Hi All,
I've been umming and ahhhing about building a line array and am trying to pick drivers. Hardest thing to do when you are on a budget. Long story short I wanted to use a vertical line tweeter that did not involve too many drivers and that was not cost prohibited.
A line of NEO3-PDR gets expensive. Line of Fountek large ribbons gets expensive. Domes require too many holes. Anything more exotic than the above and it gets expensive.
Anyways I have a pair of AMTPRO-4 and they are ok. The FR can be equalized as I'll be running active but the horizontal dispersion leaves something to be desired when compared to the NEO3-PDR and Fountek NEO X 3.0.
I guess I should outline my design goals for this project.
1. Wide horizontal dispersion or else not sure a LA is worth it?
2. Crossover close to half wavelength of center to center spacing of drivers as per Jim Griffin papers.
3. Open baffle loose option. If this does not work out will add walls to enclosure and go sealed.
4. Cut front baffle only once. I do not want to do many baffles. Too much work lol.
5. Sound the same most places in the room so probably 1 above is more important to me.
To satisfy the above requirements these will most likely be 3 way line arrays. I wanted to start open baffle and if that does not work, finish the enclosure for sealed alignment. Right now I'm running a horn down to 550hz and open baffle under that with some damping. I don't like the rear wave of the open baffle sound until it gets below 300 hz as I find it a bit strange sounding. Sealed is ok but definitely loads the room different. I don't like the boom boom. This leaves problems regarding radiation pattern but that an exercise for another day. Onto the real topic.
While costing and looking at crossover points and horizontal dispersion of many many drivers and keeping open baffle a loose option then it seems like 3-way is inevitable. Running NEO8-PDR all the way up will probably not work for me as it sounds ok but since going through all the work, might as well dive full in. Also the horizontal dispersion is a bit lacking on the higher registers so perhaps there will be more seat to seat variation?
The first candidate is AMTPRO-4 as it is one of the only reasonably priced options for me that satisfies most of the above objectives but that horizontal radiation.........could use some help.
I took off the faceplate to the AMTPRO-4 and cut a new faceplate that had a smaller opening. Used foam board roughly 2mm thick which I picked up from the craft store. The original opening was roughly 40mm, the new opening is roughly 24mm. Attached are some rudimentary off axis measurements. The measurements were meant to be quick and dirty to see if there was a reasonable improvement in off axis above 7khz. Note: Angles and levels are eyeballed but shows a definite trend. Its a bit hard to read but........
Seems like the top octaves are reduced in overall level by what seems like a shelving filter effect. But the top octaves show an improvement in off axis drop off. Not as good as the NEOCD 1.0 but closer to acceptable. Notice the modified faceplate readings were rotated to more extreme angles than the original. I wanted to make sure I was seeing a proper trend. In fact the last reading could have been more off axis. Mic was about 1 meter away and levels were reasonably loud. Distortion did not change with faceplate opening size. Will have to optimize opening size if I'm to continue.
Question: If I make new faceplates for the AMTPRO-4, barring sensitivity, do you guys think a line array with the AMTPRO-4 is worth it based on the attached measurements? The price difference from the NEO3-PDR and Fountek X 3.0 is significant for me. Also the 3.0 is only front radiating. Are there better horizontal dispersion options for the roughly the same cost/performance that I am missing?
Targeting a crossover frequency of around 1200hz to probably a line of NEO-8PDR then to line of Peerless 830668 but I'm open to better options within this price range. Speaker is a little wide though.
P.S. I'll probably have the new faceplates 3D printed but if that costs to much I'll make them out of wood. It's only time right
.
I've been umming and ahhhing about building a line array and am trying to pick drivers. Hardest thing to do when you are on a budget. Long story short I wanted to use a vertical line tweeter that did not involve too many drivers and that was not cost prohibited.
A line of NEO3-PDR gets expensive. Line of Fountek large ribbons gets expensive. Domes require too many holes. Anything more exotic than the above and it gets expensive.
Anyways I have a pair of AMTPRO-4 and they are ok. The FR can be equalized as I'll be running active but the horizontal dispersion leaves something to be desired when compared to the NEO3-PDR and Fountek NEO X 3.0.
I guess I should outline my design goals for this project.
