First trials of active horn speaker

[morricab]

Since you are losing 14db have you tried the opposite approach of rolling off your lower end 600 and up? This way you bleed off the difference with minimal HF boost. That is the worst place to add power to the driver as it is already putting out higher thd there from the controlled resonanses and break up in the last octave or so. This will simulate what your passive crossover will do. You can use the voltage drive from the DSP as a reference to design and build your crossover.

Rob

Changed since then. Now digital is at zero and balance the gain at the amps through their volume controls.
I can try you idea, seems like a good one and easy to do. However, the distortion at my levels will be very low (it sounds VERY clean) anyway and not so much info up there...but every bit counts.

 

[Folsom]

Can you do me a favor and play with moving the timing back on the tweeter in a few increments, so that it's "acoustic center" is behind the woofer at a few varying degrees? And note a general perception of sound? It wouldn't surprise me if you ended up liking the sound when they are not aligned, as I suspect not only is the alignment far from crucial like everyone is obsessed with, but even detrimental to some point due to some other factors. I'm saying we hear other things that aren't the actual alignment, giving rise to the ridiculous mania around alignment that doesn't seem to be able to prove itself. Yet, the horn is another factor so I'm not sure how it'll play out.

 

[morricab]

Can you do me a favor and play with moving the timing back on the tweeter in a few increments, so that it's "acoustic center" is behind the woofer at a few varying degrees? And note a general perception of sound? It wouldn't surprise me if you ended up liking the sound when they are not aligned, as I suspect not only is the alignment far from crucial like everyone is obsessed with, but even detrimental to some point due to some other factors. I'm saying we hear other things that aren't the actual alignment, giving rise to the ridiculous mania around alignment that doesn't seem to be able to prove itself. Yet, the horn is another factor so I'm not sure how it'll play out.

Up until now I have not taken any special care for time alignment. The mid/tweeter is already behind the woofer by more than it should be for strict time alignment (the horn is relatively deep and the throat is well behind). I can adjust the time alignment digitally anyway so the horn placement in terms of depth is uncritical once I decide to do some alignment. I have positioned the drivers currently so that I get the distance between the two drivers as close together vertically without the horn drooping too much or (god forbid) sliding off the cabinet (unlikely given the driver weight) as they are just sitting freely on the cabinets. I want to minimize lobbing to have an even coverage pattern if possible. I will eventually replace the HDTs with a pair of Supravox Alizeés that have the mid/woofer much closer to the top of the cabinet so I am hoping for an even better blend with that driver (and it is 3db more sensitive...).

 

[Folsom]

So, they are already sounding good with the tweeter behind, as I suspect might be the case. But again I do ask if possible to try different alignments and give some subjective feedback; it's good for all. Time alignment is marketing foo foo, but that doesn't mean there's no effect on the sound by moving drivers so that their patterns change what you hear.

 

[morricab]

So, they are already sounding good with the tweeter behind, as I suspect might be the case. But again I do ask if possible to try different alignments and give some subjective feedback; it's good for all. Time alignment is marketing foo foo, but that doesn't mean there's no effect on the sound by moving drivers so that their patterns change what you hear.

With digital time delay, I can experiment with time alignment WITHOUT changing the radiation pattern or at least the physical positioning of the drivers.

 

[kodomo]

Can you do me a favor and play with moving the timing back on the tweeter in a few increments, so that it's "acoustic center" is behind the woofer at a few varying degrees? And note a general perception of sound? It wouldn't surprise me if you ended up liking the sound when they are not aligned, as I suspect not only is the alignment far from crucial like everyone is obsessed with, but even detrimental to some point due to some other factors. I'm saying we hear other things that aren't the actual alignment, giving rise to the ridiculous mania around alignment that doesn't seem to be able to prove itself. Yet, the horn is another factor so I'm not sure how it'll play out.

-Disclaimer: I am writing these below considering tweeter response over 8-10khz-

Well, I have tried this extensively. Tweeters behind, tweeters forward etc. It does make a difference and you can measure it very easily. It works together with the crossover so can not be thought without considering crossovers. Just move them behind and measure/listen to the response at the crossover region of treble and mids or upper mids (whatever you are crossing to), there will be dips or rises that were not there. This also happens when you move. Perfect treble response is very much a single location phenomena, at least that is what I hear, measure and experience.

