Author Topic: Help deciding on a tool to run frequency response curves / graphs  (Read 2257 times)

0 Members and 1 Guest are viewing this topic.

Offline mapleLCTopic starter

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: us
I am in the process of setting up my lab, and I have the requirement to run frequency response graph on different types of components.

My understanding is that a Spectrum Analyzer is required for this?  I see some of these units out there, they look 20 or 30 years old.  Has this need I have been solved a better way from then until now?  Perhaps using a PC, or a cheaper piece of lab gear?

I am seeing these analyzers over $1300, which is too much for the frequency with which I need to use it, but I do need it, and also need it to be pretty accurate.
 

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #1 on: April 03, 2022, 01:50:27 pm »
What kind of frequency do you need? Do you need the level, or also the phase? If you are talking audio frequencies, then a (USB) sound card with the right software (like RMAA, REW or Daqarta) might well be sufficient. The software will usually give some recommended sound cards that are known to work well with it. The resolution and dynamic range should be quite good (better than a scope).

For up to a few dozen MHz a function generator plus digital storage scope (DSO) might work. The Digilent Analog Discovery is more convenient and should have a higher resolution than your normal DSO? For RF and higher I'd think either a spectrum analyzer with tracking generator, or even better a vector network analyser (VNA). There are some very cheap hobby-grade VNAs that might be sufficient like NanoVNA and LiteVNA. This forum has some long threads about them. They are not super accurate, but you'd have to quantify what you need.

« Last Edit: April 03, 2022, 03:23:21 pm by alm »
 

Online TimFox

  • Super Contributor
  • ***
  • Posts: 9000
  • Country: us
  • Retired, now restoring antique test equipment
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #2 on: April 03, 2022, 02:12:35 pm »
Watch out for the details on spectrum analyzers:  most units now in production can go to very high frequencies > 1 GHz, but only down to 9 kHz.  Also, they typically have 50 ohm input impedance.
 
The following users thanked this post: mapleLC

Online Grandchuck

  • Frequent Contributor
  • **
  • Posts: 754
  • Country: us
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #3 on: April 03, 2022, 03:15:47 pm »
Tim is correct.  The graphs you posted are in the audio range.  Perhaps a scope with a built-in function generator would serve well.  The sound card idea is also a good one.
 
The following users thanked this post: mapleLC

Offline mapleLCTopic starter

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: us
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #4 on: April 03, 2022, 03:48:29 pm »
Watch out for the details on spectrum analyzers:  most units now in production can go to very high frequencies > 1 GHz, but only down to 9 kHz.  Also, they typically have 50 ohm input impedance.

Tim is correct.  The graphs you posted are in the audio range.  Perhaps a scope with a built-in function generator would serve well.  The sound card idea is also a good one.

https://eevblog.com/forum/testgear/rigol-ds1054z-function-generator-fft/

My scope doesn't have it... argh.  It's a great idea, that way the scope can serve 2 needs.

Regarding the sound card, that would serve for line level components, but I don't see how I could it with a set of speakers... would u just eliminate the voltage and pass the signal right to speakers?

 

Offline pope

  • Frequent Contributor
  • **
  • Posts: 370
  • Country: pl
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #5 on: April 03, 2022, 04:28:45 pm »
Watch out for the details on spectrum analyzers:  most units now in production can go to very high frequencies > 1 GHz, but only down to 9 kHz.  Also, they typically have 50 ohm input impedance.

Tim is correct.  The graphs you posted are in the audio range.  Perhaps a scope with a built-in function generator would serve well.  The sound card idea is also a good one.

https://eevblog.com/forum/testgear/rigol-ds1054z-function-generator-fft/

My scope doesn't have it... argh.  It's a great idea, that way the scope can serve 2 needs.

Regarding the sound card, that would serve for line level components, but I don't see how I could it with a set of speakers... would u just eliminate the voltage and pass the signal right to speakers?

So you want to measure the frequency response of a speaker? For starters you need a very flat omni microphone and an anechoic chamber. Good luck with the latter :)
 

Offline KevinA

  • Regular Contributor
  • *
  • Posts: 57
  • Country: gb
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #6 on: April 03, 2022, 04:43:19 pm »
Only one option here, Audio Precision.
 

Offline cdev

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #7 on: April 03, 2022, 04:49:38 pm »
You could use a script with a signal generator and USB scope to make it easy and standardized to run the same tests on different equipment and make it easy to photograph to facilitate the realistic comparison of the results. Thats what the various magazines do.

Start each test by resetting to initialize everything, then set the test up and run it.
"What the large print giveth, the small print taketh away."
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9963
  • Country: us
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #8 on: April 03, 2022, 07:15:21 pm »
The attached graphs are from a Digilent Analog Discovery 2.  You can find the specs here:
https://digilent.com/shop/analog-discovery-2-100ms-s-usb-oscilloscope-logic-analyzer-and-variable-power-supply/

It works up to about 10 MHz
 

Offline mapleLCTopic starter

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: us
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #9 on: April 04, 2022, 01:04:54 am »
Only one option here, Audio Precision.

Can you elaborate please?
 

Offline mapleLCTopic starter

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: us
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #10 on: April 04, 2022, 01:08:13 am »
The attached graphs are from a Digilent Analog Discovery 2.  You can find the specs here:
https://digilent.com/shop/analog-discovery-2-100ms-s-usb-oscilloscope-logic-analyzer-and-variable-power-supply/

It works up to about 10 MHz

All those pins... I'd be... |O
 

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #11 on: April 04, 2022, 07:53:45 am »
For the frequency response test you'd only have to use the BNC connectors on this adapter board
 
The following users thanked this post: mapleLC

Offline mapleLCTopic starter

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: us
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #12 on: April 04, 2022, 08:05:37 am »
For the frequency response test you'd only have to use the BNC connectors on this adapter board

Oh, so this adapter plugs into ports in the scope and the other 2?  The unit being tested?  I like the idea.

Can this be adapted to test speakers?
 

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #13 on: April 04, 2022, 08:29:07 am »
I guess, if you added an amplifier on the output and microphone with a very linear frequency response. But unless you're dealing with an ultrasonic transducer that needs the extra bandwidth, I think you'd be getting better performance with a good sound card.

Offline pope

  • Frequent Contributor
  • **
  • Posts: 370
  • Country: pl
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #14 on: April 04, 2022, 12:33:20 pm »
If you want to measure the response of speakers, a decent soundcard and REW would do the job just fine. But as I already mentioned, you need a very flat microphone and access to an anechoic room. Neither is cheap.
 

Online themadhippy

  • Super Contributor
  • ***
  • Posts: 3264
  • Country: gb
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #15 on: April 04, 2022, 12:52:15 pm »
If you've got a spare time machine nip back to 2003 and download a copy of the tms-1 software from dazyweblabs,unfortunately all traces of it seem to have vanished.
 

Online HKJ

  • Super Contributor
  • ***
  • Posts: 3086
  • Country: dk
    • Tests
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #16 on: April 04, 2022, 01:33:31 pm »
It is always possible to use TestController for it, it can control soundcard for input/output and can also combine soundcard with other equipment.

Download at bottom of page: https://lygte-info.dk/project/TestControllerIntro%20UK.html
 

Online 5U4GB

  • Frequent Contributor
  • **
  • Posts: 623
  • Country: au
Re: Help deciding on a tool to run frequency response curves / graphs
« Reply #17 on: October 08, 2023, 10:31:50 pm »
Updating an older thread, the Analog Discovery 2 has been discontinued, it's now replaced by the Analog Discovery 3.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf