Attachments
In the studio realms, I had Amphion one18 and I thought these were amazing. I then bought harbeth p3esr that cross their midbass to the tweeter around 3.5khz and I preferred greatly how the mid/treble sounded. I then bought the new atc scm7v3 sealed and I find them good, crossing around 2khz, but not as good as the P3esr. then I tried the kef ls50 which cross around 2khz, again I prefer the P3esr. oh I also had the jbl lsr32, and no contest, the small P3esr in the mids are more coherent, better and more integrated, by far.
I have had Tannoy 12, 15, in gold, hpd version and now I use in my living room for my secondary system LGM 12. all those cross around 1.2khz. all modified with custom crossover or stock, I dont hear the supposedly advantage of lower crossover. I do find that a CP sounds more dynamic, but less detailed at least in the tannoy. and its no contest, my harbeth shl5 plus give much better resolution even with female vocals, and also better imaging, even though the tannoy is coaxial and should do better theoretically.
I guess I may be in the "wide bander" camp.
it seems that so far, I find that the higher the mids cross, the better it sounds. My favorite diy speaker used a 6.5 inch audax mid from 300hz up to 5/6khz and I loved it. My actual speakers now are harbeth shl5plus, which uses a 8 inch up to 3.2khz and its so far the best "compromise" ive heard. I dont notice the beaming problem since my room is treated and my head is always at the same optimal place.
I have had Tannoy 12, 15, in gold, hpd version and now I use in my living room for my secondary system LGM 12. all those cross around 1.2khz. all modified with custom crossover or stock, I dont hear the supposedly advantage of lower crossover. I do find that a CP sounds more dynamic, but less detailed at least in the tannoy. and its no contest, my harbeth shl5 plus give much better resolution even with female vocals, and also better imaging, even though the tannoy is coaxial and should do better theoretically.
I guess I may be in the "wide bander" camp.
it seems that so far, I find that the higher the mids cross, the better it sounds. My favorite diy speaker used a 6.5 inch audax mid from 300hz up to 5/6khz and I loved it. My actual speakers now are harbeth shl5plus, which uses a 8 inch up to 3.2khz and its so far the best "compromise" ive heard. I dont notice the beaming problem since my room is treated and my head is always at the same optimal place.
Last edited:
^I too think that it is a better compromise to cross tweeter near 4K in three-ways and with well behaving midwoofers in 2-ways. I give two possible explanations:
- Many rooms show longer RT/decay times between 1-4kHz than 300-1k and above 4k, so some beaming is good to achieve balanced subjective room response
- Crossing between 1-3kHz means that we get phase warp there, which makes problems to harmonics of female voice and most instruments. Very often xo's are not even close to perfect in commercial speakers!
DSP xo gives the user possibility to easily test various xo points and configurations. We must ofcourse be aware of driver response anomalies and beaming characteristics. As to LR2 vs LR4 sound, I prefer LR2 but it is more difficult to get measuring really well.
- Many rooms show longer RT/decay times between 1-4kHz than 300-1k and above 4k, so some beaming is good to achieve balanced subjective room response
- Crossing between 1-3kHz means that we get phase warp there, which makes problems to harmonics of female voice and most instruments. Very often xo's are not even close to perfect in commercial speakers!
DSP xo gives the user possibility to easily test various xo points and configurations. We must ofcourse be aware of driver response anomalies and beaming characteristics. As to LR2 vs LR4 sound, I prefer LR2 but it is more difficult to get measuring really well.
I've wondered about this (low vs high crossover point) and tried two well designed crossovers for two identical pairs of speakers. The speakers used an Usher 9950 tweeter and an Usher 8948 7" midwoofer. One pair were crossed at about 2700hz the other at about 1700hz, these were both 4th order although I did later try 2nd order with a latter delay network at about 2700 also.
Unfortunately the measurements and details are long gone, this was about 6-8 years ago. I do however remember being surprised how similar they sounded and even measured (there were differences especially off axis), maybe I was expecting too much. I found I had to move off axis quite a bit to notice a difference and even then it really depended quite a bit on the recording.
It's seems the Diy-ers like to push the crossover as low as possible for more even power response while the manufacturers seem more conservative and maybe worry about burning up tweeters.
I was satisfied with the sound of both approaches.
Unfortunately the measurements and details are long gone, this was about 6-8 years ago. I do however remember being surprised how similar they sounded and even measured (there were differences especially off axis), maybe I was expecting too much. I found I had to move off axis quite a bit to notice a difference and even then it really depended quite a bit on the recording.
It's seems the Diy-ers like to push the crossover as low as possible for more even power response while the manufacturers seem more conservative and maybe worry about burning up tweeters.
I was satisfied with the sound of both approaches.
Your assumption is not inaccurate. The surprise was how little I could hear the difference in my regular listening position. They were both very good sounding.
I've done the same and can't stand the higher x-over, moving head just a tiny bit alters the response. But as long as head is in one place
) The sound does not change much...Same with my Genelec 1030A in far field, they are crossed at 3.2 kHz. However, these are made for near field...
I have asked friends of mine to listen. Quite often, they can't hear any problems until told what to listen for...
I've done the same and can't stand the higher x-over, moving head just a tiny bit alters the response. But as long as head is in one place
Same with my Genelec 1030A in far field, they are crossed at 3.2 kHz. However, these are made for near field...
