I need to start collecting some calibration equipment to test some new speakers I am building. Currently I am looking at the Earthworks Audio Measurement Series M30 microphone. I also need a 48v phantom power, does it matter what brand I use?
I need some test software so I am open to suggestions on what works well and reasonably priced. I also figure I need to use my laptop or one of my desktops sound cards. Are there any minimum requirements to work with this microphone / recommended software?
Really new to this side of speaker building and would like to buy the right gear first time. I would really appreciate any assistance you can offer.
I need some test software so I am open to suggestions on what works well and reasonably priced. I also figure I need to use my laptop or one of my desktops sound cards. Are there any minimum requirements to work with this microphone / recommended software?
Really new to this side of speaker building and would like to buy the right gear first time. I would really appreciate any assistance you can offer.
You could try the free ARTA software suite.
I picked up a new Superlux ECM999 measurement mic on ebay, for a very reasonable price, after reading some reviews and looking at the frequency response and other specs.
Just to be able to play around with the measurement mic and ARTA, I bought a "TubeMP Studio" mic preamp on ebay, which also provides the 48V phantom power. I don't know how good it is but I got it new for about $40, last year. They seem to have several different TubeMP models on ebay, now.
I also got the necessary cables from another ebay seller. The TubeMP Studio has both three-pin XLR female and 1/4-inch mono female input jacks, and both 3-pin XLR male and 1/4-inch mono female output jacks. I used the 3-pin XLR input for the measurement mic's cable, with 48-Volt phantom power provided by the TubeMP, and used the 1/4-inch mono output so I could easily use an adapter on the other end of the cable for my soundcard input.
If you do some searches, you'll find that there are LOTS of good threads here at diyaudio about equipment and software for speaker measurements.
Regards,
Tom
I picked up a new Superlux ECM999 measurement mic on ebay, for a very reasonable price, after reading some reviews and looking at the frequency response and other specs.
Just to be able to play around with the measurement mic and ARTA, I bought a "TubeMP Studio" mic preamp on ebay, which also provides the 48V phantom power. I don't know how good it is but I got it new for about $40, last year. They seem to have several different TubeMP models on ebay, now.
I also got the necessary cables from another ebay seller. The TubeMP Studio has both three-pin XLR female and 1/4-inch mono female input jacks, and both 3-pin XLR male and 1/4-inch mono female output jacks. I used the 3-pin XLR input for the measurement mic's cable, with 48-Volt phantom power provided by the TubeMP, and used the 1/4-inch mono output so I could easily use an adapter on the other end of the cable for my soundcard input.
If you do some searches, you'll find that there are LOTS of good threads here at diyaudio about equipment and software for speaker measurements.
Regards,
Tom
Last edited:
Thanks for the advice Tom. I want a measurement microphone that comes with a calibration chart to dial out any imbalance in the microphone, so I need to look at something a little more upmarket.
I did look at the EarthWorks Audio ZDT 1021 but it’s like $1300, between it and the M30 it is over $2000 with GST and freight, and while looking at the TubeMP you mentioned, I noticed the affordable phantom power supply only seem to go to 20kHz-25kHz, so perhaps I should be looking at the marginally cheaper M23 microphone ($459 + shipping) if I am not going to spend the dollars on the megabuck preamp?
I downloaded the user manual for the ARTA software (which by the way is 79 euro if you want to use all the features) and it was filled with math equations. Is there any software where I don’t need a degree in math to use?
I did look at the EarthWorks Audio ZDT 1021 but it’s like $1300, between it and the M30 it is over $2000 with GST and freight, and while looking at the TubeMP you mentioned, I noticed the affordable phantom power supply only seem to go to 20kHz-25kHz, so perhaps I should be looking at the marginally cheaper M23 microphone ($459 + shipping) if I am not going to spend the dollars on the megabuck preamp?
I downloaded the user manual for the ARTA software (which by the way is 79 euro if you want to use all the features) and it was filled with math equations. Is there any software where I don’t need a degree in math to use?
