Looking at the narrow vertical and wide horizontal dispersion characteristics of ribbon tweeters, I don't really see how one would properly integrate these types of tweeters into a system.
The only way I see it could be done is to cross very high with a low order XO to a small midrange. This way the horizontal dispersion between the two drivers is matched (more or less), and we get a graceful transition to the narrow vertical dispersion of the ribbon due to the wide overlap by the low order XO.
But then again, due to the typical height of these tweeters, you can't really get the center-to-center spacing close enough to minimize lobing.
Am I missing something here?
The only way I see it could be done is to cross very high with a low order XO to a small midrange. This way the horizontal dispersion between the two drivers is matched (more or less), and we get a graceful transition to the narrow vertical dispersion of the ribbon due to the wide overlap by the low order XO.
But then again, due to the typical height of these tweeters, you can't really get the center-to-center spacing close enough to minimize lobing.
Am I missing something here?
A thing that further complicates matching the dispersion is that planar tweeters usually cannot be crossed low.
This seems to be a good example where both horizontal and vertical dispersion are good: https://www.diyaudio.com/forums/mul...vity-horbach-keele-filters-3.html#post5501424
This seems to be a good example where both horizontal and vertical dispersion are good: https://www.diyaudio.com/forums/mul...vity-horbach-keele-filters-3.html#post5501424
Last edited:
Crossing low takes care of the issues and IMO in critical listening positions the vertical dispersion is also a non issue so long as the ribbon is less than about 100 mm long.
Yes If you can hear well above about 12k AND you are convinced you have to have flat response out their when you stand up from listening seat then I suppose there is some concern but I suspect even this one is over blown by most.
So yes that leaves us with the issue typical in small ribbons and planers, ya cant go low enough with the designs presently offered.
I know from experience that once you overcome the low crossover issue the other "issues" dont have as much weight as they are often given.
My present fave is a 10" Al cone crossed to a 16mm by 75mm ribbon at 1 khz.
Yes If you can hear well above about 12k AND you are convinced you have to have flat response out their when you stand up from listening seat then I suppose there is some concern but I suspect even this one is over blown by most.
So yes that leaves us with the issue typical in small ribbons and planers, ya cant go low enough with the designs presently offered.
I know from experience that once you overcome the low crossover issue the other "issues" dont have as much weight as they are often given.
My present fave is a 10" Al cone crossed to a 16mm by 75mm ribbon at 1 khz.
Last edited:
Crossing low means a steep filter, which means a sharp transition in vertical dispersion.
Might not be as audible as a sharp transition in both horizontal and vertical dispersion, but it's nevertheless an issue that can easily be solved by using a regular dome.
Mine are crossed with 12db/oct LR filters.
Even with steep filters, at 1khz the present ribbon and planer offerings on the market will disintegrate and there distortion profiles are not good below about 2 Khz
are there really small domes that can do 1khz well to say 110 db peaks with a 12 db/oct cross ? And how does there horizontal dispersion up around 10K compare with a 16mm wide ribbon?
Even with steep filters, at 1khz the present ribbon and planer offerings on the market will disintegrate and there distortion profiles are not good below about 2 Khz
are there really small domes that can do 1khz well to say 110 db peaks with a 12 db/oct cross ? And how does there horizontal dispersion up around 10K compare with a 16mm wide ribbon?
Last edited:
If your goal is equal vertical dispersion, your choices are really use a line array of mids next to it, or a single mid-woofer which is never going to have the same vertical dispersion.
Part of the sound quality is the narrow vertical dispersion. It reduces floor and ceiling reflections, enhancing clarity at the listening location.Within normal listening angles, this doesn't really matter very much. I mean, most listening areas are wide, but narrow.
Part of the sound quality is the narrow vertical dispersion. It reduces floor and ceiling reflections, enhancing clarity at the listening location.Within normal listening angles, this doesn't really matter very much. I mean, most listening areas are wide, but narrow.
You seem to contradict yourself here? I.e saying that ribbons shouldn't be crossed below 2 Khz, but you cross yours at 1 Khz?
And crossing low with a 12db/oct filter, you will end up with a much steeper acoustical slope (due to the natural roll off of the tweeter that low in freq) which you will then need to match with steep high pass filter on the woofer. Which in turn means a sudden change in power response at XO.
By my understanding you generally want to match directivity at XO, or use a
shallow slope to get a graceful transition in directivity. Otherwise the tweeter will be saying "Here I am!"
In other words, they wont integrate well.I definitely see the advantages of the narrow vertical dispersion of a ribbon tweeter. But I just don't see how they can be integrated well in a system.
Hi Defo,
I don't think this needed to match identically in H and V spaces. The better question is, at your target listening locations, like, a couch, do you have similar FR with gross transitions? Or can you move from one end to the other, lie down, sit up, and hear a similar output?
