How do you check for this? What test signal do you use and at what spl ?
I've been involved in loadsa Listening Tests, many of them DBLTs, on the audibility of speaker distortions so this is a real question.
Agreed. It’s getting harder to find reasonably good speakers at a reasonable price, but it is possible.
I owned a pair of Usher S520 and they sounded so much better than a lot of towers out there. Very balanced sound, detailed. And at $400 a pair, quite the steal.
Also had some Mission speakers in the 80’s and 90’s. Very good performers, even the smaller 700 series.
True that these days, finding a pair of is harder. Navigating through the internet mess where hashtags have replaced actual data is quite daunting.
The concept is good, but the mass consumer final product is not necessary sounding as your tastes, i haven't listened Devialet so i will not say anything but i think that they know what they are doing with their market... i'm targeting a 60-70dB at the listening point and the masterpiece is the Peerless PLS-65F25 in a very stuffed and closed volume crossed at 100Hz (LR24). I'm using a Thomann DSP and TA2024 AMP for around 400€ for the complete system.
No, that's the idea, you cannot. A lot of speaker defects are not measurable, at least not at home with mainstream software. You have to listen, and listen for many, many hours, unless it's so bad that it's immediately obvious.
Obviously there are measurable defects, but if the measurement shows no defects it doesn't mean it's a good speaker.
Though I have been involved for most of my previous life in dreaming up new ways of measuring speakers, my greatest interest is LISTENING TESTS.
You said
so I want to know what you played when you heard these audible distortions ... and the conditions ...
This can be music. Certain types of music are good for picking up certain types of speaker distortion in DBLTs
Listening tests are tricky for me, i've decided to limit the low end rendering of my system in order to control my feelings because i'm unable to tune the DSP after an extended listening period... it's always seems not engouh.
When I come in the morning of after a long period, the low end is very deep and defined, it become thin and unconsistent each time i push it immediately over 80dB or more than one hour under 80dB.
I never exceed the limit and enjoy the few minutes of unbiased perception in 24 hours when i tune my DSP.
When I come in the morning of after a long period, the low end is very deep and defined, it become thin and unconsistent each time i push it immediately over 80dB or more than one hour under 80dB.
I never exceed the limit and enjoy the few minutes of unbiased perception in 24 hours when i tune my DSP.
For example right now my rig has two cases when it doesn't sound right:
1. Baritone saxophone and trumpets, I believe either the primary note 440 Hz or harmonics 880 Hz. One well recorded instrument solo may sound ok but when it's multiple instruments it sounds harsh
2. Just adding more instruments of any kind. When it's 4-5 separated it sounds nice, when it's like 10 instruments it sounds like something is really wrong and some of them are trying to mess up against the main music
And because I can't really tell bad recording from speaker error I need to listen like 10 hours of this stuff to conclude the problem is repeating.
I do have some measurable problems in my 1st diy speaker and I'm rebuilding it from scratch, no conclusion until I fix what I can.
And the opposite problem is that some tracks are so well recorded and somehow so carefully avoiding anything that can stress the speakers that I think they are specifically made to sell audio equipment at the dealerships. For example My Funny Valentine Emiel Scholsberg sounds like I have $10k speakers without even a minor problem, it sounds fabulous. I would be a millionaire having working at the audio shop if all customers will find it sufficient to buy after this single track 🙂.
Please read this link Trumpet Peculiar Frequency spectrum why trumpet could be one of the most peculiar ways to test the system, it has a lot about psychoacoustic and harmonics. However one single well recorded trumpet solo won't do it, to break the system it needs to play as part of something larger. A sax must be similar, but opens up different problems. I believe some of the orchestra playing loud at low frequencies to increase the cone excursion does the trick. And I believe the cone excursion is one of the major factors breaking the sound quality, we may discuss it later.
Can you tell us the Track or CD please.
It may be that the only desktop speaker that will satisfy you is this. You can remove the legs so it sits on your desk.
Attachments
coming up with these findings. did you finalize the XO on the W6? And built the second box?
if not, then all of this is moot as you don't really have a working system.
@perceval that's right, that's why I don't worry too much. My second box is so labor intensive that I am still working on it, but it's so much fun that I don't mind. I am using only plywood, no MDF, and there is no 90 degrees junction between any two walls. The only instruments I have are a handsaw and a drill, so I have a lot of quality outdoor time on my backyard. Every wall except base takes 3 layers of plywood or wood, so 16 panels total. Air tighteness at odd angles takes effort.
I pulled out some foam core HyperCube speakers cabinets from storage and decided to use these as the basis for a new desktop nearfield speaker. The cabinet has heavily sloped sides so the diffraction from the baffle is minimal and imaging is very good. Interior will be sealed and stuffed with polyfill. I used 3M 77 contact cement to apply a new square baffle to the old cutout hole so that I can pick a new driver to play with. I also decided to spray paint the foam core for the first time to deal with the white foam DIY look. I used flat black spray paint and then an over coat of automotive (Toyota) Phantom grey pearl BTY1614. It gave it a nice modern, if a bit Deathstar/Battlestar Galactica space look. This suits me fine. Waiting for paint to dry then I will install probably a Dayton PC83-8 3in glass fiber full range that I have had lying around for years. Never heard them play.
Finished. Will listen and measure next. Then develop passive filter for BSC etc.
Immediate comparison between your new speaker and known accurate headphones/in ear monitors can make judging issues like this easier. Using something that's closed back or mechanically sound isolating also speeds up the switchover, since you don't even have to turn the speaker off in most cases. Levels need to be matched, obviously.
I designed a filter for it after measurements. Listening to it now. Sounds great - very nice image and detail. This driver has a lot of “air”.