Not sure what you mean by dialing out the imbalance in the microphone but if you were planning to EQ it out then doing so will most likely make measurements worse.
Also unless you have access to anechoic chamber there isn't much point dropping 2 grand on an earthworks as the measurements would be fairly meaningless without a treated room.
Preamp and ADC is just as important as the room and microphone. You may want to look at a solid state preamp over a tube preamp for measurement purposes. fmr has a great pre that has superb transparency that isn't too expensive.
The truth is theres no easy way to make meaningful measurements at home, and spending money on equipment without knowing the limitations is just wasting money. What exactly are you trying to calibrate?
Thanks pby I downloaded the pdf for the FMR Audio RNP8380 and will take a more detailed look at it tonight. I like his honest approach and at ~$500 is a little more palatable than $1300.
No I don’t have a anechoic chamber, but I am looking to acquire test equipment that is not going to be dissatisfactory 5 minutes after I buy it. Experience has taught me not to skimp and have to spend twice.
So I am looking for something that is marginally better than middle of the road first time up. Not looking for over the top but don’t want garbage either.
If I spend $1500 all up on test gear and I spend $10k on speaker parts, to me that seems like a reasonable balance to achieve the end goal of quality sound.
But I don’t want to spend mega bucks on the speaker parts only to be let down by my test gear when it really matters.
Sorry if I haven’t used the right terminology as I am fairly new to all this but I wanted a quality microphone that provides a calibration chart so any slight imperfections that are inherent in the microphone can be dialled out by including the discrepancy in the software to cancel the imperfections out, giving a neutral microphone (as far as the software is concerned). Earthwork provide a graph of the imperfections for $50 extra, so the detail can be provided to the software to counteract the microphone imperfection effect.
No I don’t have a anechoic chamber, but I am looking to acquire test equipment that is not going to be dissatisfactory 5 minutes after I buy it. Experience has taught me not to skimp and have to spend twice.
So I am looking for something that is marginally better than middle of the road first time up. Not looking for over the top but don’t want garbage either.
If I spend $1500 all up on test gear and I spend $10k on speaker parts, to me that seems like a reasonable balance to achieve the end goal of quality sound.
But I don’t want to spend mega bucks on the speaker parts only to be let down by my test gear when it really matters.
Sorry if I haven’t used the right terminology as I am fairly new to all this but I wanted a quality microphone that provides a calibration chart so any slight imperfections that are inherent in the microphone can be dialled out by including the discrepancy in the software to cancel the imperfections out, giving a neutral microphone (as far as the software is concerned). Earthwork provide a graph of the imperfections for $50 extra, so the detail can be provided to the software to counteract the microphone imperfection effect.
I'm not entirely convinced this is the right approach. There are room acoustics/preamp/adc which isn't be accounted for. You cannot simply dial them out.
Unless you're prepared to go hire a chamber, budget equipment here will probably suffice for your needs. Behringer has a really inexpensive measurement microphone called the ecm8000 and you can just pair it up with a behringer mixer or preamp. While high end gear will give you more precise results, you would still be limited by the room acoustics. It would be a waste of money to spend it on mics and pres for you, as they can't correct acoustics.
I guess what I am looking for the most precise measurement I can afford, it would be great to have an anechoic chamber, but all factors considered at the end of the day I will using the speakers in a real room where there are objects that will affect the sound output.
I am currently looking at an up market digital crossovers that can do room equalisation, so I figure as long as I can measure the room and speaker accurately, I am in a good place to dial out the environment to a certain degree.
The digital crossover have such infinite adjustability I can adjust the digital crossovers to suit the room conditions should I need to move them to another room.
If we were talking strictly passive crossovers I might think quite differently about things.
I am currently looking at an up market digital crossovers that can do room equalisation, so I figure as long as I can measure the room and speaker accurately, I am in a good place to dial out the environment to a certain degree.
The digital crossover have such infinite adjustability I can adjust the digital crossovers to suit the room conditions should I need to move them to another room.
If we were talking strictly passive crossovers I might think quite differently about things.