The issue of energy in the room is compensated for vertically by the extra horizontal dispersion.
I use a large Mundorf AMT with a 6.5" midwoofer, and there's' nothing wrong with the sound.
I don't think this needed to match identically in H and V spaces. The better question is, at your target listening locations, like, a couch, do you have similar FR with gross transitions? Or can you move from one end to the other, lie down, sit up, and hear a similar output?
The issue of energy in the room is compensated for vertically by the extra horizontal dispersion.
I use a large Mundorf AMT with a 6.5" midwoofer, and there's' nothing wrong with the sound.
"You seem to contradict yourself here? I.e saying that ribbons shouldn't be crossed below 2 Khz, but you cross yours at 1 Khz? "
sorry for confusion, what Im saying is the small ribbons on the market at the moment have sharply rising distortion profiles below 2 Khz. This is unavoidable with their present designs.
"And crossing low with a 12db/oct filter, you will end up with a much steeper acoustical slope (due to the natural roll off of the tweeter that low in freq) which you will then need to match with steep high pass filter on the woofer. Which in turn means a sudden change in power response at XO."
fs on my ribbons is around 70Hz and a low Q at that. There are no issues with a crossover at 1khz.
Be careful on assumptions. My responses are not to cause trouble, they are simply to see If I myself have missed something. I am looking to learn
sorry for confusion, what Im saying is the small ribbons on the market at the moment have sharply rising distortion profiles below 2 Khz. This is unavoidable with their present designs.
"And crossing low with a 12db/oct filter, you will end up with a much steeper acoustical slope (due to the natural roll off of the tweeter that low in freq) which you will then need to match with steep high pass filter on the woofer. Which in turn means a sudden change in power response at XO."
fs on my ribbons is around 70Hz and a low Q at that. There are no issues with a crossover at 1khz.
Be careful on assumptions. My responses are not to cause trouble, they are simply to see If I myself have missed something. I am looking to learn
Last edited:
Here's a solution: Use multiple smaller drivers in an array to make a shape similar to that of your ribbon, but scaled up. Do that in multiple stages and make a fractal array. I did, and I love it.
Fractal Array Straight CBT with Passive XOs and no EQ
I am using Heils and they sound great using an active 24db L/R analog crossover and a passive network for response shaping. They integrate very well with a pair of 12 woofers crossed at 800Hz. They are tilted up a bit to keep my seated ears in the narrow verticle slot. As they are dipoles though you get enough rear splash to not make the narrow vertical as critical.
Rob
Ribbons for me are the absolute best solution for nearfield critical listening or monitoring when they can be crossed high and with a relatively benign slope.......they offer a window into detail that domes and CD simply can’t match. But out in the room at casual listening distance and in common listening rooms, the dynamic limitations leave the performance wanting. My favorite is the Fountek Neo 2.0.....the longer length opened the vertical plane and adds some air to the room
While I agree that the direct sound at all listening positions and the sound power can turn out to be OK, my experience is that it still sounds 'off' if you try to compensate a narrow vertical dispersion at high frequencies (as compared to the dispersion of the midrange) with a wide horizontal dispersion. I have this experience with a speaker with a 4" mid and a Dayton PT2C-8 planar tweeter (radiating area approx. 12 x 2 cm). It was like the highs were too loud and not loud enough at the same time. Swapping the tweeter for a 3/4" dome made the speaker sound correct.
I have two hypothetical explanations. The first is that the sound power was not OK. This is backed up by the observation that, when not standing in front of the speaker, the sound contained too little high frequency energy. The second is that the first reflections contain either too much or close to zero high frequency energy, depending on which reflection is considered.
Last edited:
Not sure what that is but I have heard something similar at times. Years of testing/ development have lead me in this direction...
Some ribbons have a difficult rising/ falling response thing to deal with. If you want flat response all the way out then you have to deal with the hump with a filter. The problem is so many of the ribbons on market do not have a nice smooth symmetrical hump which means it can be very difficult to build a filter that counters it closely. And ribbons can be quite revealing and therefore unforgiving of freq response issues. As well even if you get it smoothed out there are often stored energy issues that can produce off axis irregularities.
I have found that once I had a diaphragm design where the stored energy problems were overcome (and the off axis response problems associated), and the natural rise fall response was smooth and symmetrical on both sides of the peak, that it became easy to build a trap circuit that very closely counters the rise/ fall ( within +- 1 db), Once these two issues were overcome the "highs both too loud and too soft" thing was gone.
BTW trap circuits often get a bad rap and early on in my own development work I agreed. However once I got a well behaved diaphragm with a smooth symmetrical rise / fall and no sharp discontinuities I no longer hear an issue. I suspect the problems people are hearing with traps is really more to do with the driver not being smooth enough to match the trap to a high level.
Also I find that ribbons longer than about 80 mm are very critical of vertical alignment and getting both speakers aligned well is often no small matter.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.