I can not understand why people consider alignment a marketing phenomena. My system sounds much better when all the channels are aligned. Maybe what I call alignment is a different thing?

 

[morricab]

-Disclaimer: I am writing these below considering tweeter response over 8-10khz-

Well, I have tried this extensively. Tweeters behind, tweeters forward etc. It does make a difference and you can measure it very easily. It works together with the crossover so can not be thought without considering crossovers. Just move them behind and measure/listen to the response at the crossover region of treble and mids or upper mids (whatever you are crossing to), there will be dips or rises that were not there. This also happens when you move. Perfect treble response is very much a single location phenomena, at least that is what I hear, measure and experience.

I can not understand why people consider alignment a marketing phenomena. My system sounds much better when all the channels are aligned. Maybe what I call alignment is a different thing?

I too generally prefer the sound time-aligned. Coming from full-range planars (that were inherently this way) I came to highly value coherence as an important trait for realism. This is where a lot of large horns have troubles and why many big horn demonstrations have left me scratching my head (but not all...the best ones are WOW!).

 

[Folsom]

I did mean digitally, btw, Brad.

FR is a powerful thing, if changing your (kodomo's) tweeter position changed the FR that's all the difference you'd need to hear it.

Speakers that are "time aligned" may sound better, but it's been proven many times over the human ear can't detect physical shifts in proximity of a wavelength, by PHD's and such. That is so long as things like the FR does not shift, because then they're hearing that - or different reflections, etc etc.

Coherence is very important for me, too. That's actually why I want someone that's playing with a horn tweeter to play around with timing, just for some more feedback. It may be the position of the horn like that ultimate trumps any movement you use digitally. IME so far, getting tweeters and woofers to meld has had more to do with physical distances than any alignment electrically. I say this because non-aligned speakers sometime meld really well, but rarely with the tweeter in front.

 

[DaveC]

-Disclaimer: I am writing these below considering tweeter response over 8-10khz-

Well, I have tried this extensively. Tweeters behind, tweeters forward etc. It does make a difference and you can measure it very easily. It works together with the crossover so can not be thought without considering crossovers. Just move them behind and measure/listen to the response at the crossover region of treble and mids or upper mids (whatever you are crossing to), there will be dips or rises that were not there. This also happens when you move. Perfect treble response is very much a single location phenomena, at least that is what I hear, measure and experience.

I can not understand why people consider alignment a marketing phenomena. My system sounds much better when all the channels are aligned. Maybe what I call alignment is a different thing?

My results show it's critical within a certain frequency range. At high frequencies it matters but unless you listen with head-in-a-vice it doesn't matter as much. With lows it matter less as well, but within the critical range it's very easy to measure the interference created by overlapping drivers that aren't aligned.

With a large multi-way horn system like yours alignment is critically important imo, at least within the middle few horns, but it's also possible in other types of speakers for a wideband midrange to completely cover the range of frequencies where it matters. This is how I designed my speaker, a single wideband semi-horn loaded midrange driver, a lot like AER in a front horn but I use a smaller driver so I get significant gain via the horn as well, it's also a mechanical equalization device, a mechanical crossover at the low end, as well as controlling directivity.

 

[morricab]

Phase II has no begun with the replacement of the HDTs with Supravox Alizee speakers. These are single driver 99db speakers the use the Alnico magnet based Supravox 215–2000 full range driver. Immediately I notice better bass and integration, more refinement and better focus. Really cool.

 

[morricab]

Best part is I am getting these results without even tinkering with the crossover yet...it was just a drop in replacement!

 

[Tinear1]

Interesting... I too have been using a miniDSP-HD for integration of a horn with a how frequency system but went a slightly different direction. I was concerned that the miniDSP would dirty up the signal to my horn full range drivers (Oris-150's using the Voxativ AC-2.6 drivers) so I am running them direct from my SET 300B monoblocks and using the miniDSP as a low pass filter and output matching to the base drivers. Frankly I planned to use the DSP to define the xover filters needed and then make physical xovers but the miniDSP works so well that I didn't bother taking it out. I may re-config this weekend and try your solution. Thanks for the info.