Generally the better the quality the flatter they are, but just like a speaker none are perfectly flat so they provide a map of the imperfections so you can add it to your software, and the software artificially corrects the imperfections in the microphone by applying adjustment to the reading form the microphone.
The way I figure it the more perfect you can make the reading, the better informed you are to make a call on thing like applying adjustments to any dips in the response curve. If the reading says you high and you’re really low because of inaccurate equipment, you could potential tune an even greater problem into the speaker response.
Its the theory I’m working on I could be wrong but it makes sense to me...
The way I figure it the more perfect you can make the reading, the better informed you are to make a call on thing like applying adjustments to any dips in the response curve. If the reading says you high and you’re really low because of inaccurate equipment, you could potential tune an even greater problem into the speaker response.
Its the theory I’m working on I could be wrong but it makes sense to me...
Of course that is a correct method, in general. I am definitely not an expert but I think that some uncontrollable and some uncontrolled variations will usually provide measurement uncertainties that are at least an order of magnitude or two greater than any measurement mic's response corrections, making them insignificant, unless you use an anechoic chamber (or are investigating a particular point in a particular room, or your mic's response is much less flat than mine is advertised to be).
You may be right but being a complete newbie at all this I would rather aim higher than lower than I need, mostly because people with experience may be able to outperform my measurements with cheaper equipment simply by knowing what they are doing.
But I figure the more accurate I can start off with, the better the chance I have of getting it right. Better doesn’t guarantee right, but lesser almost guarantee wrong in the wrong hands. You could almost look it as I’m buying myself some inexperience leeway.
These are the microphones I am currently looking at http://www.earthworksaudio.com/microphones/m-series/
Along with this pre amp http://fmraudio.com/rnp.htm
Of course it’s all pretty meaningless unless I invest the time to read the manuals really well and get to understand the equipment properly.
Agree about the variables and there might be more than the mic makes up for as you say, but until I get my hands dirty I can only work with the understanding I have.
I’m curious about the need for the chamber. To me it would seem of much greater use where the variables are set like in a speaker with a passive crossover, where you only get one crack at setting it right and then forever and a day after that it’s locked in.
Whereas I will be using a crossover that is infinitely adjustable after the speakers are built in any environment I chose to move them to…
Really if I can tune the speaker in an environment where it is going to be used, and get everything levelled off with the obstacles of the environment in place, would I not be massively ahead of a perfectly setup up speaker that is moved to another environment?
Just curious because to me the chamber only seems to be of benefit to speakers with passive crossovers that are fixed.
Again only a newb laying the facts out as I percieve them at this time.
But I figure the more accurate I can start off with, the better the chance I have of getting it right. Better doesn’t guarantee right, but lesser almost guarantee wrong in the wrong hands. You could almost look it as I’m buying myself some inexperience leeway.
These are the microphones I am currently looking at http://www.earthworksaudio.com/microphones/m-series/
Along with this pre amp http://fmraudio.com/rnp.htm
Of course it’s all pretty meaningless unless I invest the time to read the manuals really well and get to understand the equipment properly.
Agree about the variables and there might be more than the mic makes up for as you say, but until I get my hands dirty I can only work with the understanding I have.
I’m curious about the need for the chamber. To me it would seem of much greater use where the variables are set like in a speaker with a passive crossover, where you only get one crack at setting it right and then forever and a day after that it’s locked in.
Whereas I will be using a crossover that is infinitely adjustable after the speakers are built in any environment I chose to move them to…
Really if I can tune the speaker in an environment where it is going to be used, and get everything levelled off with the obstacles of the environment in place, would I not be massively ahead of a perfectly setup up speaker that is moved to another environment?
Just curious because to me the chamber only seems to be of benefit to speakers with passive crossovers that are fixed.
Again only a newb laying the facts out as I percieve them at this time.
As I said, I am no expert. But I see two different things being sort-of lumped together: You are talking about designing speakers and you are talking about adjusting them to any room. The way you talk about "using an infinitely-adjustable crossover in any environment" seems to imply that nothing else will matter. So why even worry about measuring during the design?
