Is it me or is there a real line array revival going? Way more PM's than usual and line array threads popping up left and right... 
For everyone with plans: plan your build carefully. Right down till the last detail. Check those impedance curves and make them as smooth as a baby's bottom. Plan your room with the arrays in mind. Get rid of all early reflections. Use FIR filtering (with lots of tabs). And last but not least: Have a lot of fun!
For everyone with plans: plan your build carefully. Right down till the last detail. Check those impedance curves and make them as smooth as a baby's bottom. Plan your room with the arrays in mind. Get rid of all early reflections. Use FIR filtering (with lots of tabs). And last but not least: Have a lot of fun!
I think this thread is responsible for a lot of it. Also, for me personally, having built a bunch of "traditional" designs, line arrays present something new and interesting. Certainly, there are the benefits of line arrays in general... especially with the ability for me to build them into the wall. But also the FIR filtering is a new frontier.
I think so! This thread, the Stupid Cheap Line array and RA7's corner array convinced me it was worth a try.
I saw you mention a wave shape along the length of the baffle and that one of the only things you would do different would be to break up the back baffle in your sandwich into separate pieces.
Can you explain what your thinking is with this as I really don't get it.
I would think the double peak in your impedance had more to do with the standing wave associated with the height of your tower as the internal chambers aren't completely sealed and ~140Hz waves are pretty big.
As mine will be sealed in 5 separate sections it will be interesting to see if the same thing happens
I think so! This thread, the Stupid Cheap Line array and RA7's corner array convinced me it was worth a try.
I saw you mention a wave shape along the length of the baffle and that one of the only things you would do different would be to break up the back baffle in your sandwich into separate pieces.
Can you explain what your thinking is with this as I really don't get it.
I would think the double peak in your impedance had more to do with the standing wave associated with the height of your tower as the internal chambers aren't completely sealed and ~140Hz waves are pretty big.
As mine will be sealed in 5 separate sections it will be interesting to see if the same thing happens
There are a couple of reasons for me to know it's the baffle.
First - there is no straight shot trough the enclosure. Although there are holes connecting the chambers they are not in line top to bottom:
The holes are either on the side or front to back but no straight path exists. As seen in the template as well.
(plus the fact it's filled with fiberglass top to bottom)
Second - Putting O-rings under the bolts that hold down the baffle had a large effect. They broke up the double peak in the impedance plots.
This test was with rings cut from an old bicycle's inner tube. These were replaced with O-rings later on. The differences are clear though.
Third - Removing my MLV + neoprene layer from in-between the baffles and replacing it with butyl rope further improved these results. If it hadn't been the baffle that caused it, the impedance would have been the same.
The plot used for the final impedance correction.
(the correction part for the rising impedance towards higher frequencies was removed later on, leaving only the correction for the resonance bump)
There was an unlucky summing of the baffle length and the drivers own resonance making that wiggle there.
You can still see a deviation but no clear bump anymore. I've always planned to use butyl rope between the back baffle and the enclosure as well but never got around to that part, too much work involved. Your plots may look way better for starters (I hope so) with a one part enclosure, but this remedy made my enclosure itself dead silent in subsequent tests. No vibrations felt or heard even with the ear to the enclosure.
Though the baffle length was responsible for a time anomaly in the plots. That part wouldn't be fixed by changing baffle construction. It was visible in Halair's shorter arrays at a different frequency and you will "bump" into it also, I'm sure. That's a separate issue.
Last edited:
Thanks for the reply and that is all good information, the o-ring tests certainly support the notion that the baffle had a large part to play. The stuffing and non straight path would do much to dampen the length wave resonance maybe to the point that it is undetectable.
I think you are right that it is unfortunate that the length of the line coincides closely with the resonant frequency of the enclosure with the drivers in it. Not much can be done about that given ceiling sizes, driver sizes and the wiring needed to get a decent impedance.
I hope I get good results too but I doubt that they will be cleaner than yours given the effort you went to get there.
I still don't understand the acoustic theory behind why the length of the baffle would be the dominant factor. I see much about internal enclosure shape and size and position on the baffle but nothing much about the length creating issues. Maybe it's because line arrays are not the norm and therefore the issue is less common.
I see why you think the length is the issue but I still don't understand the 'why' of it.
It will be interesting to compare results given the different constructions people have been using to get to a similar place
Another thought that keeps bugging me. ra7's array is less in volume than both of our free standing arrays. I do see similar behaviour in his measurements, at a higher frequency. See this post.
Another reason to get that impedance curve as clean as you can. That's what was causing the phase wiggle if you ask me. Solvable though in DSP as my experiments have shown. Good to know as I will encounter this with the subs as well. Only of interest to freaks like me that want a time coherent system.
I've gone back and forth between blaming it on this or that, in the end it probably is the impedance peak of the complete system, after which the frequency response will normally drop off. In an ideal world with ideal enclosures one would get straight phase after straightening out the frequency curve. I've always had a small wiggle left. Even after removing all early reflections (as far as possible). Changing my DRC-FIR template to fix it separately solved the problem. An ideal driver/enclosure probably wouldn't need that fix. So it is baffle and summing drivers related. The total system.
Last edited:
Are you guys talking about the slightly flat top of the main resonant frequency peak in the impedance plot? Could it arise because each driver has a slightly different peak and shape, and combining it in series and parallel leads to slight flattening? I wouldn't worry about it too much. There are other fish to fry.
That's what I thought initially. But it isn't. It's "something" that resonates slightly out of sync with the drivers. Not the summing of the drivers. I've checked it out thoroughly long ago.
That prior discussion started here. A first sim was done by xrk971. And my research led to my baffle. A later update to the baffle fixed it even further. As seen a few posts ago.
What fish?
That prior discussion started here. A first sim was done by xrk971. And my research led to my baffle. A later update to the baffle fixed it even further. As seen a few posts ago.
What fish?
Last edited:
LOLHad been using this little TC9FD as a cutting and drilling guide, since it wasn't working. Looks like I discovered the reason why.
Sorry for the hiatus in updates on the subwoofer project. I've been busy getting the summer wheels for my car in shape. Due to lack of funds at the time I bought fake Fuchs a while ago (a complete set for the price of one genuine Fuchs wheel). Machined aluminium with clear coat is a bad idea. So right now I'm polishing the lips (by hand of course, you guys know me) having been so successful in polishing the baffles of the speakers. Being cast wheels instead of forged makes them not suitable for anodising. Powder coating is too big of a risqué due to the temperatures needed.
They (the wheels) will require more service afterwards but once the clear coat goes they look like #### anyway. My promised time line still stands, I'm hoping to start building at the end of this month/beginning of the next.
Off to the garage for more sanding...
They (the wheels) will require more service afterwards but once the clear coat goes they look like #### anyway. My promised time line still stands, I'm hoping to start building at the end of this month/beginning of the next.
Off to the garage for more sanding...
Last edited:
Someone asked me to do a tutorial on the polishing of the baffles so I shot a few pictures this afternoon.
Basically it's just a lot of work. It starts with sanding, then on to more sanding and after that: you've guessed it, even more sanding.
Depending on the starting roughness we start with pretty rough paper, I usually use dry sanding at this stage. From there on you move to a finer grid of sandpaper to sand away the scratches left by the previous round.
For these wheels I started with 60 grit to remove the clear coat, after that moving up to 150 grit dry. Then on to wet sanding with ~320 grit.
(These are pretty big steps, for the baffles I used 80 - 120 - 180 and then moving to wet sanding with 320 - 400 - 600)
You can skip the 80 if the metal is smooth enough.
This sanding process determines the amount of shine the end result will be able to get. So it makes sense to use smaller steps for a mirror shine.
Here's the wheel after the final round of wet sanding with 320:
I could have gone to 400 and then 600 but I don't care if they are polished to a mirror glaze. I'm even expecting to tone them down a bit after polishing.
Next I take my hand drill and start to machine polish. I use 2 solid polishing compounds and a couple of separate buffing cotton disks and other shapes for each step.
