• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

RTA ?

RaymondFromThePyrénéés said:
I thought i had to buy a hardware. Thank you
Click to expand...
Hardware RTAs exist, if you prefer that.

They're a lot easier to set up, but are usually way less configurable.

On the flip side, you can use a hardware RTA with your Stereo system, without having to boot up a computer.

But yeah, good hardware RTAs cost a pretty penny. For example the BDS PP-31 and PP-131.

Edit:
Here's a demonstration:
 
Last edited:
Here is his recommendation
 

Attachments

  • 20260425_232035.jpg
    20260425_232035.jpg
    512.8 KB · Views: 52
You are probably better off to stick with REW. Also if your signal is not 40dB above the noise floor it could end up being a poor exercise. Not sure where 20dB comes from - but in practice that is not what you would ever want. I did not write a book, but measured many subs over the years and resolution in low end is really problematic with simple mics like UMIK1 - unless you have a strong signal.

With subs, you can turn them up so you get some 50dB over 40dB which will be a typical daytime noise floor in most environments. Doing it late nights when noise floor drops in 30-ies is even better.

There are also some less scientific but potentially effective methods to locate optimal sub position. They are a bit odd, but actually could be quicker to locate the areas where you want to explore with actual measurements.

 
Thank you so much Oddball. This sounds like an idea coming from a genius and it really makes sens and is so much easier than REW that I still have to learn.
 
RaymondFromThePyrénéés said:
Thank you so much Oddball. This sounds like an idea coming from a genius and it really makes sens and is so much easier than REW that I still have to learn.
Click to expand...
Thanks - but definitely not genius :facepalm:. That would be more for @staticV3 and other senior members that understand all the nuts and bolts of audio. I try to help from more practical side as understand the practical problems people might be facing. But after crawling, you would still need to REW ;).
 
Oddball said:
You are probably better off to stick with REW. Also if your signal is not 40dB above the noise floor it could end up being a poor exercise. Not sure where 20dB comes from - but in practice that is not what you would ever want. I did not write a book, but measured many subs over the years and resolution in low end is really problematic with simple mics like UMIK1 - unless you have a strong signal.

With subs, you can turn them up so you get some 50dB over 40dB which will be a typical daytime noise floor in most environments. Doing it late nights when noise floor drops in 30-ies is even better.
Click to expand...

The problem with measuring bass frequencies is that the signal-noise ratio is very poor. This is because (1) the noise floor of a typical room rises at bass frequencies because rooms act as a low-pass filter, and (2) the output of most speakers drops off at low frequencies anyway. Since an RTA outputs less sound energy than a logarithmic sweep, it is even more prone to errors due to poor SNR. You can observe this by watching the meters jumping up and down. If you want to use an RTA, you have to turn averaging on and watch it slowly settle.

The SNR can be improved by (1) increasing the height (volume) of the signal, (2) increasing the width (time) of the signal, or (3) lowering the noise floor.

I do not recommend turning the volume up to improve the SNR. This is because loudspeakers are nonlinear devices - measure at a different volume, get a different frequency response. Voice coils heat up, they hit limits of cone excursion, etc. Drivers typically compress at high volumes, and they don't compress evenly over the frequency range. Audio systems should be measured at typical listening volume (yes, that is subjective). After all, what matters is how it performs when you use it. The only time you need to measure at a "standardized" volume, e.g. 76dB, is if you want to compare the performance of your system against others or if you want to perform formal testing.

This leaves the remaining two options. If you want the best SNR, choose a logarithmic sweep with a long sweep time. A 45 second sweep provides 90dB of noise rejection. Alternatively, take multiple sweeps from the same position and average them.

You need to balance the measurement method against the needs of what you are measuring. You don't need an excellent SNR if you just want to see what is going on. For this, shorter sweeps are more convenient, and even an RTA may be sufficient. Different signals and measurement methods give you different results and come with their own drawbacks. It is a waste of time to go through the painful procedure of obtaining an excellent quality measurement if you are doing a subwoofer crawl. I would use an RTA as Jim Clark suggests, it is much faster and more convenient. But if you want to DSP the result, you want the best quality measurement you can possibly take.

All this in my book in the signature.
 
And I will also do the crawling test to compare. Also, I will install a dedicated 220 line and do the sweep late ar night to avoid .traffic noise, etc.
 
