First trials of active horn speaker

[morricab]

The recommended minimum crossover frequency of the Radian is 1.2 kHz, and imo this makes sense, as its impedance peak (and presumably its resonant frequency) seems to be about 850 Hz.

It sounds to me like your crossover doesn't include any equalization of the driver's response. If that is the case, then whichever driver has the best innate frequency response on the horn you're using will probably "win".

Ime frequency response is what matters most, then once the frequency response is good, the Beryllium diaphragm's freedom from breakup starts to be a worthwhile improvement. But Beryllium (and other exotic diaphragm materials) can't compensate for having an audibly worse frequency response.

I have some eq built into the design but it is probably not exactly right. Once I measure then I will know.

 

[bonzo75]

Brad, if you are keeping this to a 2way, most bass solution require the bass being crossed:
1. with full rangers below 200
2. Otherwise anywhere up to 800

Withe Radian 475 being crossed anywhere from 1.2k to 2k, I don't see how you are going to cross over to the bass, unless I am missing something. With a Iwata of 600, minimum crossover should be 1.2k. I would suggest a driver that you can cross lower and a horn with a further lower Fc. For the size of your horn, a horn with a lower Fc would mean a tractrix.

 

[morricab]

Brad, if you are keeping this to a 2way, most bass solution require the bass being crossed:
1. with full rangers below 200
2. Otherwise anywhere up to 800

Withe Radian 475 being crossed anywhere from 1.2k to 2k, I don't see how you are going to cross over to the bass, unless I am missing something. With a Iwata of 600, minimum crossover should be 1.2k. I would suggest a driver that you can cross lower and a horn with a further lower Fc. For the size of your horn, a horn with a lower Fc would mean a tractrix.

WHat you are saying makes sense if I want to go to a 3-way design, which I will likely do. Then I would add woofers roll off below 200hz BUT the Supravox will remain full range and also contribute to the bass (DTQWT). However, what you say it makes no sense as a two-way in a TQWT. I am using a wide band 8 inch driver that goes from around 40hz to wherever I cross it.

You seem to be missing a lot, Ked about my design, despite the fact it has been described clearly and photos show. It is not a 15 or 18 inch woofer, it is an 8 inch wideband driver. You also seem to miss the fact that I have tried down to 600Hz with the larger horn and found 1Khz for that driver was optimal. For the smaller Iwata horn, it seems a high frequency works equally well for the Beyma driver and might work better fort he Radian. All you have to do is look at the Supravox driver I am using (215-2000 mKI not mkII) and you will realize that performs up to 18Khz but the dispersion starts to narrow sharply above 3-4khz.

Anyway, design a few speakers yourself and then give advice...or at least understand what I am trying to do before commenting (hint; it is not a FLH with large Altec horn).

 

[bonzo75]

you cannot cross over at less than 1.2k in a 600 Hz horn. It will sound wrong of a driver that doesn't go there. You can't do something to just do it wrong

 

[kodomo]

you cannot cross over at less than 1.2k in a 600 Hz horn. It will sound wrong of a driver that doesn't go there. You can't do something to just do it wrong

No, it is not a rule, though I adhere to it in my midrange and upper midrange. My mids are 200hz tractrix although my cut off is around 500hz, upper mid horn is 1400hz but I cross around 4.5khz but for the midbass, it does not need to be, you can use reactance annuling adjusting the size of your back chamber and drop the response down to cutoff of the horn. It all depends on the driver, the crossover and the horn itself.

600hz horn would load fully and control the pattern at 1.2khz, that is right but it would start providing gain and pattern control from about 900hz as well.

 

[bonzo75]

No, it is not a rule, though I adhere to it in my midrange and upper midrange. My mids are 200hz tractrix although my cut off is around 500hz, upper mid horn is 1400hz but I cross around 4.5khz but for the midbass, it does not need to be, you can use reactance annuling adjusting the size of your back chamber and drop the response down to cutoff of the horn. It all depends on the driver, the crossover and the horn itself.

600hz horn would load fully and control the pattern at 1.2khz, that is right but it would start providing gain and pattern control from about 900hz as well.