 

[morricab]

Interesting... I too have been using a miniDSP-HD for integration of a horn with a how frequency system but went a slightly different direction. I was concerned that the miniDSP would dirty up the signal to my horn full range drivers (Oris-150's using the Voxativ AC-2.6 drivers) so I am running them direct from my SET 300B monoblocks and using the miniDSP as a low pass filter and output matching to the base drivers. Frankly I planned to use the DSP to define the xover filters needed and then make physical xovers but the miniDSP works so well that I didn't bother taking it out. I may re-config this weekend and try your solution. Thanks for the info. View attachment 56469

Cool system! I look forward to hearing about your ongoing experiments.

 

[morricab]

Update:

Now with the Supravox Alizees underpinning the design I reoptimized the bass by basically removing the cut at 90hz I had used to tame the HDTs. Then I readjusted the horn upwards to 1000hz xover as this made the blend between the drivers seamless (at 700hz the horn was a bit spotlit) and fleshing put instruments tonally. With HDTs the added resolution running down to 700hz was welcome but the Supravox driver is more resolving and allows the higher cross. It is now in the range where i can also experiment with my other horn/driver combo (18 sound XT120 horn and Beyma CP350ti CD). I will experiment now with 900 and 800 hz to see if it is worth dipping down again below 1khz.

The sound is approaching ideal from about 80hz and up and the only wish now is to get the last couple octaves to this level.

 

[Audiophile Bill]

Update:

Now with the Supravox Alizees underpinning the design I reoptimized the bass by basically removing the cut at 90hz I had used to tame the HDTs. Then I readjusted the horn upwards to 1000hz xover as this made the blend between the drivers seamless (at 700hz the horn was a bit spotlit) and fleshing put instruments tonally. With HDTs the added resolution running down to 700hz was welcome but the Supravox driver is more resolving and allows the higher cross. It is now in the range where i can also experiment with my other horn/driver combo (18 sound XT120 horn and Beyma CP350ti CD). I will experiment now with 900 and 800 hz to see if it is worth dipping down again below 1khz.

The sound is approaching ideal from about 80hz and up and the only wish now is to get the last couple octaves to this level.

Hi Brad,

Nice update - very pleased that the full range Supravox is working out by providing that increased insight. My experience in the last year led me to similar conclusions albeit with different wide banders.

Did you try a larger horn on top with a lower cut off yet?

 

[morricab]

Hi Brad,

Nice update - very pleased that the full range Supravox is working out by providing that increased insight. My experience in the last year led me to similar conclusions albeit with different wide banders.

Did you try a larger horn on top with a lower cut off yet?

I haven’t tried a larger horn with a lower cutoff because the Beyma driver is good only down to about 600hz. I tried 600, 700, 750 and 800 previously and liked 700 the best with the HDTs. I felt it wasn’t as seamless with Supravox until I went to around 1khz then it feel into place. I might try other slopes and filter types to see if one works better than LR type. Maybe a 1st order around 2khz?

 

[Audiophile Bill]

I haven’t tried a larger horn with a lower cutoff because the Beyma driver is good only down to about 600hz. I tried 600, 700, 750 and 800 previously and liked 700 the best with the HDTs. I felt it wasn’t as seamless with Supravox until I went to around 1khz then it feel into place. I might try other slopes and filter types to see if one works better than LR type. Maybe a 1st order around 2khz?

I think it’ll be a suck it and see thing unfortunately. Keep us posted. There is something attractive about a 1st order if it works. Seems to allow more life.

 

[Folsom]

1st order with DSP? Why??????

 

[Audiophile Bill]

1st order with DSP? Why??????

I thought he was finding the optimum crossover with mini dsp and then making in analogue - well that is what he said a few pages back!!!!

 

[morricab]

1st order with DSP? Why??????

Sometimes with disparate driver types it blends better if there is a lot of overlap rather than an abrupt transition between drivers. Also, I will probably make an analog filter eventually...although the way this is working now it might not be necessary.