Dayton DSA315-8 with Pioneer PD-40
In this exercise, I am testing a 12″ woofer with the Pioneer PD-40 mid horn. For convenience, I am running it active with a Linkwitz/Riley 24dB/oct crossover. If it works well, I can then move on to designing the passive crossover.
The woofer I am using is a Dayton DSA315-8. This is the Designer Series with an aluminum cone. Frankly, this is the first time I’m crossing a 12″ at 700Hz. I am hoping that the extension in this DSA315-8 will enable me to cross smoothly with the Pioneer PD-40.
Fig 1 – Blue plot=DSA315 with Low Pass Filter | Red plot=PD40 with High Pass Filter
Fig 1 shows the frequency response of the DSA315 and the PD-40 with my active crossover set at 700Hz. No smoothing applied. Frequencies below 400Hz are in nearfield.
The two drivers appear to be crossing properly. The DSA315-8 cone breakup peak at 4kHz is sufficiently suppressed with a 4th order low pass.
Fig 2 – Summed response of DSA315 with PD-40
The summed response in Fig 2 shows the DSA315 integrates better than the Dayton DA215 that I used in the Sandpiper. There is only a light dip at 800Hz whereas with the DA215, the dip is -5dB at 700Hz.
Fig 3 – Null response
To check on the alignment, I reversed the phase of the PD-40. The Violet plot in Fig 3 shows a dip at 600Hz. This indicates the two drivers are quite close.
Fig 4 – Time aligned Null
As with most speakers with horns, the sound from the woofer reaches the listener first. To get the woofer and horn to reach the listener at the same time, I delayed the woofer until a sharp null appeared (Fig 4). This means that at 700Hz, the woofer and the horn are in phase.
Fig 5 – Time Aligned Response
To see how much of a difference the time alignment makes, I reversed the PD-40 phase back to what it was before. The dip at 800Hz (Fig 2) vanished completely. Now, the DSA315 and the PD-40 are summing properly.
Fig 6 – Toneburst Energy Storage
The Toneburst plot shows some stored energy (light blue slices) emitting from 2kHz, 6kHz and 15kHz. They are not serious as they occur at -20dB below the fundamental. Furthermore, they last not more than 8 cycles.
Fig 7 – Waterfall
Another 3 dimensional view of the behavior is the waterfall plot in Fig 7. In this plot, the Z-axis is in time instead of cycles. The stored energy recorded with the toneburst can be seen at 2kHz, 6kHz and 15kHz.
Fig 8 – Spectrogram
Where a spectrogram is useful is it shows the relative loudness at different frequencies. From Fig 8, the excess energy at 6kHz and 15kHz recorded in the toneburst is quite insignificant for they last no more than 1 msec. The one at 2kHz last slightly longer (3 msec). I’m sure I won’t be able to detect them.
the Sound of the DSA315
When I first heard the DSA315-8, I couldn’t believe my ears. It was horrible. Everything was wrong. The bass was congested and sounded compressed. Like suffering from a bad bout of constipation. And this is after a few days of continuous music with it.
I think I must have been in shock. It’s not often that I’m confronted by something like this. After gathering myself, I pumped up the volume and left my lab. When I returned a few hours later, it sounded like a new speaker. I said to myself “Damn, this woofer needs some serious breaking-in”. I am now ready to see what this woofer is made of. I threw the tracks that I use for bass, Fourplay, Jamie Valle, Santana, Bela Fleck and others and this is what I found.
The bass is tight. Not as tight as pro woofers but tighter than most Hi-Fi woofers. But that doesn’t mean it’s fantastic. The problem is there’s slight bloom in the bass notes which contributed to some loss in clarity. A bit of a one note bass effect. Upper bass is almost non-existence. Typical of what you get with subwoofers. To make matters worse, there’s still come compression in the bass notes. When all these add up, the result is a lack of texture and dynamics.
If I were to make a comparison, the DSA315-8 is one grade above the DA270-8 but a few grades below my Peerless NE265W-04. It may sound better with more break-in or it may be outstanding when I cross it at 125Hz. I don’t really know but as it stands right now at 700Hz, I am not impress. What I’m looking for is dynamics, clarity and texture. Out of 10, dynamics is 4, clarity also 4 and texture, 3. I’m not expecting 10 but a 7 or 8 in all three categories would be nice.
To be fair, I will investigate the DSA315-8 further. I must add that this is not a bad woofer. It’s just not what I’m looking for.
DSA315-8 with PD-40
This last chapter is about the integration of the DSA315-8 with the Pioneer PD-40 horn. On paper, it should be easy. It is, most of the time, except for one critical factor, the SPL.
When auditioning, I found the vocals and treble a bit harsh, especially when I pump up the volume. My crossover is correct, so is the flatness in the response. Where can the problem be. Is it that the Pioneer horn is distorting? That is highly unlikely because the sensitivity is so much higher than the DSA315-8.
After a while, I turned down the volume of the horn slightly, and I mean slightly. Now the problem disappeared. The speaker sounds perfect. I swept the response again and this is what it showed.
Fig 9 – Final Frequency Response
The Brown plot is my original response in Fig 5 above. The Black plot is when I lowered the volume of the PD-40 horn. The difference is only 1dB.
Fig 10 – Null Response
With the PD-40 horn reduced by -1dB, I reversed the phase for a null response. My new null (Fig 10) is deeper than the previous one (Fig 4) by +5dB.
The moral of my story is in the end, I still have to rely on my ears. The response in Fig 5 looks perfectly fine to me but sounded terrible. Measurements are just that, measurements. They don’t tell me how a speaker sounds like.
Update – July 18, 2020
I just realized that the DSA315-8 Mms is 115grams. There’s no way this woofer will do what I want with that kind of weight. I will have to look elsewhere.
Unless otherwise stated, all measurements were made with the mic at 36 ins, tweeter axis. Impulse Window=5ms. No smoothing applied.
Dayton DSA315-8 mounted onto a 53 liters test box of 14-1/2″ width. Box tuning at 33Hz.