proper testing methodology?

chico1st · 26th July 2011, 01:44 AM

OK i just want to make sure that I have speaker testing down right.

First I get the T&Small parameters of my driver and use that to figure out my enclosure size.

I can get my amp response using squarewave input and an oscilloscope on the output. Somehow i fix any issues there.

Then I build the enclosure and measure its output using some frequency sweep from a soundcard/audio-software with a mic. Then fix those peaks etc.

Is this right?

Loren42 · 26th July 2011, 02:13 AM

That is about the 20,000 foot view of it.

chico1st · 26th July 2011, 02:24 AM

I just want to make sure im not missing any bits.
Is there a point in testing the speaker response with no enclosure?
Also do I test SS and tube amps with no load or have my speakers attached? or does it matter?

Loren42 · 26th July 2011, 02:29 AM

You may find this link helpful:

Passive Crossover Network Design

If your goal is to build a box and design a crossover for the drivers you have, then getting the actual T/S parameters of the drivers is the first step.

Then testing the response SPL (on and off-axis) for each driver in the final enclosure is next.

Given that data you can begin the crossover work.

Then go in and verify your design by additional measuring.

Finally, tweaking the design based on those measurements and repeating your measurements again.

Loren42 · 26th July 2011, 02:32 AM

Testing an amp should always be done under a load. I would use a dummy load because it is not reactive (i.e., impedance is uniform across the audio spectrum).

I would not use speakers to test any amp. The load impedance varies with frequency and they make a hell of a noise when testing maximum power output!

chico1st · 27th July 2011, 11:43 AM

ok i've been looking through this but I have one question.
Is the typical way to get an impedance curve to put in a known frequency to your speakers then measure rms Voltage and rms Current and make a plot with a number of points? or is there something better?

Loren42 · 27th July 2011, 12:08 PM

I think you are right.

However, there are programs that you can download to automate the process.

Praxis by Liberty Instruments will measure impedance and T/S constants in it "Demo" mode. You need to make a wire harness with a resistor.

Liberty Instruments, Inc. Home Page

There are also dedicated hardware/software products like Woofer Tester 2 (Smith & Larson $150) and a competing product called WT3 (Dayton Audio at Parts Express $100) that do the same thing and may be more accurate - I don't know which is really better.

Once you progress to the phase of making measurements of the frequency response, HolmImpulse is a popular free program for making measurements of the SPL and phase.

You can also find a lot of programs for design of the box and crossover on the web as well. Some packages such as SoundEasy or SpeakerWorkshop do it all. SoundEasy is about $250 and SpeakerWorkshop is free, but very hard to use. More sophisticated programs exist and the cost escalates like crazy.

There are some advantages to doing some of these calculations by hand. You will get a better understanding of the nuts and bolts of the process doing it from pencil, calculator, and paper rather than simply pushing a button and automagically generating a design. The automagic function many times can produce something you do not want and having the experience and knowledge of the low level details helps spot mistakes.

AndrewT · 27th July 2011, 12:37 PM

If you want to understand bass loading of speaker drivers then you must at some stage do it all by hand method and by hand calculation.
Once you have achieved a sufficient understanding then you can adopt the "quick" computer methods.
I suggest "Bullock on Boxes" (isbn 0-9624191-5-x) as a good reference document to gain that initial understanding. It even shows how to automate some of the measuring/testing/calculation using your PC. His book is based on articles published in "Speaker Builder" during the 1980s. It predates most of the computerised design and testing that was/is available.

26th July 2011, 01:44 AM	#1
chico1st diyAudio Member Join Date: Sep 2010	proper testing methodology? OK i just want to make sure that I have speaker testing down right. First I get the T&Small parameters of my driver and use that to figure out my enclosure size. I can get my amp response using squarewave input and an oscilloscope on the output. Somehow i fix any issues there. Then I build the enclosure and measure its output using some frequency sweep from a soundcard/audio-software with a mic. Then fix those peaks etc. Is this right?

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Proper way to get mono?	pinobot	Analog Line Level	3	25th March 2010 02:38 AM
multiple output methodology	gearheadgene	Power Supplies	7	5th February 2006 03:23 AM
Looking for proper transformer.	contaxchen	Parts	0	15th September 2004 02:38 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

26th July 2011, 02:13 AM	#2
Loren42 diyAudio Member Join Date: Sep 2008 Location: "Space Coast" Florida, USA	That is about the 20,000 foot view of it.

26th July 2011, 02:24 AM	#3
chico1st diyAudio Member Join Date: Sep 2010	I just want to make sure im not missing any bits. Is there a point in testing the speaker response with no enclosure? Also do I test SS and tube amps with no load or have my speakers attached? or does it matter?

26th July 2011, 02:29 AM	#4
Loren42 diyAudio Member Join Date: Sep 2008 Location: "Space Coast" Florida, USA	You may find this link helpful: Passive Crossover Network Design If your goal is to build a box and design a crossover for the drivers you have, then getting the actual T/S parameters of the drivers is the first step. Then testing the response SPL (on and off-axis) for each driver in the final enclosure is next. Given that data you can begin the crossover work. Then go in and verify your design by additional measuring. Finally, tweaking the design based on those measurements and repeating your measurements again.

26th July 2011, 02:32 AM	#5
Loren42 diyAudio Member Join Date: Sep 2008 Location: "Space Coast" Florida, USA	Testing an amp should always be done under a load. I would use a dummy load because it is not reactive (i.e., impedance is uniform across the audio spectrum). I would not use speakers to test any amp. The load impedance varies with frequency and they make a hell of a noise when testing maximum power output!

27th July 2011, 11:43 AM	#6
chico1st diyAudio Member Join Date: Sep 2010	ok i've been looking through this but I have one question. Is the typical way to get an impedance curve to put in a known frequency to your speakers then measure rms Voltage and rms Current and make a plot with a number of points? or is there something better?

27th July 2011, 12:08 PM	#7
Loren42 diyAudio Member Join Date: Sep 2008 Location: "Space Coast" Florida, USA	I think you are right. However, there are programs that you can download to automate the process. Praxis by Liberty Instruments will measure impedance and T/S constants in it "Demo" mode. You need to make a wire harness with a resistor. Liberty Instruments, Inc. Home Page There are also dedicated hardware/software products like Woofer Tester 2 (Smith & Larson $150) and a competing product called WT3 (Dayton Audio at Parts Express $100) that do the same thing and may be more accurate - I don't know which is really better. Once you progress to the phase of making measurements of the frequency response, HolmImpulse is a popular free program for making measurements of the SPL and phase. You can also find a lot of programs for design of the box and crossover on the web as well. Some packages such as SoundEasy or SpeakerWorkshop do it all. SoundEasy is about $250 and SpeakerWorkshop is free, but very hard to use. More sophisticated programs exist and the cost escalates like crazy. There are some advantages to doing some of these calculations by hand. You will get a better understanding of the nuts and bolts of the process doing it from pencil, calculator, and paper rather than simply pushing a button and automagically generating a design. The automagic function many times can produce something you do not want and having the experience and knowledge of the low level details helps spot mistakes.

27th July 2011, 12:37 PM	#8
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	If you want to understand bass loading of speaker drivers then you must at some stage do it all by hand method and by hand calculation. Once you have achieved a sufficient understanding then you can adopt the "quick" computer methods. I suggest "Bullock on Boxes" (isbn 0-9624191-5-x) as a good reference document to gain that initial understanding. It even shows how to automate some of the measuring/testing/calculation using your PC. His book is based on articles published in "Speaker Builder" during the 1980s. It predates most of the computerised design and testing that was/is available.