I have been designing and building speakers for a few years now and recently came up with a new design I believe has not been done or at least hasn't had any success. The build was done with simulation software to start me off and then final voicing was done by ear and took about a year for it to sound how I would like. I believe they sound fantastic and played them for a number people who all have been pleased with them. 2 people asked to buy them, 1 asked for the drivers and build guide so I feel that others feel the same that I do. What is a good software to measure the response of the system and quantitfy the results and see what if anything needs adjustment? I have listened to a bunch of different media however it has all been from the same source and amp. Later in the week I will try other amps and sources and see how it changes the tonality. Any help would be appreciated.
If you want to get started on the cheap, to see if this is for you, there are a few free/demo programs. I use them to measure "relative" speaker response (dips, peaks).
You have the PC, probably with built in sound card, you can start with a mic (unidirectional) and install either of the following. They are fairly straight forward to use. If you find its for you, then a better method is buying a calibrated measurement mic (absolute SPL). There is the entire other issues of nearfield, farfield and room issues to deal with during measurement.
ARTA Software
https://www.roomeqwizard.com/
I have seen people use REW over on the HT forums. I will down load it tonight. I have a good external ADDA conterter. What mics would be ideal? Would I be better off going thru my adda converter and mic pres or should I just get a USB mic? I only use it for my recording DAW and not sure if it would work with REW.
baudline is a very good software package that runs on linux systems. Its live spectrogram is very detailed and fast. It is free and is a very powerful and sophisticated tool.
If you are using this to provide customer data, then often a proper calibrated mic is better like the one below is popular. If this is just for your own curiosity, then any cheaper unidirectional mic will do as you'll be doing relative measures.
https://www.parts-express.com/minid...sb-measurement-calibrated-microphone--230-332
Yes I believe a few of these designs would be able to be sold and would like solid data so I can make final adjustments as well as provide potential buyers with a cut sheet on how they perform.
Unidirectional (Cardioid) mics are not suitable for measurements. You need an omnidirectional (pressure) microphone.
I mean, omnidirectional, as RonE accurately describes. The Yamaha EQ mic is omnidirectional as well. Got my "mords wixed" 🙄
The open field (on a quiet, non windy day) removes environment reflections and gain from the measurement. You would also need to elevate the speaker on stand.
This is ideal (@1m) if you want to isolate the speaker performance or compare it to manufacturers data. They will also smooth to 1/3 octave to approx the ears response, but you may want 1/12 octave or more if you're debugging a speaker. You can usually find or create a spot indoors that may be close enough.
More ideal, is after you think the speaker is working properly is to measure it at your installed listening position as there will be lots of other room interactions that you may want to adjust for.
Last edited:
Not necessarily.
You can get ok measurements from 500Hz up with a gated measurement indoors if you have a wide open room. Below that you need to do nearfield, outdoor or anechoic chamber measurements.
1m is not always the best distance to measure from. It is a reference point for sensitivity more than anything else.
The speaker is approx 48"in tall however the drivers start at approx 2.5ft from the base. Is the elevation needed for decoupling purposes or mic placement?
What parameters should I be most concerned with?
That's a tall speaker, assuming it's a floor model 😉 As long as the ports or drivers are not close to the bottom then no stand should be required. Its for decoupling from the floor and nearby walls.
The typical spec sheet FR measurement is @1m, center of the driver(s) on axis using 1/3octave smoothing. Often an additional off axis [30deg] measurement is included as these curves still look good. The problem will be preventing room reflections in your measurements.
The typical spec sheet FR measurement is @1m, center of the driver(s) on axis using 1/3octave smoothing. Often an additional off axis [30deg] measurement is included as these curves still look good. The problem will be preventing room reflections in your measurements.
Last edited:
Yea, the towers/floor standing versions are pretty tall. I loaded the hell out of the cabs for high sensitivity, wider range, etc and dampen for better response - both add size. The sim showed a bump of about 4-5db that was able to get smoothed out by cab design. I wanted to make the front baffle as slim as possible for a few reasons which resulted in the current shape. It also put the center of drivers at ear level regardless if they are in the living room and I am on the couch or in my bed room laying down. I made 4 in ceiling ones and they have a larger front baffle so is a more compact design but same internal volume. Be interesting to see how the 2 compare and react with a different size baffle.
I'd be interested in seeing pic and results when you have them. Still, given all the variables involved, lots of people manage to get close to the sim predictions.
- Status
- Not open for further replies.