tallclyde said:
Clyde, thank you too for turning me on to them. I'mnot sure what happened between building the prototype with 4 woofers and the final product with 12 woofers, but the final array sounds significantly stronger in the bass, and I only have one running at this point.
Chipco, yes, .. a couple of corner loaded subs are in the pipeline, once I'm done with this project (mainly speaker measurement and crossover related from this point on, though I may need to build a center channel too, depending on how the pair perform in 4.1 mode). Need to build a mic. preamp and impedance jig first.
I gotta tell you... line arrays are the coolest looking speakers I've ever seen... I'm really interested to see how this all pans out for you and what your crossover design looks like.
I too would like to work on a project like this, but I would like to use them in an HT setup similar to yours. Since you aren't tri-amping, what kind of setup are you using to drive these monsters?
I too would like to work on a project like this, but I would like to use them in an HT setup similar to yours. Since you aren't tri-amping, what kind of setup are you using to drive these monsters?
I'm going to use ARTA to measure impedance and frequency response on these things. These instructions look fairly dummy proof (hint, .. hint)
http://soundhobby.com/measurementsystem.htm . From my experiments with the 4th order (24db/octave) active crossover, there doesn't seem to be much difference crossing over low at 3KHz and higher at 4-5Khz.
I plan to build a measurement mic using a panasonic WM-61A capsule. I tweaked my little sonic imact amp over the weekend to eliminate the bass roll-off so I don't have to lug the heavy marantz around.
This weekend, I found a mic preamp I built a while ago based on one of the simple circuits at http://www.epanorama.net/circuits/microphone_powering.html .
My 7 year old IBM thinkpad died of old age over the weekend
, so I'll have to use the HTPC for the measurements.
I'll have to see if the mic input on the soundblaster live sound card in the PC can provide +5V to power the electret (see article for details)
Blagosaurous, I intend to use a passive two way 3rd order crossover on this project. Below 100 - 150 Hz (TBD), I plan to split - feed the LFE to dual subwoofers ie. dual mono bass . My HT is driven by a panasonic SA-XR10. Since I don't plan to drive more than a handful of watts into these arrays, I'm not going to bother upgrading it .
I thought a lot about doing stereo bass, but the additional complexity didn't seem worth it for an A/V application.
While I get ready to measure these lines, I also plan to start playing with damping the rear, bracing etc etc.
Stay tuned, and apologies for the painful detail in this rambling thread.
http://soundhobby.com/measurementsystem.htm . From my experiments with the 4th order (24db/octave) active crossover, there doesn't seem to be much difference crossing over low at 3KHz and higher at 4-5Khz.
I plan to build a measurement mic using a panasonic WM-61A capsule. I tweaked my little sonic imact amp over the weekend to eliminate the bass roll-off so I don't have to lug the heavy marantz around.
This weekend, I found a mic preamp I built a while ago based on one of the simple circuits at http://www.epanorama.net/circuits/microphone_powering.html .
My 7 year old IBM thinkpad died of old age over the weekend
, so I'll have to use the HTPC for the measurements.I'll have to see if the mic input on the soundblaster live sound card in the PC can provide +5V to power the electret (see article for details)
Blagosaurous, I intend to use a passive two way 3rd order crossover on this project. Below 100 - 150 Hz (TBD), I plan to split - feed the LFE to dual subwoofers ie. dual mono bass . My HT is driven by a panasonic SA-XR10. Since I don't plan to drive more than a handful of watts into these arrays, I'm not going to bother upgrading it .
I thought a lot about doing stereo bass, but the additional complexity didn't seem worth it for an A/V application.
While I get ready to measure these lines, I also plan to start playing with damping the rear, bracing etc etc.
Stay tuned, and apologies for the painful detail in this rambling thread.
