I'm struggling with power tapering.
In Dr. Griffin's white paper, it says: "Experience has shown that a slight power taper (less than a 2 to 1 ratio) will improve the sound (i.e., reduced sound bloom) with minimal impact to the overall line array operation."
What does that refer to? Is is that the power to a driver at the end of the array needs to be less than twice the power to a driver at the centre of the array? Or that as you move from centre towards the end the step changes in power need to be less than 2 to 1?
I searched but couldn't find anything and I don't really have access to academic publications like those referenced in the paper.
In Dr. Griffin's white paper, it says: "Experience has shown that a slight power taper (less than a 2 to 1 ratio) will improve the sound (i.e., reduced sound bloom) with minimal impact to the overall line array operation."
What does that refer to? Is is that the power to a driver at the end of the array needs to be less than twice the power to a driver at the centre of the array? Or that as you move from centre towards the end the step changes in power need to be less than 2 to 1?
I searched but couldn't find anything and I don't really have access to academic publications like those referenced in the paper.
I come down on the side of minimal taper under most conditions. The tradeoff is to maintain line array sound without reducing the sound to point source scope.
Tapering can help better adapt line array sound to the room in my experience. For example consider normal ceiling height rooms (example would be an 8 feet high ceiling and a 6 feet high array). My experience is that a small power taper will be advantageous under those conditions. But if you have a floor to ceiling array, I would not taper. Nor does it make much sense to taper a line which is appreciably shorter than the ceiling height.
Often another reason to taper is attain a more reasonable speaker impedance.
Bottom line: My original point it that a small amount of taper can be useful when appropriately used.
Tapering can help better adapt line array sound to the room in my experience. For example consider normal ceiling height rooms (example would be an 8 feet high ceiling and a 6 feet high array). My experience is that a small power taper will be advantageous under those conditions. But if you have a floor to ceiling array, I would not taper. Nor does it make much sense to taper a line which is appreciably shorter than the ceiling height.
Often another reason to taper is attain a more reasonable speaker impedance.
Bottom line: My original point it that a small amount of taper can be useful when appropriately used.
Last edited:
Many thanks for your detailed reply. Your reply does raise a few more questions.
I was thinking along the line, no pun intended, of a large number of small 3" drivers, arranged into an 8 ft speaker with an 8.5 ft ceiling height. I would prefer to NOT taper, as I'm into tubes so I can easily make an OTL to drive the speaker directly if all drivers were in series.
However, I would not want to compromise speaker performance for this reason alone, and so would use power tapering if you recommend it.
I have a lot of flexibility in driver quantiy. Does having an even number or an odd number of drivers fit ino the power tapering issue?
I was thinking along the line, no pun intended, of a large number of small 3" drivers, arranged into an 8 ft speaker with an 8.5 ft ceiling height. I would prefer to NOT taper, as I'm into tubes so I can easily make an OTL to drive the speaker directly if all drivers were in series.
However, I would not want to compromise speaker performance for this reason alone, and so would use power tapering if you recommend it.
I have a lot of flexibility in driver quantiy. Does having an even number or an odd number of drivers fit ino the power tapering issue?
From what I have read, for a line that goes from ceiling to floor, there is no real need to do any power tapering because the vertical dispersion of the array is tightly controlled through the range anyways. In situations where the line is only a fraction of the height of the room, power tapering is a good idea to help offset the effects of gradually tightening higher frequency vertical dispersion- that couples into greater output at those higher frequencies than at lower frequencies (I'm guess that is what Griffin is talking about when he says "sound bloom.")
As long as you are dabbling in it, I found Don Keele's Keynote on CBT line arrays very interesting. You can find it on youtube. He employs a similar power tapering combined with an arc shaped line.
------
My personal opinion.. Your implementation might play friendly with a compression horn in the center of the line to help eliminate combing effects at higher frequency. For 3" drivers, I'd cross at 2K with the widest dispersion waveguide or horn I could find. The sensitivity of ~30 x 3" drivers would probably actually be very similar to many available compression drivers.
As long as you are dabbling in it, I found Don Keele's Keynote on CBT line arrays very interesting. You can find it on youtube. He employs a similar power tapering combined with an arc shaped line.
------
My personal opinion.. Your implementation might play friendly with a compression horn in the center of the line to help eliminate combing effects at higher frequency. For 3" drivers, I'd cross at 2K with the widest dispersion waveguide or horn I could find. The sensitivity of ~30 x 3" drivers would probably actually be very similar to many available compression drivers.
