At such aspect ratio, the box might act more like a MLTL. Try doing some search for that, quite a lot of discussions and examples already.
I asked if the port was near to the back - it looks long enough that the lining you've added may interfere, in which case, you would experience a loss of bass.
I think that SS 6,5" bass driver in big enclosure should give bass even without reflex port if room has some gain. I have had couple of standmounters and all of them gave 40-50 Hz audible (not room shaking) even with blocked reflex port. (I use speakers quite near wall, though)
Not enough BSC or wrong amplifier?
Not enough BSC or wrong amplifier?
When you play the 30hz tone, does the woofer move quite a bit? If the amp is filtering or the crossover is wrong, it probably won't move much. With that tone it should be moving visibly. It won't really make much sound, but it'll respond electrically and move the cone.
Hooking the woofers up full range won't hurt them at all. It's not going to sound very good, but the range of frequencies <200hz is what you need to check, which should be unaffected by removing the low pass filter.
Barring those problems, I'm going to say room response is what's happening. My main speakers will go down to 42-43hz without subs, but if I hook them up full range, because of their placement out in the room, they are very anemic. Before you change the crossover, you could always try wedging them in the corners of the room, flat up against the back and side walls and see what the bass sounds like.
I took out all the lining at the port region as per your advice
Okay, i managed to try out the speakers with another amp. The bass response is similar. I also tried them out full range without the XO. Same result. So i tried something else.
All this while i have been listening to them at quite a high height; the tweeter is about a head's length below my ears. So i sat on a smaller chair with the tweeter at eye level this time. And there is a lot more bass, the tonal balance being much better. Sounds like it might just be a room acoustic issue. I will try the speakers out in another room tomorrow (bigger room) to see how it sounds there.
There is another thing that bugs me though - why does the port output sound so... for lack of a better word, restrained? It feels like it wants to pump out more air but can't do so. I am really wondering if it's a result of the bracing being too close to the port, and i'm thinking of getting rid of the middle bracing.
But at the same time, i'm worried it'll make the cabinet ring much more
You can probably take this brace out, and perhaps move the top one a little closer to the woofer. You may also try widening the holes in the bracing a little bit, too.
It shouldn't affect it that much though. It might just sound a little thin in the bass if you've been listening to a speaker that's been designed with "the bass lump" as some are, like some of the ones you get from large chain stores that are made to sound thumpier than normal.
I'd probably have thinner stuffing or acousta-stuf in there if it was mine, too. Is that stuffing part of the Zaph design?
It shouldn't affect it that much though. It might just sound a little thin in the bass if you've been listening to a speaker that's been designed with "the bass lump" as some are, like some of the ones you get from large chain stores that are made to sound thumpier than normal.
I'd probably have thinner stuffing or acousta-stuf in there if it was mine, too. Is that stuffing part of the Zaph design?
Attachments
Last edited:
His original design recommends more use of stuffing. I toned it down a little since rockwool does a better job of killing standing waves than most generic foam. Thanks for the suggestions, will keep them in mind!
The baffle needs to be completely airtight, so why not try using a gasket before you clamp it to the enclosure?
the box is tuned too low, Zaph seems to prefer a low end that rolls off gradually. If you want a more solid low end (as I prefer) you need to tune the box so that it has a flatter (or even slightly rising) response, this will be at the expense of low end extension
^That could be true at least if room has major suck out at 35-60 Hz region. There's no use to design tuning to 25 Hz if you don't use speaker for lowest movie basses or if you are not afraid of cone excursion below tuning frequency. That is not an issues in any normal music listening setup.
i'm wondering why the need for more extension and slower roll off if the output at that range is not really meaningful..? After looking at the 18w 8531 datasheet, the roll off at the low end i'm experiencing seems to be normal. And taking into account the fletcher-munson curve, maybe what i'm hearing is perfectly normal. I do, however, prefer more punch/slam in the mid bass, which seems to be lacking a little now. I'm pretty sure it is related to the less BSC option i went with, so the midrange is quite loud relative to the bass.
