The speaker we're talking about in this thread is only a small one, meant for portability, so I was never expecting much bass, but it did turn out to go pretty low in the end, and being two small drivers it has pretty fast bass which I like. I did mess with EQ a bit on my phone before I gave him the speaker and I feel like raising the low bass 3db did improve the overall sound. But I'm not sure how close my phone is to outputting a flat response even though it says it's flat on the levels. So in the convo with him I mentioned he may have his phone set to something that is changing the sound a lot, asked him to mess around with it a bit to see if he can get the sound he wants, haven't spoken to him since though so I don't know how that went.2 cents worth >
A truly 'flat' speaker that uses high quality low distortion drivers can simply be made to sound 'exciting' or 'impressive' just using EQ.
However, small speakers can never have quite the bass impact of large ones without adding a Subwoofer.
AND sooo much comes down to ROOMS & REFLECTIONS + MTM speakers are very critical of 'speaker height'.
( I have reservations regarding the mentioned amp being used )
I got the info about taking measurements from toid directly on his forum, he helped me a lot when I was starting out, basically taught me most of what I know, but I never got to go into more detailed methods as he's a busy guy. I'll attempt to do as you mentioned above next time I do some measurements, but figuring out the software is the trickiest part for me, but thanks for taking the time. I don't have my computer with me for a few weeks so I can't upload anything really, but I realise it would have been easier for everyone to see it, apologies for that.I'm not sure what the current state of Dayton microphone calibration files is, but last time I did some testing the manufacturer provided calibration were worthless - no basis in reality. These mics are typically hot by a few db or more in the treble, which means your treble could be down quite a bit. I recommend replacing the mic with another Dayton mic calibrated by CSL. I have personally verified CSL calibration agrees with the calibration from Earthworks and NTI mics as well as a DIY microphone I built 20 years ago that was calibrated by Kim Girardin.
I can't find TOID's measurements tutorial, but I did find a video of him recommending the Dayton mics for measurement - so you know what I think of that. Again, nothing wrong with Dayton mics but they need to be calibrated by a 3rd party.
Measuring inside is problematic because you need to splice the near field bass measurement, and the measurements are smoothed at lower frequencies. So you really have no idea what's going on with the bass, and resonances can be masked at lower frequencies. Jack Hidley of NHT said he considers this type of measurement worthless.
Its best to go outside, put your speaker as high in the air as possible and take some measurements that way. Do one measurement that is an average of a few on axis measurements (eg -20, -10, 0, 10 20 horizontal and +-10 vertical. Do a second measurement that is a 360 degree loop with measurements every 10-20 degrees to get a rough idea of the power response. The on axis should be flat and the power should be flat descending.
Can you share the drive units, crossover frequency/topology and any measurements you might have? This might be helpful.
Also, again, nothing wrong with a metal-mdf sandwich. If the adhesive you used has damping properties you can claim it's constrained layer construction! But I don't think the cabinet is your problem.
I'm really not sure, all I know is he said he has a very small living room, he lives to far away from me to do any kind of measuring, it's why it's been so hard to diagnose the issue, if I could sit in the room with him I might see straight away what the issue is, maybe he's put it right upto the wall, or on a shelf in a cabinet or even on the floor, or his phone setting is messing with it, or he's playing bad quality recordings, I just don't know but I will find out all of that the next time I speak to him hopefully.I didn't read whole thread carefully, but it seems no one suggested the most important thing - what is the in-room frequency response, measured at the listening position? If the room-induced null of the frequency response falls in the 60 - 100 Hz range, of course there will be no punch. Try several different loudspeaker position and/or listening position and measure each frequency response (no gating, 1/12 and 1/6 dB/octave smoothing).
