Trying to improve low end bass response - from 80Hz (current) down to 35Hz...

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
So, once upon a time a set of drivers (3*Bass, 2*Mid, 2*tweeters) from a pair of B&W XT4 loudspeakers appeared on ebay.

Thinking they were a bargain I snapped them up and began my first attempt at a DIY loudspeaker. Along the way I've learnt a lot more about crossovers and drivers, but my understanding of cabinet frequency response and tuning is sadly lacking.

My understanding was that any cabinet has two resonant frequencies, that from the internal box volume, and that from the port (if we're talking about ported cabinets that is) which balance to produce your tuned frequency.

So i built the bass driver (single, not pair) into it's own 8.5L enclosure, vented with a 2" diameter pipe of about 14" length running up the rear of the box, then making a rapid right-angle turn out the back (centre of the rear cabinet face)

My problem is that the bass response is frankly terrible. I've managed to get it down to about 80Hz, but no amount of experimentation has seen anything better. (have tried an internal folded transmission line, ported at the front base of the cabinet, tried various 'port lengths' by adding different lengths of pipes to the rear port)

My confusion is that adding/removing lengths of pipe to the box doesn't seem to have much effect on the FR of the speaker, although there is clearly an improvement in bass response between the ported and closed version, it's not much. The problem is compounded by the impdance of the drivers - 4 ohms for the mid and tweeter, 8 for the bass, plus bass is woefully lacking in efficiency compared to the mid/tweeter units.

As the mid range is clearly the same as that used in the 2-way XT2, I considered usings it in a 2.5 way system, but I was worried by the possible distortion that could be created if the air volumes of the mid and bass were linked (or is this an unfounded concern)

At the moment the system is set up as a 3-way, with a lot of padding on the mid and tweeters, and very good (flat) FR and overall sound, but it is lacking that bass end, which seems ridiculous.

Adding another bass driver in parrallel just increases the overall bass output, and results in a BIG spike (well, hump) at around 100Hz.

Have I made a massive error with the 8.5L cabinet volume?
Obviously with the amount of work that's gone into everything I'd rather keep what I can, but there's no way I'm going to be happy with a speaker of this size that can't produce anything noticeable below 80Hz... :headbash: more importantly the idea of a 3-way design that struggles after 90Hz is ludicrous.

please help!
 
Have you got any Thiele-Small parameters for the woofer?

You are not going to get down to 35Hz with a small woofer, especially not in that size box. I think you have isolated the mid box from the bass box, but can I just confirm that is the case?

You seem to be tuning about 45Hz, so adding port length is unlikely to make much difference as you are already near the bottom of what your tuning should be.

Possibly you have a wiring polarity problem - if you listen to just the bass of one speaker on its own are things any better?
 
Last edited:
Unfortunately no thiele-small parameters, but FS is about 65Hz.
The original XT4 Cabinet seems to have an internal volume of about 16L and obviously used two bass drivers in parallel. Would building a new, larger cabinet make much difference to the bass response?

The bass is isolated form the mid, when testing I only had the bass woofer plugged in, but no problems with polarity when plugging the mid back in.
 
Just another Moderator
Joined 2003
Paid Member
Time to get yourself a copy of REW and make up a simple cable jig Cables you need the impedance cable. You can then (provided you have a sound card with line in and line out) measure the T/S parameters yourself.

Actually you do need one other piece of equipment, either a test box of known volume, or a set of accurate scales. Depending on whether you you delta mass or delta compliance.

without the T/S parameters you are stabbing in the dark!

Tony.
 
Do have a design program such as Win ISD (available as freeware) or something similar that can model different box responses? You can model different volume boxes and tuning and see how it affects the response curve. It will teach you a lot about what you can do to shape the bass response.

Ideally you want the T/S parameters for your woofer, but you can probably substitute the numbers of a similar sized woofer from another manufacturer which has published specifications. Most quality small woofers are going to be roughly similar, with an Fs near 60 Hz and Qts in the 0.3 to 0.5 range.

You are not going to get usable response down to 35 Hz with this size woofer. 60Hz-70Hz is probably reachable.
 