Filter developed in Xsim (blue) and raw measurement (green):
Response and harmonic distortion for 2.0Vrms at 0.5m:
P2P XO with Wago connectors:
Have a listen (with Caldera 12 subwoofer for bass):
Filter developed in Xsim (blue) and raw measurement (green):
Response and harmonic distortion for 2.0Vrms at 0.5m:
P2P XO with Wago connectors:
Have a listen (with Caldera 12 subwoofer for bass):
Last edited:
I apparently can't handle boxes, or even any realistic amount of baffle, in the nearfield anymore. I had been using a set of Lii Audio Fast-8s with their magnets strapped to Kanto stands. Then I found ESS selling Heil AMTs for basically half price, which sent me down the rabbit hole of learning how to do crossovers for real (which I had basically been allergic to my entire time in DIY). So now I have them crossed to the Fast-8s around 525Hz, second order. The Heils are as advertised, clear as light. really wish there were such a thing as a giant Heil driver that could get down to 80Hz or so ... keep dreaming, right?
Anyway, after my usual over preparation, my initial crossover makes these two drivers sing in the nearfield, at about 90dB average sensitivity. I use a mildly slot loaded line array of close out 4"x6" drivers (originally used by some no name maker as the "subwoofer" in some large-ish sound bar) gotten from Parts Express years ago, driven by a cheap little Dayton DTA-100LF subwoofer amp for my bottom end. The Fast-8s in no baffle are good to about 130Hz before they start rolling off. The manufacturer's graph is accurate, showing them to be good around 60Hz, but that doesn't account for all the extra roll off for not having a large test baffle that the graphs were taken on. The line array of small drivers makes an ideal match as it gets sound generation up to ear level, basically no matter what the listener's level is, while the Heil/Fast-8 combo has really good, nearly point source dispersion. Oh, I'm in a corner with my broad desk across the corner so the sub is standing in the corner behind the desk. Its only a foot wide, so its got lots of room.
With the Heils on sale, I'm out less than $500 in drivers. Crossover parts, which I went middle of the road on since it was also basically my first real soldering project, were about $120. The sub amps can be had between $100-$120. Other materials and supplies maybe add up to $50. So, $700-$800 total. Add another $250 if the Heils are regular price. Sounds like a lot, well, to me it does. I've never had this much money in a system before. Guess there's one advantage to getting older ;-p
Anyway, the Fast-8s alone, no crossover, with a sub, were really fantastic for a long time. The main drawback being that if they weren't aimed directly at you're ears, you shifted into a more laid back tonality due intensely focused directivity as you go up in frequency (large full rangers, right?). So, I had them strapped to Kanto stands sitting on these tiny lazy suzans, so I could aim them at will with a touch. For active listening, aim them dead on to how you're sitting. for anything else, it hardly matters as they sound practically the same no matter where you are until you get nearly flat to their frames, which I couldn't because of the corner sitaution. So, a good place to start, a nice full ranger of some sort, mounted however you can manage, with a DIY line array sub, or just lifting a sub that plays well up high to ear level to tone down the source location between 80Hz-130Hz (or wherever your chosen full ranger can reach).
After that, its a tweakers dream system. I'm working now on some different ways to get my stereo separation down lower, hopefully to the holy grail mark of 80Hz or lower, so then I can play around with different subwoofer ideas without collapsing my sound stage. FUN, FUN, FUN ...
Anyway, after my usual over preparation, my initial crossover makes these two drivers sing in the nearfield, at about 90dB average sensitivity. I use a mildly slot loaded line array of close out 4"x6" drivers (originally used by some no name maker as the "subwoofer" in some large-ish sound bar) gotten from Parts Express years ago, driven by a cheap little Dayton DTA-100LF subwoofer amp for my bottom end. The Fast-8s in no baffle are good to about 130Hz before they start rolling off. The manufacturer's graph is accurate, showing them to be good around 60Hz, but that doesn't account for all the extra roll off for not having a large test baffle that the graphs were taken on. The line array of small drivers makes an ideal match as it gets sound generation up to ear level, basically no matter what the listener's level is, while the Heil/Fast-8 combo has really good, nearly point source dispersion. Oh, I'm in a corner with my broad desk across the corner so the sub is standing in the corner behind the desk. Its only a foot wide, so its got lots of room.
With the Heils on sale, I'm out less than $500 in drivers. Crossover parts, which I went middle of the road on since it was also basically my first real soldering project, were about $120. The sub amps can be had between $100-$120. Other materials and supplies maybe add up to $50. So, $700-$800 total. Add another $250 if the Heils are regular price. Sounds like a lot, well, to me it does. I've never had this much money in a system before. Guess there's one advantage to getting older ;-p
Anyway, the Fast-8s alone, no crossover, with a sub, were really fantastic for a long time. The main drawback being that if they weren't aimed directly at you're ears, you shifted into a more laid back tonality due intensely focused directivity as you go up in frequency (large full rangers, right?). So, I had them strapped to Kanto stands sitting on these tiny lazy suzans, so I could aim them at will with a touch. For active listening, aim them dead on to how you're sitting. for anything else, it hardly matters as they sound practically the same no matter where you are until you get nearly flat to their frames, which I couldn't because of the corner sitaution. So, a good place to start, a nice full ranger of some sort, mounted however you can manage, with a DIY line array sub, or just lifting a sub that plays well up high to ear level to tone down the source location between 80Hz-130Hz (or wherever your chosen full ranger can reach).
After that, its a tweakers dream system. I'm working now on some different ways to get my stereo separation down lower, hopefully to the holy grail mark of 80Hz or lower, so then I can play around with different subwoofer ideas without collapsing my sound stage. FUN, FUN, FUN ...