It will do it but everyone who had a good horn that I heard recommended it at 2 times. Brad has already tried lower and he said it didn't sound right. Universum gets it right with the Radian because that's what they do for the mids (from what you are saying you do the same with your radians). So Brad will end up using his Radian driver only for treble if he keeps going higher

 

[kodomo]

Yes, I guess I just do not like generalizations as they would make you miss a lot of details and keep you from learning

ps. Also, I do not think midbass of Leif is crossed an octave above and universums midbass band is not covered completely by a flh horn

 

[bonzo75]

Yes, I guess I just do not like generalizations as they would make you miss a lot of details and keep you from learning

ps. Also, I do not think midbass of Leif is crossed an octave above and universums midbass band is not covered completely by a flh horn

Universum is a hybrid, that is one if it's slight shortcomings, but that is a separate point. I am referring to their 950 in the 200T, which crosses just under 400. Leif's 4003 crosses at 550 in a JMLC 270.

 

[kodomo]

Universum is a hybrid, that is one if it's slight shortcomings, but that is a separate point. I am referring to their 950 in the 200T, which crosses just under 400. Leif's 4003 crosses at 550 in a JMLC 270.

Yes, that is what I do as well and even more as I have 6db filters, I need wider linear response, hence 200 horn, 500 xo, and 1400 horn and 4.5khz xo

 

[bonzo75]

Yes, that is what I do as well and even more as I have 6db filters, I need wider linear response, hence 200 horn, 500 xo, and 1400 horn and 4.5khz xo

Well that's exactly what I suggested to Brad, first the 1200 was suggested due up him mentioning 600 Fc on the iwata back in March. Then it was mentioned last Thursday as that was the minimum, though Radian users suggest much higher, but if his plan was the do it lower initially he needs a different horn and driver

https://www.whatsbestforum.com/threads/first-trials-of-active-horn-speaker.28267/page-4#post-634254


https://www.whatsbestforum.com/threads/first-trials-of-active-horn-speaker.28267/page-12#post-671184

 

[morricab]

you cannot cross over at less than 1.2k in a 600 Hz horn. It will sound wrong of a driver that doesn't go there. You can't do something to just do it wrong

Well, actually you can and if it works great, if not try again. At hifi levels there is not so much strain on the drivers like there is at PA levels. That said, the Beyma works fine (it is solid down to 600hz) and the Radian was not fine. So, I have now gone to 2Khz with the Iwata 600 horn but I may come down from there if the Radian sounds great. Then I will go down until it sounds like it did before and then come up again to where it works best.

For the larger Beyma + XT1464 horn 1khz is not a problem at all and sounds really great.

 

[morricab]

It will do it but everyone who had a good horn that I heard recommended it at 2 times. Brad has already tried lower and he said it didn't sound right. Universum gets it right with the Radian because that's what they do for the mids (from what you are saying you do the same with your radians). So Brad will end up using his Radian driver only for treble if he keeps going higher

I don't see it as a problem if I need to use the Radian "only" for treble (BTW. 1Khz is already lower treble). The Supravox is a superb high frequency speaker and only the highest treble and beaming is an issue. The tweeter has to be damn good to keep up.

I was trying lower to see if it works better or worse...for the Iwata and Beyma 1Khz was ok for the Radian not. For the bigger horn 650Hz was not ok, 800hz was better and 1Khz was best (above that I didn't like how the drivers blended). Rather than being dogmatic with a 2x rule, I do actual experimentation and find the sweet spot for the driver/horn. With an active xo it is easy to experiment.

 

[morricab]

Well that's exactly what I suggested to Brad, first the 1200 was suggested due up him mentioning 600 Fc on the iwata back in March. Then it was mentioned last Thursday as that was the minimum, though Radian users suggest much higher, but if his plan was the do it lower initially he needs a different horn and driver

https://www.whatsbestforum.com/threads/first-trials-of-active-horn-speaker.28267/page-4#post-634254


https://www.whatsbestforum.com/threads/first-trials-of-active-horn-speaker.28267/page-12#post-671184

Ked, I don't need you to tell me the "rules" of horns. Do you seriously think I don't know about the recommendation about 2x from the horn cutoff? Do you really think I haven't done any homework on this ?? That I need you, who has neither designed nor built anything (this is my fourth pair of speakers over the years...including a very nice pair of ribbon hybrids), to instruct me on obvious and basic things?