Keith_W said:
The problem with measuring bass frequencies is that the signal-noise ratio is very poor. This is because (1) the noise floor of a typical room rises at bass frequencies because rooms act as a low-pass filter, and (2) the output of most speakers drops off at low frequencies anyway. Since an RTA outputs less sound energy than a logarithmic sweep, it is even more prone to errors due to poor SNR. You can observe this by watching the meters jumping up and down. If you want to use an RTA, you have to turn averaging on and watch it slowly settle.

The SNR can be improved by (1) increasing the height (volume) of the signal, (2) increasing the width (time) of the signal, or (3) lowering the noise floor.

I do not recommend turning the volume up to improve the SNR. This is because loudspeakers are nonlinear devices - measure at a different volume, get a different frequency response. Voice coils heat up, they hit limits of cone excursion, etc. Drivers typically compress at high volumes, and they don't compress evenly over the frequency range. Audio systems should be measured at typical listening volume (yes, that is subjective). After all, what matters is how it performs when you use it. The only time you need to measure at a "standardized" volume, e.g. 76dB, is if you want to compare the performance of your system against others or if you want to perform formal testing.

This leaves the remaining two options. If you want the best SNR, choose a logarithmic sweep with a long sweep time. A 45 second sweep provides 90dB of noise rejection. Alternatively, take multiple sweeps from the same position and average them.

You need to balance the measurement method against the needs of what you are measuring. You don't need an excellent SNR if you just want to see what is going on. For this, shorter sweeps are more convenient, and even an RTA may be sufficient. Different signals and measurement methods give you different results and come with their own drawbacks. It is a waste of time to go through the painful procedure of obtaining an excellent quality measurement if you are doing a subwoofer crawl. I would use an RTA as Jim Clark suggests, it is much faster and more convenient. But if you want to DSP the result, you want the best quality measurement you can possibly take.

All this in my book in the signature.
Click to expand...
We have to distinguish though, as (from REW's Help) :

The Mode can be set to Spectrum for a spectrum analyzer plot or to various RTA resolutions from 1 octave to 1/48 octave. The difference between spectrum and RTA modes is how the information is presented. In spectrum mode the frequency content of the signal is split up into bins that are all the same width in Hz. For example, with a 64k FFT length and 48 kHz sample rate the bins are 0.732 Hz wide. The plot shows the energy in each of those bins. In RTA mode the bin widths are an octave fraction, so their width in Hz varies with the frequency.
Click to expand...

Correcting narrow ranges with RTA window may not be the best idea, no matter the resolution if done in RTA and not at spectrum mode as amplitude is presented in different way, and depending how we correct (which software, etc) the info we insert to it may not be what it needs.
 
An example of the two modes and their Delta (purely electrical) :

delta.PNG


Look at how easy one can end up with a pair of chainsaws instead of speakers with the wrong mode.

Edit: and another with aligned SPL at 20Hz and full PN at same resolution 1/48 for better view:

delta.PNG
 
Last edited:
Jim Smith writes two pages on the problems occurring by using circuits of other appliances and since I am anyway installing a dedicated circuit for a sauna (a swede cannot live without a sauna) it takes little effort to install 2 circuits on the same time from the fuse box.

What I wonder though is how much wattage I should count for this HIFI circuit ?
I guess if in the future if I want to buy an amp for 2 or more subs,
a 500 watt amplifier would be the maximum ?
 
also, all the cables in this house are old, perhaps installing a new fresh cable cannot do any harm.

And I have a friend electrician who will do all the work for little money.
 
You must log in or register to reply here.

Similar threads

daniboun
Bromo/Fidelia hybrid Bookshelf speakers DIY (pictures /measurements inside)
Replies
4
Views
1K
daniboun
daniboun
O
Speaker Level Test & Measurement Interfacing - Correct Method
Replies
5
Views
517
MaxwellsEq
MaxwellsEq
M
Help with a customer - LCR speakers with Acoustic Transparent Screen
Replies
0
Views
371
meesbaker
M
Keith6690
OPPO DVD
Replies
5
Views
417
Keith6690
Keith6690
T
Noise floor level changes depending on which app is in foreground
2 3
Replies
46
Views
2K
DavidR
D
Back
Top Bottom