Measuring T/S parameters is really not that difficult, although it requires a bit more time the first few times around to get familiar with the procedure. The "Vas" parameter is the trickiest one to get and requires precision with either the fixek-volume or added-mass method, but is ultimately required to calculate enclosure parameters.

Measuring Loudspeaker Driver Parameters

It really is the way to go if you are to design a bass enclosure that is optimal for unknown drivers. The signal generator can simply be your computer with a generator software, plenty of free ones available online.

Once this is achieved, look up "baffle-step" if you are not already familiar with the concept, this is something that'll also make it seem like the bass output is weak.

IG
 
Last edited:
from these responses I'm not entirely sure the best way forward, do i measure the driver parameters first or is there a way to increase bass response from what I have?

I guess what I'm asking is if there is a 'this will always improve low end response' technique, or if I need to know the parameters and have a design goal in mind in terms of port length...

If it all comes down to port length and frequency tuning, can i do it by trial and error? or is there a set design that will give 'most bass' for this size of cabinet?

The woofer (bass) section is approx 14x16x50 (cm, WxDxH) so I wondered if maybe a rear-loaded horn type set up would work better?
 
You need the T/S numbers.

Quick and dirty: build a 1.8m TQWT. (Internally 130mm width and 250mm footprint depth minimum.) Place the driver at around 700mm from the base. Place your port at the base of the enclosure. Start with about 6 inches of your 2" port tube and play around with length and stuffing in the top half until it sounds right.

You'll have to get lucky but this might be a good start.

Then add the other drivers.
 
Last edited:
Will EQ get you anywhere, or does it get limited by excursion then?

so...the set was missing 1 bass driver? I don't think the real complete speakers will go very low. 80 hz is not bad for one small driver.

Answer: add subwoofers and a dedicated amp.

It has a quite a large x-max and power handling it would seem. Im happpy to buy another driver, but it's £65 a time, and until I get at least one of the cabinetts responding a bit lower with either one woofer or two I wouldn't think about buying an extra one....
 
Just another Moderator
Joined 2003
Paid Member
My morel MW144's do 80 Hz in 5L sealed, and they are a 5" driver. They have a lower resonant frequency though at around 50Hz. I'm not sure about the prevous comment about modelling some similar sized drivers, I think you will find that you can get BIG variances between drivers of different specifications.

I've taken my (measured) T/S params of my MW144's and done a comparison in unibox (with the fs adjusted from the real 52Hz to 65Hz).

in a 8.5L box a 60 hz tuning seems ok. attached is the comparison between sealed 5L, vented 14L (which will give about the most extension) and vented 8.5L. These are only really valid for my woofers, but gives you an idea of what you can do if you have the T/S params.

The smaller box with higher tuning frequency will give you a better transient response at the cost of less bass extension. These days I aim for somewhere in the middle of these two. For these woofers if I were to go vented (which I didn't) I'd probably settle for something around 10 to 11L.

Tony.
 

Attachments

  • mw144 comparison.gif
    mw144 comparison.gif
    25.9 KB · Views: 62
Just another Moderator
Joined 2003
Paid Member
and to show how different woofers specs will vary the results. Attached is a comparison between the MW144 results and a HiVi 5" driver (also with fs adjusted to 65Hz, interestingly it has the same 52Hz in reality as the morel). The HiVi is a kevlar and paper cone woofer.

Box volume and tuning are the same, I just put in the HiVi's T/S params.

You can see that you really need the specs for YOUR drivers!!

edit: sorry about the legends, not sure what has gone wrong there, they normally are perfectly legible, my excel has a gremlin...

Tony.
 

Attachments

  • mw144_vs_hivi.gif
    mw144_vs_hivi.gif
    21.7 KB · Views: 63
Last edited:
Exactly, but there's really no need to even get into that far. The OP has a woofer with ~65Hz Fs, and knows the ballpark cabinet volume they were meant to go in. Measuring parameters would be great, but clearly an octave below Fs is not the appropriate tuning. A much shorter port should get something that at least resembles their intended response.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.