OT - can an EQ help compensate for an undersized box?

[3rutu5]

A -3/-5db shelf filter on the mids+highs should give about the same result as a +3/5db boost on the lows (assuming the overall EQ shape is about the same and neither is causing clipping, noise, nor too little input)...because the woofer's excursion and/or the amp's wattage will give the same maximum headroom either way.

There IS still a decent chance the EQ shape might be a little different between the two options though; because -db mid+high shelf will likely leave the bass response/shape alone while a bassboost will likely boost a particular low-end frequency while leaving the frequencies below that boost slanting lower...so the bassboost might not cause as much excursion below the selected frequency (a positive in its favor unless you're trying to reach as low and flat as possible).
A bassboost "shelf" wouldn't have that difference/possible-benefit though.

Something like this:

Keeping in mind the maximum headroom of all three will be very similar (though the middle, bassboost won't run into super-low-frequency excursion problems as early as the other two).

interesting, i wasnt sure if i should be boosting the higher ends as well, i drew this up in paint.net, so not sure if it would actually work in the way im picturing this as i think im thinking of this more like the transfer chart in Winisd.

I really need to do some trial and error with this, which not quite sure how this will be done after i solder up the sides and it will block my little pogo pin bench/clip from being able to access the sides.

[3rutu5]

Well, an old trick is to use a too small cabinet, ported. This creates a peak that can be trimmed with EQ, reserving some amplifier power in that passband. Apply a bit of eq on the bottom end to lower the F3, will consume more power than you reserved by notching the peak but still somewhat more efficient.

At the end of the day, it is a 2" driver - what are reasonable expectations?

I wont quote actual numbers, but to my ear, if i could run it as a test at varying volumes and have similar bass reproduction to that of a sound link mini (assume less than) i would be happy. These little projects are fun and lowfi at best and the expectations cannot be to great as like you said its a 2inch in 0.015cuft enclosure. But in the same breath, ND65 is about the same size and it sounds good getting into the 60hz range, but also in a 4-5x larger space.

back when i built my quareenteenie (BF37 build) i was watching a hell of a lot of youtube videos of people making these small builds to compete with the bose, albeit larger. This was when i found out about these low powered 5v BT/AMPs (CSR6###, CSR8### and QCC3003) which had fully programmable chipsets. As the more successful one is something that is for sale AUKITS, who has provided a custom EQ setting on the board if you buy it, i know it can be done and also know he wont provide the secret herbs and spices to the general public. One suggestion i've heard is if you have anything above the zero DB line, to lower the entire setttings by the same amount, so if 60hz is set to 5db, apply to the master setting -5db. Another suggestion was roll off anything less than 60hz, which to be honest i'd be surprised if the little speaker wouldnt distort in that range, so like someone mentioned above roll off around 80hz. I did find it interesting the AUKIT one does mention they can get the little 1.5inch bose clone drivers down to 60hz, they dont mention at what F this may be, but though the headphones on youtube it sounds pretty nice......but probably being recorded 30cm away from the unit and not really a true representation of how it would be on a table if having a bbq etc

anyone that is interested in the 3d print itself, i've done a quick assembly video, just phone quality..

https://youtu.be/BoLTSlS8DtA

[3rutu5]

Got distracted from the other 3d printing project......designed the screw top to house the 18350 battery, BT/Amp and charging/step up module. Then if (when). Something fails it is easy to access and swap over. I found that vapers use a 18350 cell which seems to have a greater capacity than the 16340/Cr123 ones. Still only 1200mah but should be enough for a.little bit of playback time at a normal volume.

It's a raw print at the moment, so need to figure out how to finish it and still need to look into the best EQ settings for this to sound good.

[3rutu5]

oh thats right, the simple testing rig, i was actually going to give that a go...thanks for the reminder, completely forgot about that.

[Chris Roemer]

You COULD use the (free) ARTA "LIMP" module to get an impedance sweep plot, then you could SEE your tuning freq. (pretty sure this has been suggested to you before ... but you don't really want to "measure" anything, right?).
OR, you can aim your active driver straight up and put a few rice grains on the cone and do some FR sweeps. At the Fb, the rice (bouncing on the active driver) should nearly come to a standstill.

[3rutu5]

With that sized driver, I'd try to get maybe 100 hertz and up sounding good and full and smoothly roll off below that. You may be surprised how well that works, don't worry about some magic number.