The polishing disks come in a variety too, some are meant for the pré polish stage and some for the final polish.
Anyway, to give an idea what I used:
After the first round with pré polish:
(wiped clean with mineral spirits for this shot)
Then on to the final finish:
Wiped down with some metal polish and a paper towel. Cleaned up with water and buffed with a microfiber towel and paper towel.
A comparison to a yet unfinished wheel:
And another:
Can't stress enough, the end result will depend on the sanding. This wheel was done casually in two afternoons. The baffles got way more time from me. All of it spend on sanding to get that mirror finish.
Hope this helps
Basically it's just a lot of work. It starts with sanding, then on to more sanding and after that: you've guessed it, even more sanding.
Depending on the starting roughness we start with pretty rough paper, I usually use dry sanding at this stage. From there on you move to a finer grid of sandpaper to sand away the scratches left by the previous round.
For these wheels I started with 60 grit to remove the clear coat, after that moving up to 150 grit dry. Then on to wet sanding with ~320 grit.
(These are pretty big steps, for the baffles I used 80 - 120 - 180 and then moving to wet sanding with 320 - 400 - 600)
You can skip the 80 if the metal is smooth enough.
This sanding process determines the amount of shine the end result will be able to get. So it makes sense to use smaller steps for a mirror shine.
Here's the wheel after the final round of wet sanding with 320:
I could have gone to 400 and then 600 but I don't care if they are polished to a mirror glaze. I'm even expecting to tone them down a bit after polishing.
Next I take my hand drill and start to machine polish. I use 2 solid polishing compounds and a couple of separate buffing cotton disks and other shapes for each step.
The polishing disks come in a variety too, some are meant for the pré polish stage and some for the final polish.
Anyway, to give an idea what I used:
After the first round with pré polish:
(wiped clean with mineral spirits for this shot)
Then on to the final finish:
Wiped down with some metal polish and a paper towel. Cleaned up with water and buffed with a microfiber towel and paper towel.
A comparison to a yet unfinished wheel:
And another:
Can't stress enough, the end result will depend on the sanding. This wheel was done casually in two afternoons. The baffles got way more time from me. All of it spend on sanding to get that mirror finish.
Hope this helps
Last edited:
I do recall having shown or shared some pictures somewhere before in this long thread but I'll show a few, it's a 1982 911 SC 3.0 (European model):
It really isn't as pretty as the pictures might suggest. It's my daily driver and has been for the last 11+ years. So it's a little rough around the edges but in good condition where it matters most. Top picture shows the winter wheels.
If something is up with it, my garage will not look that clean, it will resemble this picture:
Actually, my garage has never looked that clean ever again after taking the top picture. You might recognise the environment from the Array build pictures. The Car also has a full stereo setup which is covered here.
I can only afford to drive a car like that because I do all the work (maintenance and repairs) on it myself. Prices have skyrocketed in the last couple of years even for this model range. I love the car, but I also love the fact that they are dependable and well build, capable of being driven day in day out. On a 35 year old car some parts do break. Yet I've managed to use it as my main form of transportation. Which means I have the usual weekends every once in a while where I'm under the car fixing something to be able to get to work the following week. It has become a way of life...
It really isn't as pretty as the pictures might suggest. It's my daily driver and has been for the last 11+ years. So it's a little rough around the edges but in good condition where it matters most. Top picture shows the winter wheels
.If something is up with it, my garage will not look that clean, it will resemble this picture:
Actually, my garage has never looked that clean ever again after taking the top picture. You might recognise the environment from the Array build pictures. The Car also has a full stereo setup which is covered here.
I can only afford to drive a car like that because I do all the work (maintenance and repairs) on it myself. Prices have skyrocketed in the last couple of years even for this model range. I love the car, but I also love the fact that they are dependable and well build, capable of being driven day in day out. On a 35 year old car some parts do break. Yet I've managed to use it as my main form of transportation. Which means I have the usual weekends every once in a while where I'm under the car fixing something to be able to get to work the following week. It has become a way of life...