Currently fighting with ARTA impedance measurements @ http://www.diyaudio.com/forums/showthread.php?s=&threadid=103243 
Project is looking good! Im working on my own line array, and thinking of using the fosters and the apex jrs also, are you glad you choose these drivers so far? Also, wht are the O.D. measurements on the mids, cant seem to find the measurements on there website, keep up the good work, looking forward to the end result.
http://www.hometheatershack.com/forums/diy-speakers/4960-ideas-front-mains.html
http://www.hometheatershack.com/forums/diy-speakers/4960-ideas-front-mains.html
Mongrel714 said:
What's OD ?. If you mean outer diameter, .. I mentioned this a pages earlier ie. between 133 and 135 mm.
I bought some parts to build the "official" ARTA jig this afternoon. Let's see what happens.
Re. the choice of drivers, I'm quite happy with the fosters so far FOR THE APPLICATION I choose ie. open baffle A/V speaker. For a true high-fi application, I would look for a different midrange. Ditto if I were looking at a sealed or ported application (these fosters really want to be free air mounted). You also haven't mentioned a budget.
As for the tweeters, they're okay but wiring those many of them was a real job. They don't seem to extend as I high as I'd ideally want but I'll reserve final judgment on this when I'm done with the project (or at least the tweeter measurement). So don't take that as a negative. In any case, with the benefit of hindsight, I can suggest that if you decide to use a conventional dome tweeter line, try and leave enough space so you can retrofit planars / ribbons at a later point if you want.
Last week, I braced the wings and used some polyfill on the sides and back of the cabinets to
1. increase cabinet rigidity.
2. kill spurious mid/high frequency reflections off the sides and back walls while letting the bass through. This has improved the intelligibility of voices etc ..
I still have some more to add but here's a preview at
An externally hosted image should be here but it was not working when we last tested it.
I've also been given some foam reflectors which I think could work well behind these arrays.
The fosters are running full range for now until I sort out my ARTA jig issues (pre-requisite to FR / impedance measurements needed for crossover development). They can get real loud without any sense of discomfort and are fun to crank up. They won't even break a sweat hitting 95 - 100dB. I can't wait till I get some movies running through this setup.
Still stuck with ARTA !!. Apologies. I finally got the impedance curves to look like they should, .. but the absolute values are wrong Browse over to my ARTA thread if you want to feel my pain
About the tweeter, a couple of things
1. I feel the 8 ohm impedance rating is misleading, ... it should really be 6. something ohm ( planet10's measurement at http://www.diyaudio.com/forums/attachment.php?s=&postid=1177687&stamp=1175916108 ) As a result, my tweeter line impedance is 4. something opposed to the 6 ohm woofer line. I think I should add 1-2 more series groups of tweeters (currently 5 series elements, each element consisting of 7 drivers in parallel) to equalize the impedance. I hate retrofits at this stage, but it should be okay if I'm careful about it. Now you understand why I absolutely need to get the impedance measurements working.
2. sound quality.I haven't played with the tweeter line for some while now, after I've stuffed the rear of the cabinet (running the woofers full range to break them in). In any case, I feel they may be rolling off a little early (maybe some comb filtering roll off too, can't say for sure), .. not a bad thing, .. but just don't expect tons of air &detail up there. When I get this running, and after I add some high freq. boost (as suggested by magnetar), Ill make my final verdict on the tweeters. If I still need more high freq. at that point, I have these radioshack supertweeters lying around that go out to more than 30KHz..... Rodent control devices, anyone ?
In anycase, they're dirt cheap and easy to work with. Price out an equivalent line with $4 dayton neo's and compare the price difference.
Browse over to my ARTA thread if you want to feel my pain About the tweeter, a couple of things
1. I feel the 8 ohm impedance rating is misleading, ... it should really be 6. something ohm ( planet10's measurement at http://www.diyaudio.com/forums/attachment.php?s=&postid=1177687&stamp=1175916108 ) As a result, my tweeter line impedance is 4. something opposed to the 6 ohm woofer line. I think I should add 1-2 more series groups of tweeters (currently 5 series elements, each element consisting of 7 drivers in parallel) to equalize the impedance. I hate retrofits at this stage, but it should be okay if I'm careful about it. Now you understand why I absolutely need to get the impedance measurements working.