1. Wide horizontal dispersion or else not sure a LA is worth it?
2. Crossover close to half wavelength of center to center spacing of drivers as per Jim Griffin papers.
3. Open baffle loose option. If this does not work out will add walls to enclosure and go sealed.
4. Cut front baffle only once. I do not want to do many baffles. Too much work lol.
5. Sound the same most places in the room so probably 1 above is more important to me.
To satisfy the above requirements these will most likely be 3 way line arrays. I wanted to start open baffle and if that does not work, finish the enclosure for sealed alignment. Right now I'm running a horn down to 550hz and open baffle under that with some damping. I don't like the rear wave of the open baffle sound until it gets below 300 hz as I find it a bit strange sounding. Sealed is ok but definitely loads the room different. I don't like the boom boom. This leaves problems regarding radiation pattern but that an exercise for another day. Onto the real topic.
While costing and looking at crossover points and horizontal dispersion of many many drivers and keeping open baffle a loose option then it seems like 3-way is inevitable. Running NEO8-PDR all the way up will probably not work for me as it sounds ok but since going through all the work, might as well dive full in. Also the horizontal dispersion is a bit lacking on the higher registers so perhaps there will be more seat to seat variation?
The first candidate is AMTPRO-4 as it is one of the only reasonably priced options for me that satisfies most of the above objectives but that horizontal radiation.........could use some help.
I took off the faceplate to the AMTPRO-4 and cut a new faceplate that had a smaller opening. Used foam board roughly 2mm thick which I picked up from the craft store. The original opening was roughly 40mm, the new opening is roughly 24mm. Attached are some rudimentary off axis measurements. The measurements were meant to be quick and dirty to see if there was a reasonable improvement in off axis above 7khz. Note: Angles and levels are eyeballed but shows a definite trend. Its a bit hard to read but........
Seems like the top octaves are reduced in overall level by what seems like a shelving filter effect. But the top octaves show an improvement in off axis drop off. Not as good as the NEOCD 1.0 but closer to acceptable. Notice the modified faceplate readings were rotated to more extreme angles than the original. I wanted to make sure I was seeing a proper trend. In fact the last reading could have been more off axis. Mic was about 1 meter away and levels were reasonably loud. Distortion did not change with faceplate opening size. Will have to optimize opening size if I'm to continue.
Question: If I make new faceplates for the AMTPRO-4, barring sensitivity, do you guys think a line array with the AMTPRO-4 is worth it based on the attached measurements? The price difference from the NEO3-PDR and Fountek X 3.0 is significant for me. Also the 3.0 is only front radiating. Are there better horizontal dispersion options for the roughly the same cost/performance that I am missing?
Targeting a crossover frequency of around 1200hz to probably a line of NEO-8PDR then to line of Peerless 830668 but I'm open to better options within this price range. Speaker is a little wide though.
P.S. I'll probably have the new faceplates 3D printed but if that costs to much I'll make them out of wood. It's only time right
.Attachments
Take a look at the Aura NSW1 . Its a 1" full range driver that can be placed at a 40 mm pitch. I haven't heard them but they look good on paper. Use enough of them - I know that means using too many drivers but that is what it takes to do it right - and they will play low enough to cross directly to your woofer line in a 2 way config.
Directivity is simply a function of the radiator's shape and dimensions. The AMTPRO-4 is going to have very narrow directivity above 10khz because of that.
There's a few ways to address this:
1) Mask off part of the diaphragm with foam or felt, like BG does.
2) Build a narrow waveguide
3) both
Check out my threads on planar and ribbon phase plugs
There's a few ways to address this:
1) Mask off part of the diaphragm with foam or felt, like BG does.
2) Build a narrow waveguide
3) both
Check out my threads on planar and ribbon phase plugs
..if you want
a long line, and
dipole, and
good sound, and
keep the cost within reason, then
-perhaps a DIY Electrostat line tweeter.
(..a DIY ribbon is also possible, but it may be more difficult to construct/assemble. I'm guessing the price for the magnets on a ribbon vs. the transformer and power supply for the ESL will be comparable, but I could be wrong.)
a long line, and
dipole, and
good sound, and
keep the cost within reason, then
-perhaps a DIY Electrostat line tweeter.
(..a DIY ribbon is also possible, but it may be more difficult to construct/assemble. I'm guessing the price for the magnets on a ribbon vs. the transformer and power supply for the ESL will be comparable, but I could be wrong.)
