HOw close does driver need to be to floor to avoid floor bounce?

[Kanga]

Hi

I'm building a pair of bass boxes to turn my 2 way AE1 speakers into three ways. The bass boxes will be about 650mm high x 275 wide x 425mm deep (about 26" x 11" x 17" for the US) and they'll have one S-S 25W8565-01 driver each and powered by a 100W AKSA amp. At the moment I'm thinking that the driver centres will be near the top of the enclosure, around 500mm from the floor. Crossover will be active 12 dB/Oct at around 300 Hz to the AE1s.

Having heard about the evils of floor bounce dip at around say 150Hz, I wondered how close the driver would be to the floor to avoid floor bounce. I assume that once you get close enough the floor the driver is radiating effectively into half space and so you don't get a reflection. This may put the driver too far away from the bass/mid driver of the AE1s (which would be at a height of around 800mm / 31.5"). The normal rule here is to keep driver spacing within 1/2 of the XO frequency wavelength, which would be about 580mm (22.5") here. That would mean that the bass driver centre could be around 220mm (8.5") from the floor.

Most of the 3 way designs I've seen seem to put the driver up the top near the mid driver.

Any thoughts on this out there?

Mick

[Jimmy154]

Originally posted by Kanga
Most of the 3 way designs I've seen seem to put the driver up the top near the mid driver.

I've seen that also. It's a good idea for sound quality. But some people also say for a 10 or 12 inch makes more bass about 12 inches from the ground. From the middle of the woofer.

[pinkmouse]

Ah, those damn laws of physics...

For true 1/2 space radiation, the centre of radiation of the driver should actually be at floor level, obviously not possible, so most people settle for getting it as close as possible.

Then the problem, as you say, comes in placing the other drivers! You are correct in saying that diffraction problems occur when centres of radiation are a significant fraction of a wavelength apart at crossover frequency. So, where does this lead us... I can see only one solution, build speakers like stage monitoring wedges, all the drivers close together, at floor level, and sloped upwards to ensure on axis radiation at the listening position!

Seriously, I can't see that going down very well on asthetics, but one other possible solution is to use a multi driver box, where the coupling of the acoustic centres will reduce this problem, but add others, such as limited sound staging.

Ah, those damn laws of physics...

[f4ier]

Yavuz A's Room Response Calculator and Paul V's Frequency Response Combiner might interest you. You can find both from the FRD Group site

[jackinnj]

all those DJ's with their monster speakers mounted on tripods -- I've mentioned to two of them, on separate occasions, that they would get more base out if the units were floor, more against a wall, and even more if they were in a corner !

just the tweeter/mid or horns need to elevated.

of course, this is if the sound were listenable anyway.

[Kanga]

Thanks for the responses so far.

One thought that I had is that if I build the box symmetrical top to bottom, then I can try it with the speaker at the top or the bottom just by tipping the cabinet upside down and see how I like it before putting feet/spikes on to the finished cabinet.

Mick

[pinkmouse]

then I can try it with the speaker at the top or the bottom just by tipping the cabinet upside down

Yup, the true spirit of DIY!

We can all talk physics, psychoacoustics and other nonsense for days, but if it sounds good to you then go for it!

BUT... bear in mind that the higher range of the driver, the more it is directional, so to give the common example, tweeters need to be at about ear level, or angled to point at that level when listening.

Have fun...

[navin]

i did something simialr and found that in my case the wofoers sounded better near teh floor.

my woofers were 8", the top was a 6" 2 way. since i was building the speakers from scratch alli did was re arrange the baffle nto teh 75kg cabinets.

hope this helps

[Kanga]

Well the bass cabinets finished apart from a few details and I've been listening for a few days now. Full report still coming.

Unfortunately I didn't manage to build the cabinets symmetric so that I could turn them upside down. The centre of the driver is normally 55cm (21.5") from the ground. However in measuring the output from the speakers with a sound level meter at the listening point, I noticed that there was a noticable drop centred on 200 Hz. Using 1/3rd octave pink noise source from a CD, the level is down around 4dB at this point.

Turning the cabinets upside down so that the driver is only about 22.5cm (9") from the ground, the response is only 1 dB down. The response a bit higher up is a bit more ragged though, presumable because of the distance between the bass and mid driver. On listening the sound is definitely more even through the bass. Unfortunately leaving the cabinets permanently inverted is not really practical, due to the way that I've built them.

What to do then

Maybe I could actively equalise. Putting another op-amp in the ESP active crossover where the unused output buffer opamp normally goes wouldn't be that hard, maybe with a bit of bodgeying. All the infrastructure is there.

Alternatively I could lower the XO point to 200 Hz from 300 Hz, and overlap the XO points slightly to get rid of the dip. This brings the XO point a bit closer than I'd like to the natural rolloff of the AE1s.

Any suggestions on possible solutions?

Mick