You are here

DIY Speaker Frequency Testing Methods

Customising, building or repairing your own gear? Need help with acoustic treatment or soundproofing? Ask away…

DIY Speaker Frequency Testing Methods

Postby pkalder » Thu Jan 31, 2019 7:25 pm

Hello all! Trying to decide on the best way to test some speakers frequency response diy with minimal equipment.

I can't decide between half space on the ground outside vs elevated on a ladder. On the ground half space would be the most simple, but do I gave to dig a hole? I don't want to. Does it have to be flush?

I want to test multiple speakers, and they have different cabinet depths. I have a relatively cheap condenser mic which is calibrated.

Half space or elevated? If half, no objects in radius of the mic right. Any other suggestions? Will the csd be accurate? I just began learning all this. Measurement diy is very sparse online.

Image
pkalder
New here
Posts: 9
Joined: Wed Jan 23, 2019 5:00 pm

Re: DIY Speaker Frequency Testing Methods

Postby James Perrett » Thu Jan 31, 2019 9:36 pm

I remember certain hifi mags used to test speakers on a high pole in the absence of an anechoic chamber.
User avatar
James Perrett
Moderator
Posts: 8258
Joined: Mon Sep 10, 2001 12:00 am
Location: The wilds of Hampshire
JRP Music - Audio Mastering and Restoration. JRP Music Facebook Page

Re: DIY Speaker Frequency Testing Methods

Postby pkalder » Fri Feb 01, 2019 6:00 am

used to test speakers on a high pole in the absence of an anechoic chamber.

Thanks james. This is what I can't decide. I guess I could try both, but I wouldnt know which is the most accurate. Also, I'd have to make sure the test ended before I get reflections from the ground however many feet below into the mic. Most people online I have seen do the elevated test.

I'm wondering why more dont do the half space with their speaker facing the sky. No ladder or mic suspension. It would make it much more convenient. I am also wondering why more people dont create CSDs for themselves. Most companies dont give out these graphs, and it makes sense why.
pkalder
New here
Posts: 9
Joined: Wed Jan 23, 2019 5:00 pm

Re: DIY Speaker Frequency Testing Methods

Postby Wonks » Fri Feb 01, 2019 9:25 am

One problem is that 'outside' is definitely not as quiet as inside an anechoic chamber, so you'll certainly be getting some noticeable influence from that at the upper and lower ends of the frequency range where the outputs are lower.

As measurement mics are (almost always) omnis, you will be picking up sound from all around. And unless it's a totally still day, you'll get wind noise. You can fit a 'dead cat' windscreen to mitigate that, but it will still have an effect on the frequency response, albeit slight. But you still don't want that at all when doing measurements.

What sort and size of speakers are you talking about measuring?
User avatar
Wonks
Jedi Poster
Posts: 8892
Joined: Thu May 29, 2003 12:00 am
Location: Reading, UK
Correcting mistakes on the internet since 1853

Re: DIY Speaker Frequency Testing Methods

Postby James Perrett » Fri Feb 01, 2019 1:23 pm

Do a search for James Moir - he almost certainly wrote some papers in the mid/late 70's on his open air test methods.
User avatar
James Perrett
Moderator
Posts: 8258
Joined: Mon Sep 10, 2001 12:00 am
Location: The wilds of Hampshire
JRP Music - Audio Mastering and Restoration. JRP Music Facebook Page

Re: DIY Speaker Frequency Testing Methods

Postby Hugh Robjohns » Fri Feb 01, 2019 1:28 pm

pkalder wrote:Hello all! Trying to decide on the best way to test some speakers frequency response diy with minimal equipment.

What exactly are you trying to achieve? Measuring speakers on poles, or in holes in the ground, is a rather labour-intensive and a tricky methodology because of external noise, wind, etc, and all the practicalities and safety issues of rigging speakers at heights.

I'd have thought you'd be much better off using something like FuzzMeasure software (now RodeTest, and Mac only, I'm afraid), which our very own Phil Ward uses for his own speaker assessments/reviews. It principally uses impulse response testing, and there's a time-windowing function to gate out any local strong reflections to improve accuracy (although that can also limit the LF measurement extension if you have nearby boundaries).