Last edited:
Thanks nc535 I'll check it out but that is a lot of holes!
Thanks Patrick. In my too long original post I created a test faceplate that seemed to improve horizontal dispersion. Will have to play with the opening size but yes reduced horizontal opening works a bit better in the upper range. Hope to not loose too much sensitivity up there though. Listening to it now but still on the fence. Need to do more testing. Waveguide is out of the question for this round. Will check out your phase plug threads!
Thanks Patrick. In my too long original post I created a test faceplate that seemed to improve horizontal dispersion. Will have to play with the opening size but yes reduced horizontal opening works a bit better in the upper range. Hope to not loose too much sensitivity up there though. Listening to it now but still on the fence. Need to do more testing. Waveguide is out of the question for this round. Will check out your phase plug threads!
..if you want
a long line, and
dipole, and
good sound, and
keep the cost within reason, then
-perhaps a DIY Electrostat line tweeter.
(..a DIY ribbon is also possible, but it may be more difficult to construct/assemble. I'm guessing the price for the magnets on a ribbon vs. the transformer and power supply for the ESL will be comparable, but I could be wrong.)
haha lets add not too long a a design phase to the list. I'm not ready to tackle that kind of R&D
. Do you have a design for one of those suggested above?I would take a look at some of the DIY arrays in these forums that feature a line of only 3 or 4 inch full range type driver that is equalized via DSP. Or the new line array kit at Parts Express. I know some will scoff at the full range concept, but I have hear the Russell IDS-25's and thought they were terrific. IDS-25 stereo column speaker array
haha lets add not too long a a design phase to the list. I'm not ready to tackle that kind of R&D
..I'd probably contact this guy to see if the idea is feasible at a low cost (and if it is - try and get his help):
Jazzman's DIY Electrostatic Loudspeaker Page
-this forum also has a planar section.
Ah the jazzman! I've been through most of his site before, pretty great! It's definitely more time than I can afford right now. A lot of material stuffs!
The easiest way to 'visualize' what's going on is to understand what causes the narrowing in directivity:
pathlength
For instance, if one side of the diaphragm is one meter from your microphone, and the other side of the diaphragm is 1.01 meters from your microphone, you're going to get a dip at 17khz. This is because 34khz is 0.01 meters long, and when the pathlength difference is half a wavelength, you get a dip. (Due the 180 degree difference in phase between the two pathlengths.)
Hope that makes sense.
An externally hosted image should be here but it was not working when we last tested it.
In the NEO3 PDR, all they're doing is cramming some felt into the side of the ribbon. The felt rolls off the highs. This means that only the center is radiating high frequencies.
Masking off the diaphragm with tape can work too, I've done that. The nice thing about felt or open cell foam is that it's an acoustic low pass filter.
Why do you want an acoustic low pass filter you might ask?
Well that gives you the best of both worlds:
1) it gives you the extended highs of the ribbon
2) along with the extended low frequencies of a larger diaphragm
Felt or open cell foam basically makes the ribbon behave like a crude coaxial.
The diypages here have been great for information! I need to be more active in commenting but I lurk well
. I've read through a number of very long threads, great stuffs. I'm bent on realizing a multiple unit rectangular (or 1 long vertical element) tweeter array outlined in the Jim Griffin bible and hope to check off most of the goals outlined in post 1. The other options seem difficult to implement.
I wish I could hear some of those fullranger arrays. I'm sure they sound great. Lots of tech involved!
The Dayton PT2C planar worked well for my project. I haven't specifically measured for off axis response but by ear, the off axis response is very good even with changes in height and distance.
I did a build thread here but unfortunately most of the pics are on photobucket and they have changed the way they do things.
Rob's OBLA-33 (Open Baffle Line Array) project. | Audiokarma Home Audio Stereo Discussion Forums
Based on my own experience, I strongly suggest you build them nearly floor to ceiling to reap all of the possible benefits of a true line array.
Video here: YouTube
I did a build thread here but unfortunately most of the pics are on photobucket and they have changed the way they do things.
Rob's OBLA-33 (Open Baffle Line Array) project. | Audiokarma Home Audio Stereo Discussion Forums
Based on my own experience, I strongly suggest you build them nearly floor to ceiling to reap all of the possible benefits of a true line array.
Video here: YouTube
Hi Rob41,
Thanks for the links! Will check them out. Yes I'm planning on 75 to 85 inches in height. That is the driver array height. Speaker will be a bit slightly taller because of the required room for the base and baffle.
That thread is epic. Nice work! Will take some time to get through properly.
The Dayton PT2C looks like it has better off axis than the AMTPRO-4, hrmm..... I'm thinking of dividing the work up and splitting the LA into 2 "modules" per side. Getting scared at the amount of work.
The more I listen my test set up I think I want to stay with sealed till about 125 to 175 hz. That makes a tweeter line and mid woofer line in a sealed enclosure. Then flanked by the OB LA. Also if the 2-way works out then no need for the OB LA!
Getting closer to driver selection based on Post 1 guidelines
AMPTPRO-4 (modified) and 6FE100.
What is the lowest crossover for the PT2C in a LA? Do you think going lower than 2k would be doable? Hard to maintain CTC spacing at that crossover.
Thanks,
The Dayton PT2C looks like it has better off axis than the AMTPRO-4, hrmm..... I'm thinking of dividing the work up and splitting the LA into 2 "modules" per side. Getting scared at the amount of work.
The more I listen my test set up I think I want to stay with sealed till about 125 to 175 hz. That makes a tweeter line and mid woofer line in a sealed enclosure. Then flanked by the OB LA. Also if the 2-way works out then no need for the OB LA!
Getting closer to driver selection based on Post 1 guidelines
AMPTPRO-4 (modified) and 6FE100.
What is the lowest crossover for the PT2C in a LA? Do you think going lower than 2k would be doable? Hard to maintain CTC spacing at that crossover.
Thanks,
Awesome. No floor or ceiling first order reflections and those surfaces will actually enhance the ability of the arrays.
Something that was pointed out to me by an iconic speaker designer was to flank the tweeters by the mid-woofers on both sides, rather than just along one side.
There will be lobing created by mounting them along only one side and while the lobing will still be present in a flanked design, it will be symmetrical which is a far better scenario.
Have fun with your project!
Thanks.
I'm crossed at 2700Hz because I'm using a fourth order crossover point. They recommend 3k or above and I sometimes use it there as well and you can get away with a much shallower cross at that point. Much will depend on the drivers you're crossing down to. I've used them down to 2300Hz but you're not going to get any lower and I'd recommend not going lower than 2500Hz.
Thanks for the information! I'd like to get a bit lower crossover so I think I'll stick with the AMT Pro. Too bad as the dispersion on those drivers are nice.
I think the easiest way to deal with the AMT is to use Patricks idea. Take off the faceplate, black tape the opening to the desired width, re-install the faceplate. The faceplate is aluminium. I was toying of the idea of getting custom faceplates done to reduce the center to center spacing both horizontally and vertically but that is cost prohibitive.
I'm currently testing the 5FE120 and have the 6FE100 on order. One of my primary goals (that did not make it onto the list in Post 1) is.......good listening at low volume level. The 5FE120 does not seem to be that lively at low listening level. Hoping the 6FE100 is a bit better. As a last resort I might go with a line of Alpair 10.3 crossed over the AMT. The 10.3 is on the high side of driver cost but I don't want the "I should have used xxx drivers instead" once all is finished. I'm not after absolute SPL.
The 6FE100 arrived! Set them up with the AMTPRO-4 for testing.
I'm not getting the subjective results that I wanted. Put up against a TW29RN and PM180 monitor, there is a gap in subjective sound quality. I know the radiation pattern of the line will be different and will load the room differently so it's not a fair test; however, for the entry price I have a number of other projects I'd rather do.
Perhaps the long game is the way to to go. I will start collecting a number of PM180 for another project and when I get tired of that project I'll requisition them (PM180) for the line array main event.
I'm envious of you line array guys! Maybe someday soon I'll get to hear a nice pair!
I'm not getting the subjective results that I wanted. Put up against a TW29RN and PM180 monitor, there is a gap in subjective sound quality. I know the radiation pattern of the line will be different and will load the room differently so it's not a fair test; however, for the entry price I have a number of other projects I'd rather do.
Perhaps the long game is the way to to go. I will start collecting a number of PM180 for another project and when I get tired of that project I'll requisition them (PM180) for the line array main event.
I'm envious of you line array guys! Maybe someday soon I'll get to hear a nice pair!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.