Offsetting midwoofers from mid line?

[HiFiNutNut]

Hi,

I am re-designing my MTM box.

To reduce deflection I normally see people offsetting the tweeter from the mid line. However, I have not see people offsetting the midwoofers.

The XO point is at 3k using LR4. The midwoofers are Peerless HDS 5 1/4 (850488).

Is there any point of offsetting the midwoofers?

Would the 1 : 1.618 golden ratio work the best?

Thanks.
Bill

[bjorno]

Hi Bill,

Is there any point of offsetting the midwoofers?

My advise to you is first to visit this thread:

My plans for my NSBs - feel free to make comments!

See post # 11 and take a look at the submitted picture that shows the M- (horizontal offset T)-M is a much better choice than a vertical aligned if you must crossover high for a unknown reason.

The polar plot’s are for 4” drivers but for 5.25” a 31% increase in diameter.
For example, by multiplying the frequency with 1.31 x the frequency shown gives the next plot (about1.25 kHz) in order, which is close enough to be valid for the 5.25” drivers i.e. The polar pattern at 1 kHz for the 4” drivers corresponds (about) to the polar pattern at 1.25 kHz for the 5.25”drivers.

Would the 1: 1.618 golden ratio work the best?

You can always use the Golden Ratio as a start, but the reality its more complex than a simple rule of thumbs, play with The Edge and you’ll see.

Download Svante’s The Edge at Tolvan:

http://www.tolvan.com/edge

Another useful program is Xdir that shows the vertical directivity pattern for 2-3 point sources in an infinite baffle.

Insert the driver c-c distances; woofer and tweeter phase lag dependent on your MTM crossover and you can study the impact on the vertical polar patterns:

http://www.tolvan.com/xdir/

Look at the picture: Xdir_MTM.gif where I put in your approximate numbers. The first one (1) is your MTM in a vertical straight line. It doesn’t look so good, the crossover point is to high for the c-c distance that should be equal or less than a wavelength at the crossover frequency and what can be seen in plot # (4) is what you really should target when designing MTM’s.

b

1(1)

[DSP_Geek]

Originally posted by HiFiNutNut
Hi,

I am re-designing my MTM box.
[...]
Is there any point of offsetting the midwoofers?

Would the 1 : 1.618 golden ratio work the best?

The Golden Ratio is a bit much. If you're doing edge rounds, try offsetting the drivers by the width of one round. For example, if the round has a 1" radius in a 15" wide cabinet, then you might want to offset the drivers by about an inch, to wit 7" from one edge and 8" from the other, so the surface wave is off one edge before hitting the second. Perhaps 2" would work also, 6.5" to 8.5", which is not far from 1:1.414 so as to spread the discontinuity evenly across the spectrum.

As Bjorno says, Svante's Edge is your friend. Excellent program. He likely has a point about dispersion, but nothing says you can't offset the mids and tweeter by a different amount off the center line.

Cheers,
Francois.

[HiFiNutNut]

Thank you so much for your help and your information. I have not had access to the forum in the past a few days hence I am getting back to this thread late.

I have downloaded the software and will be playing with them.

I am now planning to ditch the HDS 5 1/4" and go for the ultimate of using the 6.5" Scan Speak revolator (with a diameter of 182mm) and the diyAudio Esotar T330D (140mm). So the c-c distance will be 161mm at the very miminum when all drivers are placed next to each other without gaps, which corresponds to a wave length of 2,136Hz.

So in theory, it is OK to cross over at 2,136Hz or lower. But would it be better to be lower than right at 2,136? I guess I can find it out from the Edge and Xdir.

A 2k Hz XO point for the SS seems reasonable. The Esotar T330D has the Fs at 750. I intend to use LR4. Do you think the Esotar can be crossed at 2k Hz with a LR4?

Regards,
Bill

[noodle_snacks]

I believe the idea of narrow cabinet in relation to the woofer is to keep the baffle diffraction ripple out of its passband, or where the woofer starts beaming, an optimally offset tweeter would then have a pretty smooth response, and you end up with a final response that is very smooth. Rather than using rules of thumb to decide the best driver positions, I would use simulation software instead, for a more accurate answer. AFAIK, the only fault with this type of simulation is the fact that they do not account for woofer beaming. I personally like the previously mentioned "The Edge" program. I have not used "xdir" before.

The second method for removing diffraction issues seems to be the use of relatively wide baffles, so one can offset the woofer(s) and tweeter in such a way to get smooth response again. Some people argue that this method has poorer imaging (diffraction causes a more spread out sound source), but I couldn't comment on that beyond speculation. Narrow baffles would have a higher "WAF" though.