Yeah I asked in an earlier post (I think) if I should have added some kind of curve to the response rather than making it flat, but that just answered my question, thank you.@Dameo182 Please look up loudness curve and also human hearing range and perception compared to volume. You will notice that a flat response speaker sounds as if it has little bass or treble to humans at lower volumes. This effect tapers off as volume increases. Around these parts we compensate this human condition with the classic smiley face eq curve. Because this effect tapers off you would not want to build this into the crossover unless you know at what level the speakers will be driven all the time.(as well as room response and other things) I guess I'm saying that flat sounds dull but it is the best place to start when you have to do the inevitable room and listening level corrections.
Oh and also I am not an expert so now you have 4cents (if the 2cents worth are in series)
Jeremy
@classicalfan
Thank you, really, I looked for years for the amp you just linked, it's over three times as powerful as the ones I found and from the description better quality, but without typing in the model name of the amp into the UK amazon site it wasn't showing at all. Now I've managed to find it I've just bought one to try it out. It seems a much better amp, I don't know if they're new or just weren't showing up for me before but thank you. I'm actually excited yo see what projects I can build using that amp if it delivers what it claims.
Previously I tried using the stereo amps and losing one side of the audio, I hated it, so I used a circuit (can't remember it's name now) of resistors to blend both sides into a mono signal, it worked, but was a lot of extra 'to do' and I just wanted to find an easier way with a ready made mono amp.
Thank you, really, I looked for years for the amp you just linked, it's over three times as powerful as the ones I found and from the description better quality, but without typing in the model name of the amp into the UK amazon site it wasn't showing at all. Now I've managed to find it I've just bought one to try it out. It seems a much better amp, I don't know if they're new or just weren't showing up for me before but thank you. I'm actually excited yo see what projects I can build using that amp if it delivers what it claims.
Previously I tried using the stereo amps and losing one side of the audio, I hated it, so I used a circuit (can't remember it's name now) of resistors to blend both sides into a mono signal, it worked, but was a lot of extra 'to do' and I just wanted to find an easier way with a ready made mono amp.
Last edited:
A bit off topic, but after reading some replies relating to the calibration of the mic I'm using, what are the opinions on the new omnimic due to be released? Same issue or has it been more refined for in room measurements? I know it has new features like indicating 1watt but is it worth replacing my current one with that?
I wouldn't buy the omnimic unless you can find someone else who bought one and verified the calibration against something known. Also, as AllenB said, you don't want a USB mic because of the latency issue.
I would get a CSL calibrated EMM6, or a used NTI M2010 from Toronto surplus, or Earthworks M23 or Isemcon mic.
I would get a CSL calibrated EMM6, or a used NTI M2010 from Toronto surplus, or Earthworks M23 or Isemcon mic.
Some are concerned about setup but it's not difficult to use a regular mic with the right guidance like you'll find here.
Some mics want 'Phantom power', and while many of us DIY, you can buy soundcards that supply it simply. The soundcard itself can be USB if you don't have a desktop computer, that's no problem because the reference channels are together on the card.
Stereo channels out, one goes to the speaker/amplifier and one goes to one of the stereo input channels. The mic goes to the other input channel.
This way the computer compares the test signal directly with the measured signal.
Some mics want 'Phantom power', and while many of us DIY, you can buy soundcards that supply it simply. The soundcard itself can be USB if you don't have a desktop computer, that's no problem because the reference channels are together on the card.
Stereo channels out, one goes to the speaker/amplifier and one goes to one of the stereo input channels. The mic goes to the other input channel.
This way the computer compares the test signal directly with the measured signal.
@pulexirritans @AllenB
It seems like a complicated process to someone like me who has very limited experience with measuring equipment, but I'm willing to learn. I'm not sure how available those mic's are here in the UK but I'll have a look for one when I get the chance, I'm assuming they have to be used with REW? I have zero experience with that software and have heard it can be a steep learning curve as it's not very intuitive?
It seems like a complicated process to someone like me who has very limited experience with measuring equipment, but I'm willing to learn. I'm not sure how available those mic's are here in the UK but I'll have a look for one when I get the chance, I'm assuming they have to be used with REW? I have zero experience with that software and have heard it can be a steep learning curve as it's not very intuitive?