I do realize that you didn't mean it that way. But to me, just saying it tends to imply that you might be a lot more surprised than you currently think, by the complexities of the dynamics of sound waves in a room, and of attempts to achieve meaninglful measurements in that type of environment. Even with just a single tone, everything is different at every mic position and at every instant of time. Echoes and other effects fill the room with complex interference patterns (sums of multiple interesecting waves, with variable amounts of reinforcement and cancellation, everywhere), causing the amplitude to vary significantly for even small changes in mic position. Even if you had a theoretical perfect point source, perfect equalization would only be completely valid at one point in the space.
I would think that for measurements during speaker design, as many room variables should be eliminated as possible, at least once you know what types of propagation patterns are desirable, without room effects. But finding out what propagation patterns can be made to perform best WITH room effects might require in-room measurements, if that is also unknown, initially. (Or maybe it's already common knowledge, for all I know.)
Not being an expert, what I've attempted to convey, above, could all be seriously flawed. PChi's suggestion of doing some reading of relevant books is a very good one. Software and measurement equipment will be mostly useless unless you understand what can be measured and exactly how to set up the measurement software and hardware.
I do realize that you didn't mean it that way. But to me, just saying it tends to imply that you might be a lot more surprised than you currently think, by the complexities of the dynamics of sound waves in a room, and of attempts to achieve meaninglful measurements in that type of environment. Even with just a single tone, everything is different at every mic position and at every instant of time. Echoes and other effects fill the room with complex interference patterns (sums of multiple interesecting waves, with variable amounts of reinforcement and cancellation, everywhere), causing the amplitude to vary significantly for even small changes in mic position. Even if you had a theoretical perfect point source, perfect equalization would only be completely valid at one point in the space.
I would think that for measurements during speaker design, as many room variables should be eliminated as possible, at least once you know what types of propagation patterns are desirable, without room effects. But finding out what propagation patterns can be made to perform best WITH room effects might require in-room measurements, if that is also unknown, initially. (Or maybe it's already common knowledge, for all I know.)
Not being an expert, what I've attempted to convey, above, could all be seriously flawed. PChi's suggestion of doing some reading of relevant books is a very good one. Software and measurement equipment will be mostly useless unless you understand what can be measured and exactly how to set up the measurement software and hardware.
Agree I didn’t mean to make the proper speaker design aspect of it seem incidental, as it is far from it. Guess what I was trying to say in a somewhat flippant way was that even after I do my darnedest to follow every golden rule of good speaker building, the DCN28 is going to give me a level of flexibility to fix conditional issues that a passive crossover can never do.
That doesn’t mean that I won’t be aiming for perfect cabinet, with perfect acoustics, and trying to sort the room out in a way that allows for the best possible sound, I was just saying that once all of that is done, some of the pressure of doing everything spot is taken away if can be manipulated after the speakers are finished to the best of my ability.
Guess what I need to be careful of is not using a crutch to fix a problem that could be fixed with some more thought. So yes I guess you are right a chamber would be very helpful to get the speaker and the sound right as possible, and then rely on the DCN28 to compensate for the room effect rather than relying on it to fix problems that could have been prevented with more effort.
Unfortunate I don’t have a room for such testing, but perhaps I could make some kind of make shift room come that time.
I’m not a big one for reading but in order to get the most out of the equipment I am about to purchase, I really need to step up my education level and do some dedicated reading like the one PChi mentioned.
That doesn’t mean that I won’t be aiming for perfect cabinet, with perfect acoustics, and trying to sort the room out in a way that allows for the best possible sound, I was just saying that once all of that is done, some of the pressure of doing everything spot is taken away if can be manipulated after the speakers are finished to the best of my ability.
Guess what I need to be careful of is not using a crutch to fix a problem that could be fixed with some more thought. So yes I guess you are right a chamber would be very helpful to get the speaker and the sound right as possible, and then rely on the DCN28 to compensate for the room effect rather than relying on it to fix problems that could have been prevented with more effort.