I have asked friends of mine to listen. Quite often, they can't hear any problems until told what to listen for...
Maybe it's time for me to do this experiment again with additional listeners. I have a pair of Kairos which cross pretty low at about 1800. Does anyone have a good design for the same drivers at say 2700-3000hz? I have an additional set of the same drivers, but I'm not competent enough to design a good passive crossover.
It was a huge JBL placed in a small office with a lot of clutter. I did not remember the model. They were in effect used almost as near fields and were perfectly capable of blowing ear drums. I was not overly impressed with the tonal balance. I had the feeling that the same would have sounded way better in an appropriately sized venue - hence the "need room to breath" comment.
Music reproduction consists of the reproduction of the sound and the reproduction of the experience. I buried my head too deeply in the former and my audiophile friends with deeper pockets than they have knowledge helped me understand the latter.
Think about the largest orchestral pieces, like Mahler 6, or popular test material like 1812. You can use the most expensive electrostatic headphones to perfectly reproduce the sound, but the scale is completely wrong. In the quest of "what is the best speaker?" only the most dynamic loudspeakers can reproduce the experience of large scale orchestral (realistically, that speaker has to use HLCD - high efficiency low distortion) - AND your actual listening room needs to be large enough for that reproduction of experience. The aforementioned JBL is not appropriately sized for its room (or the other way around?) so it sounds weird. The scale is wrong.
You are the resident HLCD expert here and my concepts might have been entirely mistaken. I did not mean horns cannot be listened up close - just look at all the near field monitors with horns. What I meant was certain horn designs are supposed to be BIG, and they are weird to listen up close, because the scale is wrong.
Music reproduction consists of the reproduction of the sound and the reproduction of the experience. I buried my head too deeply in the former and my audiophile friends with deeper pockets than they have knowledge helped me understand the latter.
Think about the largest orchestral pieces, like Mahler 6, or popular test material like 1812. You can use the most expensive electrostatic headphones to perfectly reproduce the sound, but the scale is completely wrong. In the quest of "what is the best speaker?" only the most dynamic loudspeakers can reproduce the experience of large scale orchestral (realistically, that speaker has to use HLCD - high efficiency low distortion) - AND your actual listening room needs to be large enough for that reproduction of experience. The aforementioned JBL is not appropriately sized for its room (or the other way around?) so it sounds weird. The scale is wrong.
You are the resident HLCD expert here and my concepts might have been entirely mistaken. I did not mean horns cannot be listened up close - just look at all the near field monitors with horns. What I meant was certain horn designs are supposed to be BIG, and they are weird to listen up close, because the scale is wrong.
Last edited:
I really wish I had more knowledge, and could figure out what I am hearing that is so different. I come from the camp of only having owned five to ten thousand dollar commercial Hi Fi speakers, for the past 32 years. I have never used Pro drivers for home use. Nor have I ever owned any hyper sensitive horn based Hi Fi speakers. I had a phone friend, who used to tell me 'If you ever hear a good 100db sensitive horn system, you will convert. I haven't ever had the opportunity yet.
But what I did find was a free pair of 96db efficient speakers, that uses a 10" Pro woofer, crossed at 2400hz, to an air motion transformer. And the speakers have many flaws compared to the Hi Fi speakers I have owned and heard .... but they get voices, drums, and cymbals/percussion down to a very eerie and realistic, in the room act. I attribute most of this to the AMT. But this odd combo is more enjoyable to listen to, than all but one pair of speakers I owned. And it has made me decide to build a much better 3 way pair using these same tweeters, but with very good Pro woofers and midrange drivers.
I am hoping to get the best of both worlds. I just wish I knew enough, to know just what is so different I hear on these 40 year old speakers.
But what I did find was a free pair of 96db efficient speakers, that uses a 10" Pro woofer, crossed at 2400hz, to an air motion transformer. And the speakers have many flaws compared to the Hi Fi speakers I have owned and heard .... but they get voices, drums, and cymbals/percussion down to a very eerie and realistic, in the room act. I attribute most of this to the AMT. But this odd combo is more enjoyable to listen to, than all but one pair of speakers I owned. And it has made me decide to build a much better 3 way pair using these same tweeters, but with very good Pro woofers and midrange drivers.
I am hoping to get the best of both worlds. I just wish I knew enough, to know just what is so different I hear on these 40 year old speakers.
Achieving the right balance for proper reproduction is a complex and involved process. Regardless, I find that it is difficult to find recordings that scale some things properly. The 1812 is a good example because the cannons may have been recorded elsewhere and brought into the mix.. not to mention the difficulty in recording them well.
Scale involves the stereo spatial information as well as the response. The recording should sound different depending on where it was made, eg made outside it will sound different in both response and spatial cues.
The regions where reproduction can be affected by a baffle becoming small, and be affected by the room are difficult to get right. Tactile feel around 200Hz or so is often not reproduced well IME.
Diffraction/horn reflections can be a problem here. For some types of horns there is no simple answer to this. Horns in a domestic situation need to be chosen carefully for their purpose.
ESS Tempest towers, with the round Heil AMT. Ancient, ugly, technically fraught with design faults. Yet does something no Hi Fi speaker ever has in my listening experience. Sounds more like real live music being played in my room. Makes me wonder if I have wasted 33 years of listening to music in a sub par way, and on a sub par audio chase.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.