T/S Vas is not correct

The place to discuss general topics about Soundeasy.
User avatar
Chad Gray
Site Admin
Posts: 46
Joined: Wed Dec 02, 2020 7:26 pm
Contact:

Re: T/S Vas is not correct

Post by Chad Gray »

Thank for the information!

I have tried 9600 sample rate and was able to drop down to 65535MLS. If i go lower than 65535 my Zin graphs get a little shaky looking.

I will try this out for awhile.
Chad
dcibel
Posts: 39
Joined: Sun Dec 13, 2020 6:39 am

Re: T/S Vas is not correct

Post by dcibel »

65535 samples at 96khz = 682ms. I generally aim for 2 seconds, so just max out the MLS at 262143 samples at 96kHz = 2.73s. For impedance, there's less worry of external noise influence on the measurement, most speakers aren't that good at being microphones. I find ESS works much better for impedance anyway, at least at low frequencies. You just have to deal with the "instant" aka take a coffee break input preview. With MLS you can always run multiple cycles for averaging to lower the noise floor a fair bit. You don't get that luxury with ESS for whatever reason.

Realistically, 5Hz sine is 200ms long, you'd think that as long as MLS was longer than 200ms you'd be able to capture everything you need for impedance sweep. For acoustic measurement, the initial thought would be that MLS would need to be at least window width + acoustic distance. I've always just maxed it out, with the understanding that at 96kHz the MLS takes half as long as it does at 48kHz.
User avatar
Chad Gray
Site Admin
Posts: 46
Joined: Wed Dec 02, 2020 7:26 pm
Contact:

Re: T/S Vas is not correct

Post by Chad Gray »

If you do not mind me picking your brain on some of this... We can start a new thread if you like.

So MLS (Maximum Length Sequence) is the quantity of samples emitted by Soundeasy and Sample Rate is the amount of data that is ingested by the sound card to Soundeasy?

I want to start learning more about these terms and data so i can apply them to CSD measurements and how to great good data and receive good data. I mean I can setup it up and get pretty graphs of FR over time with CSD, but i need to learn more about MLS and Sample rates so i understand the data i am giving to CSD and reading from CSD.

Thanks!
Chad
dcibel
Posts: 39
Joined: Sun Dec 13, 2020 6:39 am

Re: T/S Vas is not correct

Post by dcibel »

MLS length is in samples, sample rate is the number of samples per second that is played back. It's really that simple, 96kHz means 96000 samples per second.

I'd suggest to read the manual pages starting of Chapter 16.

CSD plot is simply making many FFT plots by moving the start of the window forward a bit each time

Happy Holidays!
User avatar
Chad Gray
Site Admin
Posts: 46
Joined: Wed Dec 02, 2020 7:26 pm
Contact:

Re: T/S Vas is not correct

Post by Chad Gray »

Thanks Reid! Happy Holidays also! I will start reading more. I have some free time this weekend.

Chad
Post Reply