Last edited:
For leadbelly's case I would not taper as his line approaches floor to ceiling.
I'm not a big fan of a using a line of full range drivers because of comb line issues. You need to equalize both high and low frequency response and even after that you do not attain the sound quality of a proper two-way or three-way implementation.
I like a true ribbon tweeter in the center if you wish to have a simple two-way case. But either the compression horn or ribbon tweeter creates a situation where-in the sound fall-off of the tweeter is 6 dB per doubling of the distance vs 3 dB for the array. You can make it work for seated listeners roughly the same distance from the source.
Don Keele uses power shading to attentuate one end of his line to achieve a specific weighting. Power tapering does not attenuate the signal as it retains full power without loss of energy.
I'm not a big fan of a using a line of full range drivers because of comb line issues. You need to equalize both high and low frequency response and even after that you do not attain the sound quality of a proper two-way or three-way implementation.
I like a true ribbon tweeter in the center if you wish to have a simple two-way case. But either the compression horn or ribbon tweeter creates a situation where-in the sound fall-off of the tweeter is 6 dB per doubling of the distance vs 3 dB for the array. You can make it work for seated listeners roughly the same distance from the source.
Don Keele uses power shading to attentuate one end of his line to achieve a specific weighting. Power tapering does not attenuate the signal as it retains full power without loss of energy.
Last edited:
I am resurrecting this thread to test an idea that came to my mind when circling around different ways utilize bunch of Vifa TC9FD18-08 fullrangers. I read through Jim Griffin's paper on line arrays and power tapering. I was thinking of 12 Vifa's per speaker which gives the array the height of 1m (~3').
Symmetric power tapering described in the paper (2/4/6 or 2/3/3/4) requires (IMO) that the most efficient driver group (2) will be at ear height. Given that, 1m line array with symmetric power taper would either a) be lifted so that middle drivers would be approx. 0,9-1m height making the speaker ~1,5m high or b) left as 1m and listen to them lying on the sofa. Neither option is favorable for my application
So if i wanted to keep this speaker 1m long and have normal listening height, I was thinking could power taper be used asymmetricly? So that 2/4/6 taper is descending from top to bottom drivers. I assume (based on light theoretical background) this would lift the image up and help to address floor boundary effects of a short line.
Feel free to shoot this idea down
Symmetric power tapering described in the paper (2/4/6 or 2/3/3/4) requires (IMO) that the most efficient driver group (2) will be at ear height. Given that, 1m line array with symmetric power taper would either a) be lifted so that middle drivers would be approx. 0,9-1m height making the speaker ~1,5m high or b) left as 1m and listen to them lying on the sofa. Neither option is favorable for my application
So if i wanted to keep this speaker 1m long and have normal listening height, I was thinking could power taper be used asymmetricly? So that 2/4/6 taper is descending from top to bottom drivers. I assume (based on light theoretical background) this would lift the image up and help to address floor boundary effects of a short line.
Feel free to shoot this idea down
... So if i wanted to keep this speaker 1m long and have normal listening height, I was thinking could power taper be used asymmetricly? So that 2/4/6 taper is descending from top to bottom drivers. I assume (based on light theoretical background) this would lift the image up and help to address floor boundary effects of a short line.
Feel free to shoot this idea down
Yes, that has been used. For example Harrin-kaiutin (do not remember model name) but it was presented in Pekka Tuomela's book too.
There was only 4 mid range elements but the method was what you explaned.
I tested power tapering long time ago and it seems to be better than using capacitors to shorten line with frequency because that leads to phase problems and therfore some nasty issues in response depending on the distance. However the same idea to get line high depend on frequency can be done by power tappering.
My idea what I tested and it worked well was to limit low frequencies of the elements which has the highest power. In the practice the to add series capacitor for the 4 and 6 in 2/4/6 line.
Result is that on higher frequencies power tapering gives same effect for amplitude as using capacitors on parallel of the 4 and 6 but without phase difference. On the lower frequences the series capacitors limits the amplitude (and cone movement) so that every element have about same total cone movement.
Result wa very good but unfortunately in my prototype the radiation pattern matching between line and ribbon tweeter was not good enough and therefore I did not ocntinue with it.
I will have to look up the design in Tuomela's book. There are some issues that I might not be able to address in this setup, so for the time being this is just theoretical speculation. For starters, Vifa's have Fs on 125Hz, therefore this line needs to have woofer of some sort. the other is HF combing if I run them full range. Those will not be problems if I decide to address them, the question is do I want to
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.