Here's a review of the old FM4:
https://www.soundonsound.com/reviews/supermegaultragroovy-fuzzmeasure-4

And here's a link to the Riodetest webpage:
https://www.rodetest.com/
H
User avatar
Hugh Robjohns
Moderator
Posts: 24035
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: DIY Speaker Frequency Testing Methods

Postby Mixedup » Fri Feb 01, 2019 2:09 pm

Hugh Robjohns wrote:What exactly are you trying to achieve?

Yeah, I struggled to understand that. A commercial speaker manufacturer might want to measure the speaker only... but they'd surely want to measure it properly, not in a sort of nearly-good-enough DIY test. But surely everyone elese would want to test the whole listening system... including the room, and from the listening position(s).


Hugh Robjohns wrote:I'd have thought you'd be much better off using something like FuzzMeasure software (now RodeTest, and Mac only...

There's also Room EQ Wizard (REW), which is free for Mac, Windows and Linux.
User avatar
Mixedup
Jedi Poster
Posts: 4336
Joined: Wed Sep 03, 2003 12:00 am
Location: Cambridgeshire, UK

Re: DIY Speaker Frequency Testing Methods

Postby Studio Support Gnome » Fri Feb 01, 2019 9:01 pm

Mixedup wrote:
Hugh Robjohns wrote:What exactly are you trying to achieve?

Yeah, I struggled to understand that. A commercial speaker manufacturer might want to measure the speaker only... but they'd surely want to measure it properly, not in a sort of nearly-good-enough DIY test. But surely everyone elese would want to test the whole listening system... including the room, and from the listening position(s).


Hugh Robjohns wrote:I'd have thought you'd be much better off using something like FuzzMeasure software (now RodeTest, and Mac only...

There's also Room EQ Wizard (REW), which is free for Mac, Windows and Linux.



I love your Innocence



Let's just say that when I was given tour of a certain speaker manufacturers production facility, while being interviewed for a job (for which they decided I was over qualified..... ) I raised eyebrows and asked some pointed questions about their test and documentation methods....

Having put my foot in my mouth, they decided I was over qualified for the service engineer's role, but were talking to the MD about creating a new role ..... one which made better use of my multi-discipline engineering skills.....

The MD however, pointed out that while he had signed off on £Xk for a service engineering role, he was not about to sign off on a management restructuring, new departmental roles and 3 times the money ...... just because I'd been politically indelicate during my interview .

fair Point I thought.....


I now work just around the corner..... applying said engineering skills for someone else....
User avatar
Studio Support Gnome
Frequent Poster (Level2)
Posts: 2824
Joined: Tue Jul 22, 2003 12:00 am
Location: UK
Now available for consultations and audio engineering jobs .  Also guitar tech work , and “rent-a-shredder” sessions .  Oxfordshire based but can and will travel .  Email maxtech.audio@me.com

Re: DIY Speaker Frequency Testing Methods

Postby Folderol » Fri Feb 01, 2019 9:25 pm

Studio Support Gnome wrote:I now work just around the corner..... applying said engineering skills for someone else....
:clap: :bouncy:
However, I wish you wouldn't keep using the 'W' word - makes me feel quite tired :tongue:
User avatar
Folderol
Jedi Poster
Posts: 8558
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: DIY Speaker Frequency Testing Methods

Postby pkalder » Sat Feb 02, 2019 12:44 am

what are you trying to acheive
I'm building diy auratones, and i want to measure as accurate a CSD/waterfall as possible. Learn some stuff two. As of a month ago I had no idea what frequency response was.

every one else would want to test the whole system
Then once I have the baseline response of the speaker, i can see where i need to treat the room. I will be using Room eq wiz as well.

what sort of and size speakers?
The diy auratones i am building and some behringer ms16 "monitors." I want to see how bad the behringer ms16 are. I got suckered into these a year ago. I need actual monitors, and I want to dive into analysis and measurements. I may get some useful monitors in the future, and at my price points they dont have CSD plots.
pkalder
New here
Posts: 9
Joined: Wed Jan 23, 2019 5:00 pm