Unfortunate I don’t have a room for such testing, but perhaps I could make some kind of make shift room come that time.
I’m not a big one for reading but in order to get the most out of the equipment I am about to purchase, I really need to step up my education level and do some dedicated reading like the one PChi mentioned.
Last edited:
Hello,
The book 'Testing Loudspeakers' includes a short section on Microphones for Loudspeaker Testing and also describes the use of MLSSA and CLIO to create a quasi anechoic response where the effect of room reflections is minimised by analysing the response before the reflections return.
In my opinion it covers a complex subject in a readable way.
I have only tested speakers used in Mobile Phone Handsfree Units and also handsets and appalling is overstating the sound quality.
In my opinion because the loudspeakers are a box of delayed resonances the crossover and electronics can't realistically compensate. I would say that the drive units have the greatest effect on sound quality. Unfortunately I don't know whoose are good. I have some Focal units that seem OK and SEAS are respected. I have not been imppressed by KEF units.
I have been in an RF anechoic room when performing EMC testing that used carbon loaded foam wedges. It was also pretty good acoustically and was weird to be in, really quiet.
The book 'Testing Loudspeakers' includes a short section on Microphones for Loudspeaker Testing and also describes the use of MLSSA and CLIO to create a quasi anechoic response where the effect of room reflections is minimised by analysing the response before the reflections return.
In my opinion it covers a complex subject in a readable way.
I have only tested speakers used in Mobile Phone Handsfree Units and also handsets and appalling is overstating the sound quality.
In my opinion because the loudspeakers are a box of delayed resonances the crossover and electronics can't realistically compensate. I would say that the drive units have the greatest effect on sound quality. Unfortunately I don't know whoose are good. I have some Focal units that seem OK and SEAS are respected. I have not been imppressed by KEF units.
I have been in an RF anechoic room when performing EMC testing that used carbon loaded foam wedges. It was also pretty good acoustically and was weird to be in, really quiet.
Interesting comment... On axis I don’t see why it couldn’t be delayed electronically with a great deal of accuracy. If the sound is buffered and then sent to the appropriate crossover channel marginally in front or behind the other signals, it should produce fairly accurate time alignment to compensate for any cabinet time delay due to all drivers being on the same baffle board.
Off axis where reflection off the wall comes into play might be a whole different ball game... What might be EQ’d perfectly on axis 1 meter in front of the cabinet, may be completely wrong in the main seating position after it mixes with the reflected sound waves.
I think this goes back to my previous post about me not being lazy in the design phase and making sure that all the drivers are time aligned as well as possible, so as not to require electronic adjustment. Again nice to have it but hopefully shouldn’t need it.
Definitely need to do some reading to understand the principals better before I start trying to measure the speakers (that have yet to be built).
I’m kind of excited about learning how to measure speakers and the room environment properly, as it takes me to a whole new place in setting up my gear. Instead of just plonking down anywhere a set of speakers that someone else made and hoping for the best, I can position them and adjust the room with acoustic deadening materials to get the most out of them.
Last time I set up the Cremona’s I did it by ear. I wound up moving them further and further away from the walls to stop reflection, ending up a couple of feet off each wall. It will be nice to be able to measure that sweet spot rather than having to guess with my ears.
Currently the Cremona’s are in storage about a 1000klms away waiting to be relocated, temporarily I am listening to some Sharp 2 way speakers that belonged to a 3 in 1 that is about 20 years old. My ears haven’t started bleeding yet but there not far off it.
Maybe they would be a good starting point to practice on, so I’m not concerned about trying to make them better and can just concentrate on the measuring.
Attachments
Hello,
I like the pictures of the Cremona's.
I have drifted some way from your initial question. I had a look at the Earthworks Audio Measurement Series M30 Microphone and it appears to be an electret type. Even the cheap capsules as used in Mobile Phones were pretty good with a reasonably flat frequency response and relatively well defined sensitivity. Hopefully the Earthworks isn't just a cheap capsule in an expensive housing.