But my concern is if i go with the original, full BSC option, the midrange will be too recessed and laid back. The easy way, of course, would be to try out different inductor values, but it will mean $$$. My L10 is at 2.0mH now. I'm wondering if going with a value between 2.0-2.7mH will be advisable? Also, does anyone know how much BSC is there with the original 2.7mH value? Zaph didn't state this in the website. i'm not too confident of my ability to measure inductance accurately even if i buy a inductance meter , so unwinding the coils to change the inductance just doesn't seem like a comfortable thing for me.
Also, if the bracing is not too near the port and the port is not too near the back wall, is there a reason why it seems like the port isn't functioning properly? It doesn't feel (amount of air pushed) and sound like the port is doing much, especially when i run a 30Hz test tone - which is the port's tuning frequency.
At the PE forum, someone suggested testing the tuning frequency with a multimeter, how can i do that?
A big thank you to everyone for all your help! 😀
But my concern is if i go with the original, full BSC option, the midrange will be too recessed and laid back. The easy way, of course, would be to try out different inductor values, but it will mean $$$. My L10 is at 2.0mH now. I'm wondering if going with a value between 2.0-2.7mH will be advisable? Also, does anyone know how much BSC is there with the original 2.7mH value? Zaph didn't state this in the website. i'm not too confident of my ability to measure inductance accurately even if i buy a inductance meter , so unwinding the coils to change the inductance just doesn't seem like a comfortable thing for me.
Also, if the bracing is not too near the port and the port is not too near the back wall, is there a reason why it seems like the port isn't functioning properly? It doesn't feel (amount of air pushed) and sound like the port is doing much, especially when i run a 30Hz test tone - which is the port's tuning frequency.
At the PE forum, someone suggested testing the tuning frequency with a multimeter, how can i do that?
A big thank you to everyone for all your help! 😀
Last edited:
Lower tuning moves group delay lower to region where it is not audible, gives louder 30-40 Hz (bass E) and better transient acting more like closed. And it doesn't boost room modes that easy versus peaking response speaker.
I don't know how your room acts but if sound is shouting in the midrange and lacks body of male voices or guitars I believe problem is more matter of too less BSC than wrong tuning frequency. So I would try 2,7 mH. Though, strange that 30-70 Hz isn't audible like you said.
Try to remove bracings (even all!) and stuffing in middle of enclosure that air can flow better. Leave stuffing in the bottom and behind driver. You can fine tune (bracing, stuffing) speaker after it works as it should.
Last edited:
Frankly, I don't understand why you don't try to measure the impedance of the speaker. It will probably teach you if there is a problem or not.
It is really easy to build a jig that does the job. You only need a PC with an audio card. Here is the link:
loudspeaker measurement jig
This page doesn't contain anymore the pictures. I'm attaching the only one you need.
Ralf
Attachments
Loren42's post #16 explains how.
Also, two people have pointed out that the box must be airtight. You can't expect it to be airtight with a few clamps unless you use a gasket, as the baffle will flex and leak when the woofer is driven.
i'm wondering why the need for more extension and slower roll off if the output at that range is not really meaningful..? After looking at the 18w 8531 datasheet, the roll off at the low end i'm experiencing seems to be normal. And taking into account the fletcher-munson curve, maybe what i'm hearing is perfectly normal. I do, however, prefer more punch/slam in the mid bass, which seems to be lacking a little now. I'm pretty sure it is related to the less BSC option i went with, so the midrange is quite loud relative to the bass.
But my concern is if i go with the original, full BSC option, the midrange will be too recessed and laid back. The easy way, of course, would be to try out different inductor values, but it will mean $$$. My L10 is at 2.0mH now. I'm wondering if going with a value between 2.0-2.7mH will be advisable? Also, does anyone know how much BSC is there with the original 2.7mH value? Zaph didn't state this in the website. i'm not too confident of my ability to measure inductance accurately even if i buy a inductance meter , so unwinding the coils to change the inductance just doesn't seem like a comfortable thing for me.
Also, if the bracing is not too near the port and the port is not too near the back wall, is there a reason why it seems like the port isn't functioning properly? It doesn't feel (amount of air pushed) and sound like the port is doing much, especially when i run a 30Hz test tone - which is the port's tuning frequency.
At the PE forum, someone suggested testing the tuning frequency with a multimeter, how can i do that?
A big thank you to everyone for all your help! 😀
We will try to help you again.
This image, taken from, the website : Measuring Loudspeaker Driver Parameters , shows how to make the measurement for impedance. You really don't need the amplifier if your generator has any balls at all.
An externally hosted image should be here but it was not working when we last tested it.
You can download free programs on the web that act as signal generators and use your PC's sound card to generate the test signal. This gets your amp out of the loop, which may be the real cause of the problem. There are also free programs that allow you to read in the data into the PC instead of a voltmeter.
The resistor can be about 10 Ohms and is available at Radio Shack.
If your amp is set to SMALL you will not get any bass. I would borrow a known good amp to test your speakers with.
You need to take the time to learn how to measure your speaker. Building a speaker requires tools like saws, drills etc. Measuring what you built is the only way to confirm you built what you intended! It is foolish not to do this and measurement is simply another tool that you must have to complete the job.
Programs like Holm Impulse do a remarkable job. You will need some kind of USB sound card and a mic and a stand. Either borrow or buy these things and get the job done.
There are books that will help you learn how to test and there are many tutorials on the web as well.
It is darn near impossible for anyone to help you here without data to determine the problem. Your ears don't count so please bring us some data so we can help you.
At the least, borrow a known working amp to test your speaker for bass to eliminate the amp as a problem.
Actually, if you read my earlier post, i did test the woofer out with my friend's amp and the bass response is similar.
I will try and measure the impedance tomorrow. Just to be sure i'm doing it properly, do i bypass the XO and wire the woofer directly? Or it doesn't matter? And assuming there really is a problem with the tuning frequency, what's the next step?
It won't matter. What you after is where the enclosure is tuned, which is not impacted by the crossover.
That assumes that the crossover does not have a problem. However, even if it does, I think you can still get meaningful data from your efforts.
You can automate the whole process with a program from Liberty Instruments called Praxis. It will run in Demo mode for free and perform the impedance test.
Liberty Instruments, Inc. Home Page
You need a sound card that is capable of 16-bit, but just about any card will do that.
I think ARTA is another program that will let you do the same thing in demo mode, but you can't save files.
You will probably need that resistor as described in previous posts.
The computer automated programs will give you a fast and reliable way to determine exactly where your port (it really is the enclosure) is tuned.
If the tuning is too high or too low you can fix it by changing the port length. However, before you do that you should do some work to understand the problem first.
Last edited:
and still ignored
and like said by others
tuning very low does not give more bass
especially not with small woofers
it does go low, but rolled off very early
could be an advantage with big woofers
but with small woofers, it easily leads to thin bass
may be called correct and 'tight' by some, and boring to the rest of us
compared to this, I bet you could just as well put in a much smaller closed box
you could try to put small bags of sand in the bottom
actually works quite well
A test tone of 30Hz with a speaker tuned (Fb) to 30Hz is going to have very little driver excursion which is the nature of a vented design, but the port should be working. See the excursion graph with about 10W input.
Having used the 18W8531 on several designs, 38 litres is too large and the Fb is too low. Have a look at Troel's designs and comments on the 18W8531 driver. Also have a look at his article on vent tuning where port length is usually 0.7 x calculated to achieve the target Fb.
For a test, try a port that's tuned to 40Hz to see if it makes a significant difference.
I've found 28 litres or smaller vented with this driver works the best and Fb > 35Hz for personal taste.
Try running the woofer without the crossover but add a 1mH inductor in series. This will flatten the response to 1.5k after which it will roll off smoothly. Even though it will not give BSC, it will give an indication of the raw woofer's performance in the box.
Having used the 18W8531 on several designs, 38 litres is too large and the Fb is too low. Have a look at Troel's designs and comments on the 18W8531 driver. Also have a look at his article on vent tuning where port length is usually 0.7 x calculated to achieve the target Fb.
For a test, try a port that's tuned to 40Hz to see if it makes a significant difference.
I've found 28 litres or smaller vented with this driver works the best and Fb > 35Hz for personal taste.
Try running the woofer without the crossover but add a 1mH inductor in series. This will flatten the response to 1.5k after which it will roll off smoothly. Even though it will not give BSC, it will give an indication of the raw woofer's performance in the box.
Attachments
- Status
- Not open for further replies.