Re: DIY Speaker Frequency Testing Methods

Postby OK1 » Sat Feb 02, 2019 10:57 am

pkalder wrote:
what are you trying to acheive
I'm building diy auratones, and i want to measure as accurate a CSD/waterfall as possible. Learn some stuff two. As of a month ago I had no idea what frequency response was.

every one else would want to test the whole system
Then once I have the baseline response of the speaker, i can see where i need to treat the room. I will be using Room eq wiz as well.

what sort of and size speakers?
The diy auratones i am building and some behringer ms16 "monitors." I want to see how bad the behringer ms16 are. I got suckered into these a year ago. I need actual monitors, and I want to dive into analysis and measurements. I may get some useful monitors in the future, and at my price points they dont have CSD plots.

The universal challenge is cost vs benefit, getting as much bang for your buck, as you have alluded to.

I think, it's best to look at this as an incremental challenge. A related example, looking good has fundamentals - eat a healthy diet, exercise, sleep, and if you can afford it some massage, smart clothes, and maybe a bit of makeup or plastic surgery or specialised care by a skin expert/and or hair expert, and of course better cameras and better lighting improve your look in pictures, and on top of this you may add some photoshop to even out blemishes.

My point is - how far do you want to take this, cos it can become a never ending journey, so some of the questions that others have asked really revolve around your destination, how far do you wish to take this and do you know when you have arrived? Otherwise it becomes an endless chase.

Some of the most important decisions and choices that affect your final quality.

1. Strategy - Build or buy - I think SOS mag, has the best collection of reviews of affordable studio monitors (and of course the more expensive ones), with very few caveats, the reviewers opinions are some of the most trusted - so suggest you start there. DIY/Building is great, nice hobby, but arriving at a decent result takes quite a bit of knowledge. And the build option is not the best use of limited funds, trial and error can quickly become more expensive than the buy option, with no guarantee of a result. Buy - if you choose wisely, is a bit of insurance, someone who does this every day and has to compete with other manufacturers every day, has put in many hours of design and manufacture. The competitive environment means speakers generally get better over time, with improvements in material science, computer aided design, 3D printing at prototype stage, DSP. With DIY, you may not be taking advantage of the latest knowledge, such as better drivers and components, cos there is less opportunity for comparison in the DIY world - with a far smaller number of speaker users in this space.

Rather than add your own learning curve to the most difficult aspect of your entire solution - the speaker itself, it may be better to focus energies and risk on only those areas that no one else can solve for you, outsourcing as much of the solution to experts (paid or pro bono), so buy wisely, with a lot of study of paid for/published information(e.g magazines), and freely available opinions on the web/forums, is highly recommended.

2. Positioning - and playing around with this to achieve the most effective sound, toe in angles, distance, speaker height, distance from walls, etc, etc.... Probably more than many other aspects - this is important.

3. Budget and where do you wish to focus your spend. Some would say it's the room/acoustic treatment, others would argue it's the speaker. I say both of these things need wisdom/knowledge to get the best value for your investment. My suggestion is to deal with this incrementally - speakers 1st, positioning next, then if you have money left over, treat the room with commercial solutions.

4. Quick optimisations - But before any expensive acoustic treatment, consider using things you may already have - bookshelves, cushions, drapes, carpets to tame most of the troublesome reflections, In a bedroom, with wardrobes full of clothes - open the wardrobe for even more absorption. Positioning yourself in the nearfield, eliminates quite a bit of the reflected sound. Moving the speaker nearer the front wall, if you have a front ported speaker, can reduce the time between the direct and reflected sound, for a more coherent result - experiment to discover what works best.

5. On the acoustic treatment - address points of 1st reflection as a priority. Do not aim for perfection, all audio reproduction is a compromise anyway - it is never perfect, aim for the most affordable, best value compromises, and stop at some point.

6. Assuming that you are listening in the nearfield, with most of what you hear coming from the direct source - the speakers, I think there is value in applying some gentle - note I said gentle - eq to "improve" your speaker. Lots of ways to do this but none of them really free - the free solution will take your time, trial, error, measurement microphone, an SPL meter (or a USB calibrated microphone), and a year or two or more to really understand how best to get a good result - just when you think you got your eq right, you realise there is a minor fault in your correction, and a slight change in the eq will improve it even further. Or the paid options like Sonarworks on the cheaper end of things. Look at this as makeup, icing on on the cake. Over time in addition to the basic eq curve correction, you'll learn what kind of personal preference curve which you may wish to implement over and above the correction curve( e.g more bass, roll of bass at certain frequencies. etc.

