hi thank you for the very kind reply
what is your take on IMD? is it an important parameter for assess a driver quality?
Sorry its me again but i really need some comments on this
As i said above i have the strong feeling that IMD measurements can tell a lot about a driver's behaviour (like also FR and CSD of course)
Looking for IMD figures/graphs on drivers datasheets is very frustrating The very rare times i find something i find not complete information
I am attaching the IMD results for a very expensive and very well regarded driver, used in very expensive hifi speakers
I have two questions
1) why the graph is truncated in the lower range ? (the 30Hz test tone is missing from the graph)
2) testing at only 80dB/1m makes much sense ? (considering that some music peaks can reach easily 100dB the IMD at let's say 95dB could skyrocket )
Please give me your opinion/advice
Thank you very much indeed
imho IMD together with FR and CSD can be very useful to select a driver
As i said above i have the strong feeling that IMD measurements can tell a lot about a driver's behaviour (like also FR and CSD of course)
Looking for IMD figures/graphs on drivers datasheets is very frustrating The very rare times i find something i find not complete information
I am attaching the IMD results for a very expensive and very well regarded driver, used in very expensive hifi speakers
I have two questions
1) why the graph is truncated in the lower range ? (the 30Hz test tone is missing from the graph)
2) testing at only 80dB/1m makes much sense ? (considering that some music peaks can reach easily 100dB the IMD at let's say 95dB could skyrocket )
Please give me your opinion/advice
Thank you very much indeed
imho IMD together with FR and CSD can be very useful to select a driver
Attachments
TO BE SUNG TO THE TUNE OF, "ONWARD CHRISTIAN SOLDIERS"
Onward Loudspeaker Designers
Marching as to war!
With our Holy Priesthood
And lots of Attitude!
(Burn the Heretic)
etc.
First of all, if the OP doesn't like the Trayf (not Kosher) Agility Factor, he can still create a SPREADSHEET of driver candidates and populate it with whatever data sets he prefers. But... reading below, it seems that the usual parameters are Greek to him. In retrospect, I should not have bothered trying to help him.
Secondly, my co-author Jim Tuomy has six patents in Electroacoustics--please list all YOUR patents for me.
Thirdly, the Reader of our paper, John Atkinson was asked to deliver the Richard Heyser Memorial Lecture by the AES, in recognition of John's having published a greater number of comprehensive loudspeaker measurements than anybody else. John Atkinson did not think that our paper was noise.
Finally, I have never claimed to be an engineer... but I have done a lot of work as a Team Leader, leading a team designing consumer products... and the products need to sell in a competitive market, and our assumption always is that our products are going to be bought by competitors and torn down and costed out, and if there is any trash they can talk in memos to their sales force, they will talk trash. So, having "name brand" subcomponents is important. For any given purpose, more expensive is not always better...
But nobody who knows the business can say that Peak Consult, Sonus Faber, or Rockport were fools for paying the price for drivers from Audio Technology.
Have a nice day,
john
Manufacturers don't post IMD because they don't have to. They sell drivers without it. Because the box design and crossover can affect IMD, measuring that on a standard baffle is not much use. You'll note the HD chart post 14 was for a packaged speaker, not just one of the two drivers. That published number didn't sell many speakers, 99.99% of the SP2(2004) model sold to bar bands that like their guitars rather distorted. They buy pedals to do just that. Hardly anybody bought SP2(2004) because it reproduced Steinway grand piano accurately @ 1/8 watt. Many that like the latter test bought Klipsch Heresy at about 3 times the price of the SP2(2004).
The graph you posted was IMD 30 hz + 255 hz, @ 80 db, so obviously 30 hz line would be the same as the 255 line shown, 80 db.
You will see from the HD chart post 14, HD was 10 db higher at 10% rated power (50 W) than at 1% rated power (5w). So if a manufacturer is puffing IMD, he wants to show it as low as possible. Besides, home users listen at 80 db quite a lot, me somewhat lower, 75 average. Bar bands OTOH play at half or more rated power 90% of the time.
Fr I suppose you mean frequency of resonance. datasheets I download more consistently post Fs al though if there is a frequency response chart you can read the Fr from the frequency of the impedance peak. CSD I don't know what you are talking about. The deltapro-15A datasheet , the fatal 15PR400 datasheet, the McCauley 8341-8, the B&C 15TBX100 datasheets don't have that. Sorry I have studied 15" woofers, I didn't intend to build a bookshelf speaker because I have never heard a good one in the bass. I've heard very little of anything that I thought was excellent. Even the automobile sized Meyersound speakers at Louisville Brown theater I thought were screechy on the violins.