 

[morricab]

Yes, that is what I do as well and even more as I have 6db filters, I need wider linear response, hence 200 horn, 500 xo, and 1400 horn and 4.5khz xo

With first order your driver needs to be able to handle the expected excursion from going quite a bit lower than a steeper filter, even with a significantly higher frequency.

 

[bonzo75]

Ok

Ked, I don't need you to tell me the "rules" of horns. Do you seriously think I don't know about the recommendation about 2x from the horn cutoff? Do you really think I haven't done any homework on this ?? That I need you, who has neither designed nor built anything (this is my fourth pair of speakers over the years...including a very nice pair of ribbon hybrids), to instruct me on obvious and basic things?

Yes I know the about the big planars in small rooms under driven on SETs which never can sound right to anyone except yourself. I guess you are going the same way your DSPed horns changing cross over from 800 Hz to 2k without any reference except those tiny Odeon in your video


But with ability to flatten out a Radian soundstage I guess you are the expert you wanted to be. Plus you will report it based on your incorrect assessment of your own set ups just like you do unfairly to most gear you walk past at Munich

 

[kodomo]

With first order your driver needs to be able to handle the expected excursion from going quite a bit lower than a steeper filter, even with a significantly higher frequency.

Yes, no worries there, there is also parallel impedance correction on my crossovers for linearity. When I say first order, people think they are simple, there are 8 separate boards for my crossovers involving vacuum tube capacitors, russian micas etc.

My distortion figures are very very low, and my measured response is very linear, do not forget, I have five channels sharing the load so not much to do for any of my drivers

 

[morricab]

Yes I know the about the big planars in small rooms under driven on SETs which never can sound right to anyone except yourself. I guess you are going the same way your DSPed horns changing cross over from 800 Hz to 2k without any reference except those tiny Odeon in your video


But with ability to flatten out a Radian soundstage I guess you are the expert you wanted to be. Plus you will report it based on your incorrect assessment of your own set ups just like you do unfairly to most gear you walk past at Munich[/QUOTE]


I am not even talking about my big planars, which sounded extremely good to everyone who actually heard them in my room (and inspired a lot of copycats). I am talking about something else I made from scratch about 15 years ago (ask Christoph...he thought they were excellent).

I am not changing blindly, although it is a matter of seconds to do so. I am also measuring in addition to listening.

Did your reading comprehension drop through the floor?? I noted that the Radian was not working well at the settings I had (it was an easy experiment as a drop in replacement and it didn't work and I duly noted this)...I am reporting the progress of development, not final a final configuration. Show me one DIYer who came up with the final design in one go that wasn't a copy of another system. Do you have any clue what goes into the development of something?? Likely not.

I don't take your BS opinions seriously...how could I when a man with no experience or experimental drive wants to try to lecture me on the "right" way to develop a horn speaker (or any speaker for that matter)?

 

[morricab]

Ok


Yes I know the about the big planars in small rooms under driven on SETs which never can sound right to anyone except yourself. I guess you are going the same way your DSPed horns changing cross over from 800 Hz to 2k without any reference except those tiny Odeon in your video


But with ability to flatten out a Radian soundstage I guess you are the expert you wanted to be. Plus you will report it based on your incorrect assessment of your own set ups just like you do unfairly to most gear you walk past at Munich

Oh, and feel free to stop commenting on my thread because you have nothing but obvious dogma to present to the discussion.

 

[morricab]

Yes, no worries there, there is also parallel impedance correction on my crossovers for linearity. When I say first order, people think they are simple, there are 8 separate boards for my crossovers involving vacuum tube capacitors, russian micas etc.

My distortion figures are very very low, and my measured response is very linear, do not forget, I have five channels sharing the load so not much to do for any of my drivers

Yes, first order crossovers can be very complicated (I remember that the Vandersteen and Thiel crossovers have A LOT of elements). Do you have impulse response measurements or step function measurements to show how well you managed to get those large horns time aligned?

 

[Alrainbow]

Yes, no worries there, there is also parallel impedance correction on my crossovers for linearity. When I say first order, people think they are simple, there are 8 separate boards for my crossovers involving vacuum tube capacitors, russian micas etc.

My distortion figures are very very low, and my measured response is very linear, do not forget, I have five channels sharing the load so not much to do for any of my drivers

Parallel as in conjugates ...?