100% right, with my BF37 build I fluked the tuning and have no idea what I actually did....as I used the Chinese passive membranes and not really sure how to model it, I put it in like a vented enclosure in winisd and then put few coins on them. Since then I realized how wrong that is, but it sounds really good for the size and I don't know what the box is actually tuned to. Other small builds I've tuned properly and they sound horrible in the low end, often breaking or sounding muffled when the bass hits.

I wasn't sure if with all those settings on the chip if I could pull off the old Bose trick of small driver good bass and low to mid volume and reduction of the doof doof as the volume gets higher
I'll do some research and have a bit of a play with the EQ graph, but totally agree with the 100hz and above

[Billet]

With that sized driver, I'd try to get maybe 100 hertz and up sounding good and full and smoothly roll off below that. You may be surprised how well that works, don't worry about some magic number.

[3rutu5]

ok, been a fun little afternoon, been able to change the name of the BT device and get to the sound settings. Any suggestions for an EQ settings that would give me a little bass and sound good without distortion? The suggestion online was to drop the overall DB's ie make it -3 so nothing peaks above 0 and remove out anything that the speaker cannot actually play like 50hz and below.

[3rutu5]

this is the EQ setting i liked for my general home setup, sounded good, but never tried it on anything like this before.

[3rutu5]

Yeah this was meant to be just a small portable project that I've been trialling and built variants that I haven't really been happy with to date, I tried a ported one over the passives and it seemed to show some promise but I still needed to work out the internals. I had this nice sounding EQ preset for my foobar called "home theater" which I was going to try first

I think from memory I was using some TS specs from a 8ohm variant as the 4ohm peerless was a special order and they won't give up the data for it....yes I haven't done any measurements. I have a 1s charging board with a voltage booster in it and going from 3.7 to 6v and using a small battery with 700-1000mah and combining it with an EQ is sounding like I'd have a pretty short run time...

Internal volume is about 0.015cuft, which that is visaton BF37 Territory but then would be restricted in overall volume of the music

[LOUT]

Question for all as I think about this, could one apply a shelf filter above the current F3 by say -3db to -5db and get the same result as boosting the low end and then just overcome with volume compensation on the amplifier? I have not modeled this but seems that it could work theoretically to a point.

A -3/-5db shelf filter on the mids+highs should give about the same result as a +3/5db boost on the lows (assuming the overall EQ shape is about the same and neither is causing clipping, noise, nor too little input)...because the woofer's excursion and/or the amp's wattage will give the same maximum headroom either way.

There IS still a decent chance the EQ shape might be a little different between the two options though; because -db mid+high shelf will likely leave the bass response/shape alone while a bassboost will likely boost a particular low-end frequency while leaving the frequencies below that boost slanting lower...so the bassboost might not cause as much excursion below the selected frequency (a positive in its favor unless you're trying to reach as low and flat as possible).
A bassboost "shelf" wouldn't have that difference/possible-benefit though.

Something like this:

Keeping in mind the maximum headroom of all three will be very similar (though the middle, bassboost won't run into super-low-frequency excursion problems as early as the other two).

[billfitzmaurice]

I expect you need a larger driver. The smallest I use are fours, only above 80Hz, and eight per cab at that.

[johnnyrichards]

Well, an old trick is to use a too small cabinet, ported. This creates a peak that can be trimmed with EQ, reserving some amplifier power in that passband. Apply a bit of eq on the bottom end to lower the F3, will consume more power than you reserved by notching the peak but still somewhat more efficient.

At the end of the day, it is a 2" driver - what are reasonable expectations?

[anunnaki]

Using EQ can be done as mentioned but you typically need a very Stout driver with a lot of overhead capabilities in xmax and thermal/power compression. This also assumes you are using an amplifier with sufficient power to be able to deliver the power needed with the boost applied.

Question for all as I think about this, could one apply a shelf filter above the current F3 by say -3db to -5db and get the same result as boosting the low end and then just overcome with volume compensation on the amplifier? I have not modeled this but seems that it could work theoretically to a point.

[fpitas]

Besides eating lots of extra power, be careful using EQ boost below the port tuning frequency. You can exceed xmax real easily. Sealed doesn't have exactly the same problem, although you can still exceed xmax easily enough.