The Bruel and Kjaer measurement microphones were of the condenser type with I believe a 200 V bias. They were used with a sound pressure calibrator to set the level. It's some while since I did any testing and it was all semi automated. I remember the microhone calibration chart and I think that adjustment coefficients were added in. Somewhat excessive for Mobile Phones.
I agree that the crossover and electronics can compensate for time alignment which is good and necessary. What the electronics can't realistically do is compensate for imperfections such as cone break up and cabinet resonances. The resonances are likely to be dependent on time, signal level, temperature and possibly humidity if they are paper cones. In my opinion these have a large effect on sound quality, dominant for the tweeter.
Also non linear distortion is very hard for the electronics to compensate.
My only experience is measuring handsfree loudspeakers and the harmonic distortion could reach many 10s of percent.
I like the pictures of the Cremona's.
I have drifted some way from your initial question. I had a look at the Earthworks Audio Measurement Series M30 Microphone and it appears to be an electret type. Even the cheap capsules as used in Mobile Phones were pretty good with a reasonably flat frequency response and relatively well defined sensitivity. Hopefully the Earthworks isn't just a cheap capsule in an expensive housing.
The Bruel and Kjaer measurement microphones were of the condenser type with I believe a 200 V bias. They were used with a sound pressure calibrator to set the level. It's some while since I did any testing and it was all semi automated. I remember the microhone calibration chart and I think that adjustment coefficients were added in. Somewhat excessive for Mobile Phones.
I agree that the crossover and electronics can compensate for time alignment which is good and necessary. What the electronics can't realistically do is compensate for imperfections such as cone break up and cabinet resonances. The resonances are likely to be dependent on time, signal level, temperature and possibly humidity if they are paper cones. In my opinion these have a large effect on sound quality, dominant for the tweeter.
Also non linear distortion is very hard for the electronics to compensate.
My only experience is measuring handsfree loudspeakers and the harmonic distortion could reach many 10s of percent.
Looked for an e-book of this but can't find one, I might need to buy the paper copy.
Agree there is no substitute for quality parts and sound engineering. If I go with Scan-Speak drivers they will be paper and if I go with AudioTechnology they will be polyprop.
The AudioTechnology website had an interesting write up on the use of paper drivers, but I digress...
Is one microphone type more suited to measurement over the other?
Unfortunately I don't know for sure if one microphone type is more suited to measurement.
Certainly the Bruel and Kjaer condenser / capacitor microphones give repeatable results. Mind you at the price they should be good. I vaguely remember that the sound pressure level calibrator alone was a few hundred pounds. The software engineers where I worked managed to loose one so a replacement was required. Another manufacturer of condenser microphones is ACO Pacific.
The electrets use the same condenser microphone principle with a bias voltage trapped in a plastic diaphram (using magic) and performance is pretty good. I think that the Bruel and Kjaer condenser microphones use a metal diaphram so are likely to be more stable and repeatable.
My feeling is that unless you have unlimited money or are developing speakers for a living the electrets should be good enough.
Thanks for the suggestion of the AudioTechnology website. The Focal units that I have use a Kevlar sandwich construction for the mid and bass. The tweeter however is titaniam.
Certainly the Bruel and Kjaer condenser / capacitor microphones give repeatable results. Mind you at the price they should be good. I vaguely remember that the sound pressure level calibrator alone was a few hundred pounds. The software engineers where I worked managed to loose one so a replacement was required. Another manufacturer of condenser microphones is ACO Pacific.
The electrets use the same condenser microphone principle with a bias voltage trapped in a plastic diaphram (using magic) and performance is pretty good. I think that the Bruel and Kjaer condenser microphones use a metal diaphram so are likely to be more stable and repeatable.
My feeling is that unless you have unlimited money or are developing speakers for a living the electrets should be good enough.
Thanks for the suggestion of the AudioTechnology website. The Focal units that I have use a Kevlar sandwich construction for the mid and bass. The tweeter however is titaniam.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.