To select a bass driver I looked at useable frequency range, sensitivity, and the F3 . I also wanted a flattish frequency chart because bumpy charts cause HD & IMD, and it takes 2 passive crossover components to even out each bump. Crossover components cost >$20 each IMHO, and I want to get away with only 4 or 6 of them per speaker.
I wanted high sensitivity because I run 35 to 70 W/ch class AB amps that run 30 years without maintenance. If you are running a Class D amp with an average useful life of 5 years, which are unrepairable by amateurs, you might not care how many watts you are using at 80 db. I also wanted to hear at least down to 54 hz -3db, since I find that realistic on piano organ and bassoon low notes. Bass reflex design can extend driver response that falls off after 100 hz down to 54 hz with a box designed for that frequency. 30 hz -3db in a 2 or 3 way seems attainable only in a $1000 JBL model not sold or stocked within 1500 miles of here.
Apparently one can achieve 54-15khz in a 2 or 3 way in a bookshelf box; see many designs on multiway forum. Wolfsteeth just posted one with a 5" woofer. I suggest you look around at those, and copy. i don't know how low the wattage would be at 75 db that I listen to all day, probably higher than a 15" 2 way because that giant cone make 54 hz very efficiently. A 6" cone will have a lot more excursion IMHO to make 54 hz @ 75 db. I have a feeling that lots of excursion causes HD & IMD, see the result for HD post #14.
JohnMarionMarks calculated number looks interesting, but a first approximation of that figure of merit would be sensitivity and useable frequency range, which nearly everybody publishes.
Last edited:
If I’m not banned for the following, I will most certainly be burned at the stake!
I’m a firm believer in defining a very clear objective for anything I do. What is the end goal? What is the purpose? When this is defined, we can then plan a route from where we are now to when we want to be In the future.
For me, the goal is simply listening to music with as much enjoyment as possible. I appreciate that not everyone hears the same and what constitutes “enjoyment“ will vary from person to person. However, since my speakers are for me, I don’t need to concern myself with this minor technicality! To avoid chasing rainbows, I’ve narrowed down my goal to “possessing nothing which detracts from my enjoyment of the music”
The major detractor for me is sloppy pitch reproduction. Some components appear to struggle with reproducing something as simple as the accurate pitch of a note. Unfortunately, I’m not aware of any metric which correlates with pitch reproduction, so the only solution is to listen to the drivers a see how they perform. The drivers don’t need to be mounted in an enclosure for this initial test: the differences are obvious. Selecting drivers this way separates the wheat from the chaff, leaving what I think of as a driver with “intrinsic musicality“.
Note that at this point nothing has been measured! But we’ve got a driver, or a selection of drivers which have been chosen to assist us in our journey towards our dream speaker.
I’m a firm believer in defining a very clear objective for anything I do. What is the end goal? What is the purpose? When this is defined, we can then plan a route from where we are now to when we want to be In the future.
For me, the goal is simply listening to music with as much enjoyment as possible. I appreciate that not everyone hears the same and what constitutes “enjoyment“ will vary from person to person. However, since my speakers are for me, I don’t need to concern myself with this minor technicality! To avoid chasing rainbows, I’ve narrowed down my goal to “possessing nothing which detracts from my enjoyment of the music”
The major detractor for me is sloppy pitch reproduction. Some components appear to struggle with reproducing something as simple as the accurate pitch of a note. Unfortunately, I’m not aware of any metric which correlates with pitch reproduction, so the only solution is to listen to the drivers a see how they perform. The drivers don’t need to be mounted in an enclosure for this initial test: the differences are obvious. Selecting drivers this way separates the wheat from the chaff, leaving what I think of as a driver with “intrinsic musicality“.
Note that at this point nothing has been measured! But we’ve got a driver, or a selection of drivers which have been chosen to assist us in our journey towards our dream speaker.
