Single Sheet Challenge (with bass) !!!

[Brian Steele]

Hi Brian,

In this particular case, I don't think it is Hornresp that is wrong... 🙂.

Thanks David for stepping in 🙂.

In attempting to insert a constriction 90 cm from the mouth, you have actually changed the tapped horn system configuration quite significantly. In the original design the throat side of the diaphragm is offset 11 cm from the closed end S1. In your modified design the throat side of the diaphragm is no longer offset along the horn - it faces directly into the 100.1 cm long throat chamber, which you are using to form part of the horn.

Good point. The sim isn't a good match for the condition then when the driver's output is offset in the throat chamber.

The original design has a pipe length of 11 + 155.1 + 12.9 = 179 cm, which when multiplied by the constant cross-sectional area of 290 sq cm gives a volume of 51.91 litres.

The modified design has a pipe length of 100.1 + 1.8 + 0.1 + 77.1 + 12.9 = 192 cm. The volume in this case is given by (290 x (100.1 + 0.1 + 77.1 + 12.9)) + (1.8 x 40) = 55.23 litres.

Yes, it looks like I got the throat chamber length wrong. I forgot about including L34 in the calculations! The throat chamber should be 179 -77.1 - 0.1 - 12.9 = 88.9 cm, not 100.1 cm. I've redone the sim to reflect this - unfortunately it's still showing that a massive dip at 200 Hz that does not appear in the FR measurements. Charted in grey in the image below. Is it possible that this big difference is only due to the S1 issue you mentioned above?

 

[David McBean]

Is it possible that this big difference is only due to the S1 issue you mentioned above?

Hi Brian,

Changing the length of the L12 offset certainly alters the position and shape of the dip, but unfortunately does not eliminate it entirely.

Out of interest, I tried inserting a 40 sq cm constriction 90 cm from the mouth of the tapped horn in an AkAbak model. The predicted response is almost the same as before - the 200 Hz dip is still there (see screenprint attached).

It seems that neither AkAbak nor Hornresp can cope with your "dogfood duct" very well 🙂.

Kind regards,

David

 

[Brian Steele]

Hi Brian,

Changing the length of the L12 offset certainly alters the position and shape of the dip, but unfortunately does not eliminate it entirely.

Out of interest, I tried inserting a 40 sq cm constriction 90 cm from the mouth of the tapped horn in an AkAbak model. The predicted response is almost the same as before - the 200 Hz dip is still there (see screenprint attached).

It seems that neither AkAbak nor Hornresp can cope with your "dogfood duct" very well 🙂.

Kind regards,

David

As the HornResp and Akabak models like pretty much identical, perhaps it's time for me to go back and review the observations.

I've found one error already BTW - my S1-S4 settings had the path length after the DFD @ 90.1 cm rather than 77.1 cm, and the DFD is located 77.1 cm from the mouth according to my last set of measurements. L23 should therefore be 77.1-12.9 = 64.2 cm. Note, this correction also requires a correction to the throat chamber length, moving it back upwards to 100.7 cm. Redoing the sim with the corrected value for L23 and throat chamber volume results in the dip being filled in a bit more, but not to the amount I observed. The predicted 5dB peak before the dip also did not show up in the measurements.

I'm going to repeat the measurements as soon as I can, to see if there's anything else that I've missed.

 

[Don Snyder]

justint has built one!

http://www.diyaudio.com/forums/subw...40hz-but-only-20-deep-help-3.html#post2539832