OB with Alpair 12P and Eminence Beta 15

[Max N]

Thanks both.
When I was using lspCAD, I seem to remember the sound card/audio interface had to be very low latency. Does the USB interface give any problems in that area?

[IslandPink]

You'll have problems getting data below 500Hz with the microphone because of room modes. I ended up using a radioshack spl meter on 'C' weighting and taking 4 or 5 sets of daya for discrete frequencies then averaging all the data in Excel, when I tested the Folded horn. You could do the same if you can get the data out of the FR traces.
Someone like Chris or Ed would probably know what the 'proper' way to do this would be.

[simon]

I keep meaning to post the the sim results

Please do Max

[Max N]

As stated before, these are pure simulation, no measurements, so take with a pinch of salt....
This is with the Betas in parallel:

The problem is that the impedance is maybe too low at low frequencies....
This is with the Betas in series:

Mark's right, they are rolled off at LF, but I think I was OK with that because I'm trying not to excite the 40Hz room mode.
This is in series but with room gain:

[MJK]

Something does not look right.

I would have expected the series and parallel connected woofers to have essentially the same shape for the SPL frequency response curves but just at different output levels, the parallel connection producing +6 dB more output for the same input voltage. Seeing the significant differences in the curve shapes does not pass my eye-ball check.

[chris661]

I'm also wondering if the simulation is for the actual driver. Most 15"s will have a breakup peak somewhere between 1-2kHz.

Chris

[MJK]

The software does not account for cone break up, it assumes a rigid piston motion for the cone.

[chris661]

The software does not account for cone break up, it assumes a rigid piston motion for the cone.

Cool, thanks.

Chris

[Max N]

These plots include passive filters, I have used different passive filters / crossover points to get a better blend taking account of the different sensitivities, if that makes sense....

[MJK]

I don't think changing a passive low pass filter in the 200 to 400 Hz will have that big of an impact on the shape of the low frequency roll-off. Looks like a modeling error to me.

[IslandPink]

Yes I agree with Martin - the roll-off looks plausible , especially the second plot, but the curve should be the same although at a different level ; in the end the efficiency per watt should all come out in the wash if you apply the appropriate drive voltage ; however the crossover would be different .
Oh, hang on, is the actual choke used in the sim ? - and does the effective Q to the bass drivers differ because you're using the same choke to driver a series (16R) pair or a parallel ( 4R ) pair ?

[MJK]

Over the past two evenings I exercised my latest MathCad worksheets for the Alpair 12P combined with one or two Alpha 15A or Beta 15A woofers. For two woofers I looked at a parallel connection for the Alpha 15A's and a series connection for the Beta 15A's.

There are definitely workable solutions with any combination of these drivers. I think and excellent OB system can be built.

I also played with BASTA a little more and was able to essentially duplicate this design work. So keep trudging forward with learning the simulation tool BASTA and I think you will be able to assess the various trade-offs and derive a very good OB design.

[Max N]

Thanks everyone for the comments. Sorry I haven't had time to respond more fully. I will check exactly what settings I have got in Basta to make sure I haven't cocked anything up. If not I'll post some more plots showing how the passive filter is modifying the roll off. Or if I have cocked it up I'll post some corrected plots.

The Alpairs are still running in, but I am a little worried that I may have damaged them, as they are sounding quite coarse. I was very careful to follow the recommendations, very low volume for the first 100 hours etc.
They sounded sweet when I first hooked them up, but not now. Anyway I will persevere. I think I'm at about 400 hours now. There is no capacitor or other protection, and I have them playing reasonably loud with visible cone movement

[Max N]

As promised, some more plots from Basta showing how the passive filter is affecting the Betas.
All of these plots are comparing the series vs parallel connection. The first plot is with no passive filter. As expected, the parallel connection gives 6dB extra sensitivity


Now with an equivalent (same corner frequency) low pass passive filter applied to both


Same as above but with a lower corner frequency


Finally, we compare the series connection with the higher corner frequency with the parallel connection with the lower corner frequency


So the different low-end roll-off is just the extra sensitivity combined with a lower corner frequency on the low-pass filter

The Alpairs are still breaking in. The sound continues to change, it has improved since the last post, so I think I am just hearing non-linearity in the suspension as it loosens up with the (slow and gradual) increase in signal I am feeding them.

[MJK]

OK, your plots make sense now.

Looking at your final plot, which produces the deepest bass output, you are using two Beta 15A woofers with a lower crossover point producing about 95 dB/2.83 V/m down to about 50 Hz. I have a very similar simulation that produces a little more output (96 dB/2.83 V/m) using two Alpha 15A woofers and a higher (less expensive) crossover frequency.

In summary, I think you can get the same low frequency results using less expensive woofers and crossover components. What performance advantage are you gaining by using the Beta 15A woofers?

With respect to the sound being course during break in of the Alpair 12P with a full range signal, I think the performance of the full range driver will improve significantly once you remove the low frequency signal and redirect it to the big woofers. A high pass crossover on the Alpair 12P driver will clean up the performance and relieve low frequency strain on the full range driver