Sorry if I didn't catch the question earlier but I can't stand arguing about syntax, slang, common incorrect uses, colloquialisms and context.Yeah I asked in an earlier post (I think) if I should have added some kind of curve to the response rather than making it flat, but that just answered my question, thank you.
WARNING This post has grammatical errors yet is contextually accurate.Also, as @AllenB said, you don't want a USB mic because of the latency issue.
Isn't there a technique to use the second channel output as a timing reference?
I think I thought I saw a vid.
This is an honest inquiry.
Jeremy
@pulexirritans @AllenB
It seems like a complicated process to someone like me who has very limited experience with measuring equipment, but I'm willing to learn. I'm not sure how available those mic's are here in the UK but I'll have a look for one when I get the chance, I'm assuming they have to be used with REW? I have zero experience with that software and have heard it can be a steep learning curve as it's not very intuitive?
Use Arta. It used to be $100 but now it looks like it's free? Arta is easy to use and get measurements out of. Later you can take the measurements from Arta and import them into something like vituixcad to design crossovers or do other stuff.
Isemcon is in the EU, but my preference would be to order the used NTI M2010 because of the titanium ACO pacific capsule. The others are plastic, which is fine, just the NTI is on another level and doesn't cost any more.
You'll also need a USB interface that works with arta such as Scarlett 2i2 or Behringer UM202HD
A Swept frequency response curve analysis . . . it is possible to over-sweep the measurement instrument.
If the Hz/second sweep is faster than the rise and fall time of the Resolution Filter of a spectrum analyzer/tracking generator; Or FFT Bin filter rise and fall time, versus the sweep generator Hz/second sweep rate, . . . then the amplitude of the signal will be attenuated, destroying the measurement.
This is often misunderstood by those who make the measurement.
A USB mic does have a delay, so it is an additional problem for the measurement.
Just slow the sweep down, and make it closer to a point by point frequency response measurement, and you can correct for the problem.
Fast, accurate, high resolution measurements are often an Oxymoron, the swept frequency response is one of those measurements.
Good measurements often take more time than you think, and more time than you want.
I solved many customers problems by getting them to slow down.
Just my $0.02
If the Hz/second sweep is faster than the rise and fall time of the Resolution Filter of a spectrum analyzer/tracking generator; Or FFT Bin filter rise and fall time, versus the sweep generator Hz/second sweep rate, . . . then the amplitude of the signal will be attenuated, destroying the measurement.
This is often misunderstood by those who make the measurement.
A USB mic does have a delay, so it is an additional problem for the measurement.
Just slow the sweep down, and make it closer to a point by point frequency response measurement, and you can correct for the problem.
Fast, accurate, high resolution measurements are often an Oxymoron, the swept frequency response is one of those measurements.
Good measurements often take more time than you think, and more time than you want.
I solved many customers problems by getting them to slow down.
Just my $0.02
I used Arta for some time and it was a pleasure to work with. Holmimpulse and many others have good things going for them. Some find REW less intuitive to use but it is supported, is cross platform and there are tutorials available to help.Use Arta.
Thanks for everyone's input, I've ordered 2 of the new amplifiers linked above and I will see how his response is to that one as a replacement and update this thread, could be a while though. As for the measurements, I really appreciate the advice given about the mic, I feel I'd need to start a whole new thread for that as it's a lot more complicated than I'd hoped, I was under the impression that the omnimic was doing what I was expecting it to do, granted I'm only taking measurements for my own projects and not selling them to hardcore audiophiles, but I'd still like to get better results if possible. I think the measurements I've taken get me in the ball park of good sound (it seems like that to my ears anyway) but I guess I'll have to improve my knowledge if I'm to improve my results in the future.
Thanks again for all the time taken helping me with this 👍
Thanks again for all the time taken helping me with this 👍
- Home
- Loudspeakers
- Multi-Way
- Quick question about a speaker I recently made