanking
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod