I also don't think a full flat response is mandatory, but knowing what you do is always better than guessing, so measuring, simulating and testing your results work better than tuning with your ears only. But the final judgement is our ears. A full flat response often is very lifeless and fatigueing to my ears, a little bass heavy sound with a very smooth roll off above 12kHz sounds way better to my ears and does sound more real than the objective real.
But one thing i learned about audio over the years, everybody hears different, and so appreciate different things and sounds. So there is no absolute law what sounds best. Only opinions. But that does not mean we can't use objective measurements and a systematic approach to get to that good sound (whatever that may be for you). So i use both ears and measurements to get where i want to be.
But one thing i learned about audio over the years, everybody hears different, and so appreciate different things and sounds. So there is no absolute law what sounds best. Only opinions. But that does not mean we can't use objective measurements and a systematic approach to get to that good sound (whatever that may be for you). So i use both ears and measurements to get where i want to be.
Yes, I agree with you regarding flat response, not always good. Of course it is better measuring and after tweak and also by ears.
Exactly everybody's ears are different.
For curiosity I recorded our speaker with a very cheap microphone. Unfortunatley the quality very low but can be heard roughly. I think it is not bad if we considering that is tuned by ear
Microphone record
Exactly everybody's ears are different.
For curiosity I recorded our speaker with a very cheap microphone. Unfortunatley the quality very low but can be heard roughly. I think it is not bad if we considering that is tuned by ear
Microphone record
The armchair philosopher speaks : if you are truly doing DIY, ask yourself: who are you trying to please? Only yourself. That's easiest. Not needing to produce a commercial design at minimum means: you don't have to worry about mark-ups etc. nor making a compromise system that will tend to sound merely "ok" in the "average" end-user location. The mere status of doing a custom design, being able to tune your design to a specific set of components, listening room, and listener puts you at a huge advantage over even the best of mass-produced speakers.
A common enemy both with DIY and store-bought is the perennial problem of Ego. Is this approval of your social circle important? If it is, your job is many times harder. It's remarkable if you can even answer the question honestly. Again, if you're the rare guy who can really say, "No, I only need to please myself," your job is far easier. Given even mdoerate starting components and test gear, you can get very close to ideal, call it "optimum." Yes, having measuring gear helps, and is cheap today, but you can go far with even just a tone generator or pink noise bands and careful listening. An active x-over is much easier to tweak and electronics are so cheap these days, there is almost no excuse for not having one. If you are old school and/or must have a true crossover, again, you can use active to get close to your ideal, then use design software or trail & error to make an analogue x-over. The ultimate test is: does it sound good to you? Screw everybody else!
: if you are truly doing DIY, ask yourself: who are you trying to please? Only yourself. That's easiest. Not needing to produce a commercial design at minimum means: you don't have to worry about mark-ups etc. nor making a compromise system that will tend to sound merely "ok" in the "average" end-user location. The mere status of doing a custom design, being able to tune your design to a specific set of components, listening room, and listener puts you at a huge advantage over even the best of mass-produced speakers.A common enemy both with DIY and store-bought is the perennial problem of Ego. Is this approval of your social circle important? If it is, your job is many times harder. It's remarkable if you can even answer the question honestly. Again, if you're the rare guy who can really say, "No, I only need to please myself," your job is far easier. Given even mdoerate starting components and test gear, you can get very close to ideal, call it "optimum." Yes, having measuring gear helps, and is cheap today, but you can go far with even just a tone generator or pink noise bands and careful listening. An active x-over is much easier to tweak and electronics are so cheap these days, there is almost no excuse for not having one. If you are old school and/or must have a true crossover, again, you can use active to get close to your ideal, then use design software or trail & error to make an analogue x-over. The ultimate test is: does it sound good to you? Screw everybody else!
Last edited:
We could/should have a contest - we specify two basic drivers woofer and tweeter that anyone can reasonably obtain as the price of admission to the contest (but not used often enough that canned XO's are on the net), and see who can make the most musically satisfying XO (usually will be neutral flat response that works with all genre's of music). We have to be on the honor system and not use a mic, REW, Xsim, or any computer or software means to design it. Only your ears and pencil and paper, maybe a handheld calculator.
After completion, share your design to a neutral third party (like a moderator) to keep secret until the end of the contest when all is revealed. At that point, you can measure it with a mic and show freq response plots, measure impedance etc. or even show how it looks on Xsim/PCD/or other software simulator.
How to judge: let people vote on the post-analysis results. If we could all be in one room we would set them up with same amp, source, and music and listen. But we can't do that.
You could really add a whole new level of complexity by making it a 3-way.
Now, how to find obscure drivers that level the playing field? The drivers need to have a good factory frequency response chart, imepdance chart, and accurate TS parameters. And be reasonably priced and available in US/EUR/AUS readily.
After completion, share your design to a neutral third party (like a moderator) to keep secret until the end of the contest when all is revealed. At that point, you can measure it with a mic and show freq response plots, measure impedance etc. or even show how it looks on Xsim/PCD/or other software simulator.
How to judge: let people vote on the post-analysis results. If we could all be in one room we would set them up with same amp, source, and music and listen. But we can't do that.
You could really add a whole new level of complexity by making it a 3-way.
Now, how to find obscure drivers that level the playing field? The drivers need to have a good factory frequency response chart, imepdance chart, and accurate TS parameters. And be reasonably priced and available in US/EUR/AUS readily.
Last edited:
Karoly, you are right a system sound should be tuned by ear if you are DIY. I started with a 6922 preamp, a very flat new amp +-0.02 db with tone stack centered. then tweak the RIAA , FM and iphone inputs to have the same frequency response balance as the CD /DVD input.
The last step is to tweak the HPM-900 midrange response to sound like exotic $200k speakers. it's in the phase and output level of the midrange.
The last step is to tweak the HPM-900 midrange response to sound like exotic $200k speakers. it's in the phase and output level of the midrange.
Last edited:
Stockrader200, yes but I think it is more individual. Wow it was a very exciting project then and in the end you have success Congratulations!
I recorded a few jazz and we really enjoy our speaker. Unfortunately with this cheap microphone the sound is rather bad, and so sensitive which position I place, but probably it is possible to imagine what is in the background. Rich instruments, many details, big soundstage, deep bass and high dynamic. I also observed that it doesn't mean what kind of room or where I put the sound is always good. Also involving and very coherent. Otherwise every year we go to the big Copenhagen Hi-Fi exhibition and honestly more than 50% of the speaker (very expensive ones) we don't like and even cannot come close to our lovely 47 years old speaker. Of course I replaced the midrange and the tweeter, only the woofer is the original Pioneer. Please, listen people our presentation. Thank you.
Record II
Congratulations!I recorded a few jazz and we really enjoy our speaker. Unfortunately with this cheap microphone the sound is rather bad, and so sensitive which position I place, but probably it is possible to imagine what is in the background. Rich instruments, many details, big soundstage, deep bass and high dynamic. I also observed that it doesn't mean what kind of room or where I put the sound is always good. Also involving and very coherent. Otherwise every year we go to the big Copenhagen Hi-Fi exhibition and honestly more than 50% of the speaker (very expensive ones) we don't like and even cannot come close to our lovely 47 years old speaker. Of course I replaced the midrange and the tweeter, only the woofer is the original Pioneer. Please, listen people our presentation. Thank you.
Record II
Karoly, can you post a picture of your speakers mine are the Pioneer HPM-900 with original ( unobtainium ) graphite cones that were re foamed last year.
your speakers sound pretty good even with a bad mic. I have been testing mine with Hello Dolly (louis Armstrong) and they sound great.
what mid and Tweet did you use?
your speakers sound pretty good even with a bad mic. I have been testing mine with Hello Dolly (louis Armstrong) and they sound great.
what mid and Tweet did you use?
Attachments
Last edited:
stocktrader200, wow I really like your speaker so much. I can imagine the sound is very very good. My midrange is Seas MCA15RCY H1262-8, the tweeter is Monacor DT-254, but I am planning to replace with Seas 27TDFC. Crossover points 560Hz and 3400Hz. Otherwise the sound indeed already good, how would be with the Seas dome tweeter? More details, more extended?
For experiment I put a circular felt material around the tweeters hole and really improved it, slightly more focused the high frequencies.
Last year I sold one of our CD player and the buyer asked, what kind of speaker do I have? I told, why? And he: because the sound is extremely involving, warm and fantastic. I told we have this vintage from 1973 and he didn't want to believe it and just remarked that do you know it has a much better value than any of the new modern one today. I just smiled at him and I told myself, yes, I know...in the end do you know what he asked? That I am willing to sell it to him? I told no, never want to sell it, I definitely would like to keep it...
For experiment I put a circular felt material around the tweeters hole and really improved it, slightly more focused the high frequencies.
Last year I sold one of our CD player and the buyer asked, what kind of speaker do I have? I told, why? And he: because the sound is extremely involving, warm and fantastic. I told we have this vintage from 1973
and he didn't want to believe it and just remarked that do you know it has a much better value than any of the new modern one today. I just smiled at him and I told myself, yes, I know...in the end do you know what he asked? That I am willing to sell it to him? I told no, never want to sell it, I definitely would like to keep it...Attachments
Last edited:
Hi all,
I came across this thread earlier this morning, and have read the first 20 posts. So far, very good! I especially thank Wintermute for the excellent posts which, in essence, highlight the important differences between generic modeling, realistic modeling, measurement, and guesswork. (Not to slight other helpful posters.) I'll read more of the thread later today.
I came across this thread earlier this morning, and have read the first 20 posts. So far, very good! I especially thank Wintermute for the excellent posts which, in essence, highlight the important differences between generic modeling, realistic modeling, measurement, and guesswork. (Not to slight other helpful posters.) I'll read more of the thread later today.
How did you get the get the High and Low pass impedance for the calculator?
I am looking at the specs for and Illuminator tweeter R3004/6620, that has a min imp of 3.9ohm and max imp of 20.6ohm.
The midwoofer has min imp 7.9ohm and max imp of 76.9ohm.
How can I derive the values from these specs?
Thank for replying.
Werner
From my experience, flat frequency response is distant third in subjective importance, after non-harmonic and harmonic distortion.
Long time ago I was also a subscriber to this speaker-building orthodoxy that flat response equals to good sound. Once I saw a tweeter on Parts Express, with exceptipnally flat response and at reasonable price. The response curve was indeed as advertized, but the tweeter was unlistenable. Then there was experience with BG Neo-8, whose response curve is horrendous, but sound quality leaves nothing to desire.
Remedies used to correct uneven frequency response, such as complex high order crossovers, are by themselves worse malady than the one they are intended to cure. Capacitors and inductors degrade the sound, with effects compounding as their numbers increase.
Folks who rely solely on measurements get themselves into dead end pursuing this path. SPL strongly depends on microphone placement and frequency - does it mean that listener's head has to be affixed to a certain position?Ridiculous, but that's the logical conclusion from the orthodoxy. In fact, a good system sounds good in broad area, not a tiny sweet spot.
Long time ago I was also a subscriber to this speaker-building orthodoxy that flat response equals to good sound. Once I saw a tweeter on Parts Express, with exceptipnally flat response and at reasonable price. The response curve was indeed as advertized, but the tweeter was unlistenable. Then there was experience with BG Neo-8, whose response curve is horrendous, but sound quality leaves nothing to desire.
Remedies used to correct uneven frequency response, such as complex high order crossovers, are by themselves worse malady than the one they are intended to cure. Capacitors and inductors degrade the sound, with effects compounding as their numbers increase.
Folks who rely solely on measurements get themselves into dead end pursuing this path. SPL strongly depends on microphone placement and frequency - does it mean that listener's head has to be affixed to a certain position?Ridiculous, but that's the logical conclusion from the orthodoxy. In fact, a good system sounds good in broad area, not a tiny sweet spot.
I'm certainly no expert and even though I've been an audiophile for 25 years and built speakers (even worked for a speaker manufacturer) I'm only now getting into measuring and simulating.
Having said that when I was playing around with a miniDSP in a 2-way a few years ago it struck me that no matter what crossover slope, point, or EQ shape (to a point) I used for those particular drivers, the sound quality itself didn't seem to get a whole lot worse or better.
Of course that was only 1 set of drivers so I'll probably get flamed for this but I'm thinking Vince Bruzzese of Totem was right when he said the biggest determinant of sound quality in a speaker is driver synergy. In other words, you have to find the right combination of woofer, mid, tweeter etc. Which translates into having to have a lot of drivers on hand and experiment with them. Not something a lot of do-it-yourselfers have the resources for.
Having said that when I was playing around with a miniDSP in a 2-way a few years ago it struck me that no matter what crossover slope, point, or EQ shape (to a point) I used for those particular drivers, the sound quality itself didn't seem to get a whole lot worse or better.
Of course that was only 1 set of drivers so I'll probably get flamed for this but I'm thinking Vince Bruzzese of Totem was right when he said the biggest determinant of sound quality in a speaker is driver synergy. In other words, you have to find the right combination of woofer, mid, tweeter etc. Which translates into having to have a lot of drivers on hand and experiment with them. Not something a lot of do-it-yourselfers have the resources for.
Last edited:
The big problem is absolutely unshakable arrogance that one's ears are anything but very very blunt instruments with only the smallest talent for hearing analytically or pronouncing anything more definitive about a frequency spectrum than "... the trumpet seems kind of bright on that recording..." or "hmmm, no bad bass emphasis" or "too distorted"..... stuff like that.
One thing that remains after you try all the options with a given set of drivers is their acoustic conditions. Under certain conditions you couldn't hope to tune properly by ear.. however when crossovers are done properly, acoustic conditions are favourable and working well with the room, and you have enough time then tuning by ear can take you all the way.
Even though I didn't say it, I in no way meant that speakers should not be measured and modeled and that one should rely solely on the ears for the entire process. What I meant was that if you for example took a 2 way speaker from a manufacturer like Revel or PSB (who design by scientific principles and do measure and model) and you replaced their crossover with a generic off-the-shelf 'textbook' one, that many would be surprised that the sound is subjectively similar to the designed crossover. IOW they've chosen/created drivers that play nice in the first place. From there on you make it as best as possible objectively and subjectively. I could be wrong but that makes sense to me.