Here is a 2 way with 7" woofer that somebody recommended on a multiway thread: http://speakerdesignworks.com/Invictus_1.html
The multiway forum thread is https://www.diyaudio.com/community/threads/new-home-new-system.394285/latest
The multiway forum thread is https://www.diyaudio.com/community/threads/new-home-new-system.394285/latest
Hi thanks for the very valuable advice In my simple mind i was thinking that the drivers performance in some ways fix a limit for what can be achieved
I mean if i have a driver with a high distortion no cabinet on the earth will lower that distortion because it is intrinseic in the driver
i agree that if a driver is used badly or mounted in a bad cabinet the resulting distortion will be worse
If i read a datasheet i would like to know everything of that driver And distortion vs Hz for me is maybe the most important thing
With Frequency Response and Cumulative Spectral Decay
👍
i would like to see the IMD products in the lower region where things can go bad very easily
I am sure the portion not shown it is much worse They do not want to show the worst side of the story
And i am talking of a driver mounted in speakers costing tenth of thousands ... i think that better drivers are needed for this kind of speakers
But they are fashionable
Yes this is my point They show what will not raise concerns The nicest parts
I am a low distortion freak because i adore accurate speakers The ones that tell me everything, for the good and the bad I wan too see in the recordings
For this reason i am abandoning the hifi speakers to go with a monitor speaker
Speaking of cheap products usually the weak point are the woofers and the cabinets (usually plastic is used) So my idea is to focus on woofers performances and cabinets design and built
actually i was meaning just Frequency Response
Cumulative Spectral Decay
https://www.diyaudio.com/community/...dth-pro-15-and-18-drivers.367679/post-6528747
the anal guy its not me But i tend to be anal me too on the distortion But also CSD ... with FR they tell a lot about the mechanical behaviour of a driver
I think that you are perfectly right But my room is 15'x20'
if i had a bigger room i would use pro drivers like Faital, B&C, etc. ... for sure
this is interesting I did not know But i do not follow You dont want, rightly, HD and IMD Nevertheless you are not curios about the HD and IMD of the drivers you use ? this makes a little sense to me If low IMD is a good thing to have let's have it directly in the datasheets
Imho IMD is what separates very good drivers from so so drivers I am sure of this Tones are linked to IMD The people is so used to IMD that when they listen to a very low IMD with a very clean sound are puzzled They are used to garbage To dirt And they talk about a too clinical sound ... they are lost
👍👍👍
very very valuable advice and thank you again for this Presently i am looking for a very good 7-8" able to work decently also up to 2kHz
I have found one very good indeed in the Visaton AL200 But it is now 200euro/each A little on the expensive side But it is a remarkable driver indeed
For sizes below 8" the bass could be too weak For bass i mean 50-60Hz max
The AL200 would be also low in IMD I guess
There was rumors that woofers of a very famous brand were made cheaply in China I could not find them
https://www.audioasylumtrader.com/images/y2016/12/165396/IMG_0113.JPG
Hi ! very interesting indeed and thank you for your kind and valuable advice
May i ask you more about your procedure to check the pitch reproduction ? do you listen specific tracks ?
i think that your fav measurement could be the spectral decay of the driver The response to impulse vs Hz i mean
hi thanks a lot and very nice indeed The woofer is a 200 euro piece It seems that the price of admission to quality is around that threshold
I was hoping something less But this is indeed the woofer i would need I just wonder if 90% of its performance can be obtained for half the price
I would be quite done
Thank you and you’re welcome! I like to listen to a variety of different types of music with a focus on naturally recorded acoustic music. I think it’s also important to take time and not rush into a decision, although tbh, if the reproduction doesn’t immediately give the impression that you’re listening to a live event, it probably never will. Thanks for the tip too, although I must admit that I struggle to link what I see with what I hear. For example, only today I tried 0.43mH inductor in place of a 0.39mH inductor. The change in sound appeared to be far greater than XSim indicated it might be, with minimal changes to the X/O frequency, impulse response, square wave etc., however the music was suddenly injected with a sense of rhythm it previously lacked. Admittedly, the new inductor was a a slightly different design (round rather than square), although with the same wire csa and air core. I measure with REW and DATS (v3. Excellent bit of kit!), but as the BBC used to say, “If the sound doesn’t agree with the predictions, the predictions are wrong”! I know this will rub some people up the wrong way, but for me, the differences are very noticeable.
Hi i agree on this Processed music (i.e. amplified) is not the best for checking tones of voices and instruments
I like minimalist recordings like MA Recordings to name one label Very simple but very high quality recording chain mic>pre>recorder
yes but usually when a system is good i feel it using the recordings i know I have selected some cd tracks to use as a tool to check dynamics, soundstage and tones Even on Youtube there are some usable tracks imho
very interesting indeed I guess the inductor was in series with a cone driver ? have you tried to connect it on the - terminal of the driver (usually they are connected on the + terminal) ?
i am sure that if we hear something different the difference can be measured in some way ... for instance with a cumulative spectral decay test
low noise and low distortion and low ripple in the response to impulse are good things to get
i got a lot insults (probably rightly) because i asked about this crossover of the ProAc Studio 140 in the past
https://www.audioreview.it/wp-content/uploads/prove/pro-ac-studio-140.html/Figura-1.jpg
it is wired a little differently from all others with the inductor connected to the minus terminal of the woofer
But they told that this makes no difference in performances
The only thing i know, and your words confirm my doubts, is that of all parts that can be found in a crossover and inductor is by far the very worst one to be used in series to a woofer
And the fact that very good ones cost an arm and leg means that the impact of cheap parts is very evident For the bad
There are some inductor less speakers I guess they work with just particular woofers by the way
In inductor opposed to the variations of current flowing across it Exactly what happens with woofers A disaster
Last edited:
The passive way to cut high frequencies from a woofer is a series inductor. Allowing highs to run into the woofer wastes watts as heat and makes the woofer perform where the frequency response is bumpy, causing distortion.
Alternative methods are a op amp high cut filter previous the amplifier, which requires bi-amping and wastes electricity every hour you listen. Also a DSP previous the amp, which requires bi-amping; also a cellphone which changes op system so often the app to program the DSP may be obsolete in 2 or 5 years.
See the Eminence Beta-10a 10" driver since you don't have room for 15". Cone breakup at 2000-2500 hz which you don't want to listen to. Then above that a 25 db dive into jagged mountains. Anything in the >3000 hz area is just producing heat from your amplifier output.
The B&C 10FW64-8 has response with even more jagged mountains 2000-2500 Hz but I can't get the **** datasheet to download to a pdf file. Only a file of every driver B&C made in 2021.
My eminence Deltapro-15a has a rise above 1000 hz before cone breakup at 2000, which I am filtering off with a series inductor since I am crossing over to a CD horn at 1200 hz.
I let ramista claim of hearing pitch shift on speaker slide, post 25, but that post makes me feel he is hallucinating. I'm very sensitive to pitch and I've never heard pitch shift from any stationary speaker. I'm a bassoon player who had to control every note pitch with my mouth since the Q of a bassoon is so bad. I was good enough to make 5th chair in the All state TXMEA band 1968. I tune pianos this decade with 2 tuning forks, as a volunteer for several churches and my own inventory of 3.
I have heard high frequency IM distortion come out of a speaker, which sounded like vibrato on a top octave Steinway piano track. Pianos don't have vibrato. Cause was a 25' 16 ga zip cord used as speaker wire compared to 8' of 10 gauge wire which had stable pitch. Effect may have been limited to one amplifier.
What passive crossover components like inductors & capacitors will do is shift the phase of the music at different frequencies. Which some people claim messes up the soundstage. A whole infinite area of discussion. I just deal with the phase shift so I only have to purchase and keep running one stereo amp instead of the 2 of a biamp or 3 of a triamp.
Last edited:
hi thank you sincerely for your great advice
i am trying to learn something hoping to make wiser choices in the future
I think i will end with powered studio monitors solution to start
the room is quite small and full of things
my listening i think is between a near and a midfield situation
i would use the cabinets and the amps of two behringer b2031a and put a Tannoy dc inside I am afraid it's the best i can do at the moment
Thanks a lot again and kind regards gino
Rude!
Since you’ve pulled the “expert” card, I have to bow to your undoubted superior knowledge. You’re obviously correct and my experience is clearly better explained by reference to a psychiatric disorder………. Not! It’s a shame that you have to adopt such a superior position in order to make your point. Also, you do realise that suggesting a change of speaker wire can unsettle a notes pitch puts you in the same psychiatric category (in many peoples eyes, but not mine, I hasten to add) as the one to which you feel I belong!
speaking of pitch i think that some standard tests can say one or two things about the mechanical behaviour of a speaker (i. e. drivers plus cross over plus cabinet)
i really don't understand why some tests routinely carry out on amplifiers are not carried out on speakers
one of the reason i think is magazines
in the past they used to do some tests
now every review is a novel
words words words when graphs for instance are much more telling
just think how many words you need to describe a graph or a plot
frequency response
cumulative spectral decay
distortion and
h and v dispersion is all we need
i really don't understand why some tests routinely carry out on amplifiers are not carried out on speakers
one of the reason i think is magazines
in the past they used to do some tests
now every review is a novel
words words words when graphs for instance are much more telling
just think how many words you need to describe a graph or a plot
frequency response
cumulative spectral decay
distortion and
h and v dispersion is all we need
I don't think your earlier example is the same as a series crossover.
The series crossover is actually a cool thing. The way the parts come together can be helpful, giving automatic approximate impedance compensation, compensation for the directivity hole near the crossover and greater than 1st order (nearing 2nd order) while using a minimum of parts.
Speakers do not have crossover distortion like amplifiers do. Speakers don't have high orders of harmonic distortion.
Speakers have simple 2nd and 3rd order distortions.