7. One simple improvement, rarely mentioned - place speakers at the front of the desk and/or on stands, positioned to avoid 1st reflections from the speaker to your desk to your ears. This simple change makes any speaker sound a lot better.

8. If you are trying to save money, as your most important constraint, I have found that speakers with larger woofers give a better impression of the low end, than these tiny woofer speakers with beefed up amplifiers and boosted bass EQ. But with larger woofers, unless you are willing to spend a bit of money on something like the Neumann KH80 DSP a small but very accurate speaker, your listening distance will increase a bit to enable the sound from both tweeter and woofer to "blend" at your ears, otherwise you may be listening to more of the tweeter and misjudging the audio.

9. Measure speaker frequency response from the nearfield, so most of what you measure is the direct sound of the speaker, rather then room reflections - and as some others have mentioned there are windowing techniques for reducing the contribution of reflections in the measurements. I do not think it makes sense measuring the room, and attempting to correct for this. Better to focus on being nearfield and your ear does a decent job of zoning out the room sound, psychoacoustically. Acoustic treatment is the better way to deal with the influence of the room - not EQ. Google Ethan Winer, he has strong opinions on this and I tend to agree with him. Probably the only psychoacoustic tweak I could think of to compensate for the room, is a slight reduction in lower frequencies, cos lower frequencies are more omnidirectional and tend to be boosted in all rooms(from stronger room reflections, than higher frequencies which have fewer reflections) - this tweak tends to help the midrange focus (subjectively) that bit more - be gentle - less EQ is more.

Only you can decide which is more important to you - your time, or your money and how much risk you can place on yourself and your personal knowledge limitations, and which is more expedient - spending money to save time(by buying the end product of accumulated knowledge), or spending your time to try and save money.

Here's my final suggestion - the Kali Audio LP6

https://www.soundonsound.com/reviews/kali-audio-lp-6

A good example of well built, recently designed, and highly rated set of speakers for £315 or about $300, per pair.

This is most likely a better way to spend both money and time, enjoying a good product. Out of the box it should sound great, yet it has lots of options, in built, to tweak the sound to your personal taste and room, or in response to your measurements.....
OK1
Regular
Posts: 80
Joined: Sun Sep 27, 2009 12:00 am
Location: Reading, Berkshire, UK

Re: DIY Speaker Frequency Testing Methods

Postby Wonks » Sat Feb 02, 2019 11:13 am

Quite a few wrong assumptions there OK1, along with some valid points. I'm not going to pick everything out, but just want to say that regardless of how close you sit to monitors speakers (and you can certainly sit too close for good listening), the room acoustics will still have a big effect on what you are hearing. It's physics, and you cannae change the rules.

If you can't treat your room to a reasonable extent, then use headphones.
User avatar
Wonks
Jedi Poster
Posts: 8892
Joined: Thu May 29, 2003 12:00 am
Location: Reading, UK
Correcting mistakes on the internet since 1853

Re: DIY Speaker Frequency Testing Methods

Postby Mike Stranks » Sat Feb 02, 2019 12:07 pm

Just a word or two on 'outside' testing...

I do a fair bit of location recording in the wide open spaces... mainly SFX material. I'm using decent mics and a have a good recording chain. Even in the depths of the Gloucestershire countryside (and I mean depths!) it's amazing how much faint residual atmospheric noise I'm picking-up. Birds, very distant traffic/aircraft, tree/wind noise, far-off dogs, sheep, cattle etc etc. Over 50% of my recordings get dumped for that reason.

And I'm not using omnis the vast majority of the time!

Our ears/brains have become attuned to filtering-out all that mush. Unfortunately decent and semi-decent recording chains haven't.
Mike Stranks
Jedi Poster
Posts: 6716
Joined: Fri Jan 03, 2003 1:00 am

Re: DIY Speaker Frequency Testing Methods

Postby pkalder » Tue Feb 05, 2019 5:06 am

Thanks everyone. Thanks OK1 on the tips. I do plan on getting better monitors in the future. I spent a couple hundred getting the acoustic piano tuned, and now all I use is the MIDI keys. I wish I put that money towards monitors.

I'll still keep playing round with this calibrated mic I have, and everyone's comments about outside noise make sense. Taking that into account, I'm thinking of going with indoor gated measurement combined with near field measurements. I've got some unknown 4.5 speakers too that I can't tell if they are full range. And a couple of 10s that I'm wondering where they drop out.
pkalder
New here
Posts: 9
Joined: Wed Jan 23, 2019 5:00 pm

Re: DIY Speaker Frequency Testing Methods

Postby pkalder » Tue Feb 05, 2019 5:56 am

I dont mean to bump the thread again, but also...

Why are there no calibrated piezio mics? It seems like this could remove much of the rooms effect on the sound output.

What would happen if I stuck a piezio mic on a bookshelf speaker cab and ran it through room eq wizard? I'm guessing piezios have their own coloration.
pkalder
New here
Posts: 9
Joined: Wed Jan 23, 2019 5:00 pm

Re: DIY Speaker Frequency Testing Methods

Postby ef37a » Tue Feb 05, 2019 9:01 am

"I'm guessing piezios have their own coloration."

Ask any guitarist trying to record an acoustic with one!

Then, I seem to remember that the radiation from cabinet walls is highly non-linear? That is why such great steps are taken to minimize it.

I wish you well. Lot of work, probably a lot of fun but ultimately pointless.

Dave.
ef37a
Jedi Poster
Posts: 10506
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: DIY Speaker Frequency Testing Methods

Postby Martin Walker » Tue Feb 05, 2019 4:38 pm

pkalder wrote:Why are there no calibrated piezio mics? It seems like this could remove much of the rooms effect on the sound output.

Probably for the same reason that all but expensive reference microphones come with a 'generic' rather than an individually-measured frequency response. For instance, if you want a shock, take a look at the frequency response results from 85 samples of a 'well-known budget' omnidirectional measurement mic:

http://www.cross-spectrum.com/weblog/2009/07/

Warning - make sure you're sitting down before you view it :beamup:

These results don't mean the product can't be used as a measurement mic in many scenarios, but that you certainly shouldn't rely on its absolute measurements below 100Hz or above 10kHz in particular :headbang:


Martin
User avatar
Martin Walker
Moderator
Posts: 14011
Joined: Wed Jan 13, 2010 9:44 am
Location: Cornwall, UK

Re: DIY Speaker Frequency Testing Methods

Postby Mixedup » Tue Feb 05, 2019 8:50 pm

Wow. Good find Martin.

Those last two individual plots... :o
User avatar
Mixedup
Jedi Poster
Posts: 4336
Joined: Wed Sep 03, 2003 12:00 am
Location: Cambridgeshire, UK

Re: DIY Speaker Frequency Testing Methods

Postby wireman » Tue Feb 05, 2019 9:49 pm

Does anyone know what the cheapest calibrated mic is? I was thinking of getting the UMIK-1 next time I'm in the USA. Don't really have a serious application but I'm interested in the frequency response of a few locations.
wireman
Regular
Posts: 267
Joined: Fri Dec 17, 2004 1:00 am

Re: DIY Speaker Frequency Testing Methods

Postby Hugh Robjohns » Wed Feb 06, 2019 12:00 am

Martin Walker wrote:These results don't mean the product can't be used as a measurement mic in many scenarios, but that you certainly shouldn't rely on its absolute measurements below 100Hz or above 10kHz in particular

Yes, there's a good reason why professional calibration mics are very expensive, and why the B8000 (and other similar budget test mics) isn't a professional calibration mic.

But having said that, for purely comparative measurements it's adequate, and if you can acquire a proper calibration chart/file it's inherent production tolerance variations can be compensated given the appropriate measurement software.

H
User avatar
Hugh Robjohns
Moderator
Posts: 24035
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Next

Who is online

Users browsing this forum: No registered users