2. sound quality.I haven't played with the tweeter line for some while now, after I've stuffed the rear of the cabinet (running the woofers full range to break them in). In any case, I feel they may be rolling off a little early (maybe some comb filtering roll off too, can't say for sure), .. not a bad thing, .. but just don't expect tons of air &detail up there. When I get this running, and after I add some high freq. boost (as suggested by magnetar), Ill make my final verdict on the tweeters. If I still need more high freq. at that point, I have these radioshack supertweeters lying around that go out to more than 30KHz..... Rodent control devices, anyone ?
In anycase, they're dirt cheap and easy to work with. Price out an equivalent line with $4 dayton neo's and compare the price difference.
While ARTA consumes my life, I've decided to go by planet10's impedance measurements on these tweeters (see first page of this log) which indicate that these tweeters are closer to 6 ohm (as opposed to 8 ohms). In any case, they measure like that with a multimeter across the terminals.
To match the impedance of the woofer line, I estimate that I'll need some more tweeters (7 series groups, each consisting of 7 paralleled tweeters totalling 49 tweeters per enclosure) ie. 14 more tweeters per side.
28 more holes per enclosure, 28 more tweeters to wire .... headbash
. I'll survive. Luckily, I have plenty of tweeters left.
To match the impedance of the woofer line, I estimate that I'll need some more tweeters (7 series groups, each consisting of 7 paralleled tweeters totalling 49 tweeters per enclosure) ie. 14 more tweeters per side.
28 more holes per enclosure, 28 more tweeters to wire .... headbash
. I'll survive. Luckily, I have plenty of tweeters left.
Got somewhere with ARTA. Really cool tool, once it's working
After trying various amplifiers, I finally used the sound card output to drive the drivers. I'll elaborate later, in my ARTA thread. For today, here are the impedance plots for the raw drivers.
apex jr tweeter
compare with planet10's measurements here http://www.diyaudio.com/forums/attachment.php?postid=1177687&stamp=1175916108&
My curve rises sharper above 10KHz, but otherwise, the curve are similar.
foster 0504 wideband woofer
According to these measurements, the woofer impedance is somewhat higher than the nominal 8 ohm at my intended crossover point (say 4 KHz).
Here's an impedance plot of the 12 driver foster array
Do I design for 6 ohm or 12 ohm (at 4 KHz) ????
After trying various amplifiers, I finally used the sound card output to drive the drivers. I'll elaborate later, in my ARTA thread. For today, here are the impedance plots for the raw drivers.
apex jr tweeter
An externally hosted image should be here but it was not working when we last tested it.
compare with planet10's measurements here http://www.diyaudio.com/forums/attachment.php?postid=1177687&stamp=1175916108&
My curve rises sharper above 10KHz, but otherwise, the curve are similar.
foster 0504 wideband woofer
An externally hosted image should be here but it was not working when we last tested it.
According to these measurements, the woofer impedance is somewhat higher than the nominal 8 ohm at my intended crossover point (say 4 KHz).
Here's an impedance plot of the 12 driver foster array
An externally hosted image should be here but it was not working when we last tested it.
Do I design for 6 ohm or 12 ohm (at 4 KHz) ????
I've been advised to first measure the frequency response of the lines before attempting a custom crossover using, say, XOverSim ( http://www.geocities.com/f4ier/newxoversim.htm )
To this end, I've ordered a few panasonic electret capsules to build electret measuring mics with.
Over the weekend, I increased the number of tweeters per side to 49 (7 x 7). Need to wire them in.
To this end, I've ordered a few panasonic electret capsules to build electret measuring mics with.
Over the weekend, I increased the number of tweeters per side to 49 (7 x 7). Need to wire them in.
Well,
here's my first attempt at FR measurements using ARTA. I haven't calibrated the software yet, but the graphs demonstrate the general trends. I used a panasonic mic capsule and a simple mic preamp . The mic is at the listener position.
