HMC-C200 DRO vibration sensitivity graph

the units of g^2/Hz is just as you have stated, power spectral density ( or Acceleration spectral density ASD). the point of this type of measurement is to remove the bandwidth (filter) from the measurement to normalize the data. you could use this data to calculate the gRMS at any bandwidth (filter) you chose.

EX. the graph shows a very small number (10^-8) for the PSD.

multiply that by the BW of you filter or the BW of interest, then take the square root to get gRMS.

Hi Jarrett

Thanks for your response. I understand what you mean, but it still is not clear to me how to interpret the graph. Maybe you can help by clarifying it to me further.

Let's say I have a DRO. Your graph tells me the sensitivity to vibration in the z-axis is

10^-8 g^2/Hz

at 100 kHz offset frequency. If my vibration occurs in a bandwidth of 100 Hz, that would give me a sensitivity of

10^-3 g or 0.001g

at 100 kHz offset frequency. What does that mean in terms of frequency deviation / phase noise under vibration? I don't understand how the amount of acceleration (g) or vibration (g^2/Hz) indicates the sensitivity of a particular device. What am I missing here?

Can somebody help me with this issue, please? Should I do more to clarify my question?

Thanks! I will have a look at the paper.

I can see you are frustrated.

the vibration sensitivity measurements have been contested before.

the best thing I think I can do is upload a paper for you to read and hopefully get a better understanding of our measurements.

this paper goes over PSG and G^2/Hz

Vibration-induced PM Noise in Oscillators and its Suppression

Archita Hati, Craig Nelson and David Howe

National Institute of Standards and Technology

Hi

I read through the paper you suggested and it has not fundamentally changed my understanding of vibration sensitivity and its measurement. The only graphs in the paper that have the same units as your graph (g^2/Hz) are graphs showing an amount of vibration. Nowhere are those units used to indicate vibration sensitivity. 

Is it possible that the graph in the datasheet is incorrect? I think its units should be 1/g. It is possible that the traces are correct if you just correct the units.

Do you perhaps have any direct measurements for this unit related to vibration sensitivity? E.g. phase noise measured under vibration conditions - the measurements that would have been done to generate the graph in the datasheet. I would appreciate it if you could send me those measurements, please.

good news and bad news.

good news is that we have discussed this back and forth a few times and we have come to realize that you are right!

the measured data should be in 1/g.

Bad news, is that now we have to change the datasheet.

thank you for calling us out on this and forcing us to unravel this internally.

we at ADI are always learning and always improving.