ETF5 question (1 Viewer)

[Richard_M]

This maybe in the wrong section, but I think most people that use these programs tend to hang around here.

Can someone please answer a question with etf5 software.

When you calibrate spl to say 80dB in ETF5 and run the test with a calibrated mic file etc. Does the output graph in the low freq response show that all frequencies at 80dB and above are what you can expect at an output of 80dB using an spl meter.

I hope this makes sense, I am trying to understand the results in a bit more detail, and understand how effective the output of subs actually are during normal listening levels. For example if the output shows say 85-90 dB at 20hz is this what you can expect if the main system is at an output of 80dB when watching a movie?

Richard

 

[DavidES]

If I understand you correctly, all frequencies would ideally have the same dB level deviating from reference level +/- 2 to 3dB over the entire 20Hz-20KHz range. Reference level is the output level around 1KHz.

In your example 20 Hz would be hot since it's greater than a 3dB deviation from reference level.

You do not want dips or peaks in the response graph greater than 3dB compared to reference level. Peaks are more noticeable especially adjacent to a > 3dB dip.

If you are able, post your graph in Speaker forum. Someone can help you interpret the results furthur if you need it.

Hope this helps

 

[DavidES]

You might find the excel spread sheet easier to use; found here.

Hope it helps

 

[Rick Guynn]

Richard, assuming you have the sub properly calibrated to your mains, then yes, your interpretation is correct. Depending on room size, it is not unusual to see a rise in response as the freq's go lower (depending on the sub's rolloff). It is a cabin-gain effect.

As a bonus, once you get used to using ETF, you can use it to blend your sub in with the mains by taking readings and studying around the crossover frequency.

RG

 

[Richard_M]

Thanks for the response Guys,

I am sorry for the confusion, but the ETF help file says;

"Input an SPL value into the Calibration SPL box for reference. 80 dB is a good value. If 80 dB is used and a measured signal is 3 dB louder, the resulting measurements will show an 83 dB level. Calibration using this method is not accurate for true SPL readings, but is very accurate for comparing levels between loudspeaker channels."


Because it says it is not accurate for true spl readings, and the result in their example is 83dB, due to the fact that the test level is no where near 80dB, at least on my system when I run the tests it is more like 60+dB this is why I was trying to come to grips with the actual output when the subs are having to output 80dB. i.e. are the results shown when tested at say 60+dB going to similar at higher levels. I also realise that alot of other factors come into play i.e. amp headroom etc.

The problem I have is that the sound card on the notebook cannot put out anymore power, and I don't want to alter the subs inputs due to having already calibrated them with an SPL meter.

Richard

 

[Rick Guynn]

That is why I stressed that "if your sub is calibrated to your mains". The SPL levels in ETF are relative, but if your sub is calibrated to your mains, and you play your mains at (for example) 80dB, the relative values still apply. Even if you were playing at 70dB, the relative differences would still apply.

RG

 

[Richard_M]

Thanks Rick

I was just a little confused.

If I can ask another question, if I input ETF into a line in of my amp say cd, what would you suggest being a level to run the tests at. For example at -10 on my amps volume I have a SPL reading of 85dB using video essentals test disk. If I ran the etf test signals at this volume would I damage my speakers?

Richard

 

[Rick Guynn]

I wouldn't pay too much attention to the relation between where your system is calibrated for normal use and where you set it for use with ETF. I will usually put my preamp about -10 below reference (reference being whatever it is set at when you calibrate with VE) then I use the volume control and channel balance on the computer to get sufficient signal strength (loudness on the test signal) while not overdriving the input.

RG

 

[Richard_M]

Thankyou Rick

Appreciate the help!

Richard