Now, I need to build a crossover. Looks like I'll have to
1.notch the peak around 600 - 700 Hz on the woofer line, though it might be worth trying & measuring a makeshift zobel first
2. maybe think about boosting the treble response.
here's my first attempt at FR measurements using ARTA. I haven't calibrated the software yet, but the graphs demonstrate the general trends. I used a panasonic mic capsule and a simple mic preamp . The mic is at the listener position.
Now, I need to build a crossover. Looks like I'll have to
1.notch the peak around 600 - 700 Hz on the woofer line, though it might be worth trying & measuring a makeshift zobel first
2. maybe think about boosting the treble response.
An externally hosted image should be here but it was not working when we last tested it.
.... here's a good link explaing the hows and whys of crossover design http://sound.westhost.com/lr-passive.htm
Here's the woofer line with a crude makeshift 8 ohm/10uF zobel, .. notice how the impedance flattens out
In other news, I played around with various slopes and crossovers in foobar 2000 / dsp crossover and came to the conclusion that 4.5 - 5 KHz /3rd order is a good crossover point, to my ears. Need to order parts.
An externally hosted image should be here but it was not working when we last tested it.
In other news, I played around with various slopes and crossovers in foobar 2000 / dsp crossover and came to the conclusion that 4.5 - 5 KHz /3rd order is a good crossover point, to my ears. Need to order parts.
Last week, I used foobar 2000's DSP filter plugin to test various filter types / x-over points and slopes . There didn't seem to be much difference (if any) between a 4KHz and 5 KHz point, so I picked 5KHz as the higher x-over point allows the tweeters to play louder with less distortion. Both the tweeter and woofer lines are around 6.2 to 6.5 ohm nominal, so I chose 6.5 ohm for the crossover calculations .
I've ordered parts for a 5KHz / 3rd order filter and a zobel ciruit for the woofer array . Madisound seemed to have somewhat lower prices than PE. The total parts cost was around $60 + shipping.
Rather than use a passive filter to tame the mid-bass / mid-range hump, I plan to use the eq. in foobar 2000 for this purpose, if I eventually find it an issue. The tweeter impedance peak is fairly benign so hopefully I won't need to suppress this peak via the crossover. Yet another reason to cross over a bit higher and further away from Fs.
Comparing the ARTA tweeter impedance curves for the two cabinets, .. one was about an ohm above the theoretical value. Looking further, I found and fixed a loose solder joint. Props to this great tool.
Here's my parts list
I've ordered parts for a 5KHz / 3rd order filter and a zobel ciruit for the woofer array . Madisound seemed to have somewhat lower prices than PE. The total parts cost was around $60 + shipping.
Rather than use a passive filter to tame the mid-bass / mid-range hump, I plan to use the eq. in foobar 2000 for this purpose, if I eventually find it an issue. The tweeter impedance peak is fairly benign so hopefully I won't need to suppress this peak via the crossover. Yet another reason to cross over a bit higher and further away from Fs.
Comparing the ARTA tweeter impedance curves for the two cabinets, .. one was about an ohm above the theoretical value. Looking further, I found and fixed a loose solder joint. Props to this great tool.
Here's my parts list
Code:
Products
------------------------------------------------------
2 x 5 mfd Solen Fast Cap 400V () = $5.60
2 x 6.8 mfd Solen Fast Cap 400V () = $6.60
2 x 10 mfd Solen Fast Cap 400V () = $9.00
2 x G-POSTBL Binding Post Black () = $2.90
2 x YT, Set of 4 Tiptoe () = $4.80
2 x G-POSTRL Binding Post Red () = $2.90
2 x 8 ohm - 15 watt wirewound resistor () = $0.90
2 x 3.3 mfd Solen Fast Cap 400V () = $4.50
2 x Sidewinder 0.1mH 16AWG () = $7.10
2 x Sidewinder 0.33mH 16AWG () = $10.20
2 x Sidewinder 0.15mH 16AWG () = $7.30
------------------------------------------------------
Sub-Total: $61.80Here's the response with the crossover, pretty good, I'd say.
An externally hosted image should be here but it was not working when we last tested it.
Effect of reversing tweeter , ... I think I'm going to keep the tweeter reversed, to avoid the cancellation dip.
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.