Build of Bagby Mandolin speaker

[mtmpenn]

Very interested to see these measured!

Recognizing the challenges with subjective comparisons of speakers (particularly when you are only building one and when you have them at different times), I’d be curious to hear thoughts of this versus the S2000.

We will have the measurements in time, but I’m impatient!

 

[Rick Sykora]

Very interested to see these measured!

Recognizing the challenges with subjective comparisons of speakers (particularly when you are only building one and when you have them at different times), I’d be curious to hear thoughts of this versus the S2000.

We will have the measurements in time, but I’m impatient!

Amir should test the S2000 soon, but a comparison to the Mandolin is a stretch. The S2000 woofer seems well designed, but is a bit of an unknown. Even if it is as good (or a bit better) than the Mandolin woofer, it is much smaller and in a smaller cabinet. Without a subwoofer, it is at a major disadvantage to the Mandolin.

Plan to do my basic measurements later today. Since I have the other S2000, can do a quick comparo.

 

[Mattcc22]

Very interested to see these measured!

Recognizing the challenges with subjective comparisons of speakers (particularly when you are only building one and when you have them at different times), I’d be curious to hear thoughts of this versus the S2000.

We will have the measurements in time, but I’m impatient!

I have a pair of S2000 MT's in my living room with a ported UM10 sub. Very detailed speakers but can be pushed to the woofers limits very easily on their own. With the sub I think they sound very good and can get plenty loud for the average listener. The mandolin has a larger woofer so I would expect it to have much more punch in the bass without a sub.

 

[Rick Sykora]

Just to show why testing is important even for an experienced speaker builder, I got a little lazy and did not mark the driver connections. If you look at my earlier posted crossover, the positive tweeter connection should have been connected in the upper right (between the resistors in the L-pad). I hooked it to second resistor and this caused a frequency drop of about 5 dB from about 5 kHz on.

I suspected something was off in my quick initial listen, but I have fixed now (and sounds lots better) and need to remeasure impedance. Then will post the graphs.

 

[Rick Sykora]

Here are my measurements...

Bagby's measurements were minimal, but the one that is available is comparable enough that am confident in the build.

Kudos to Meniscus Audio for spotting my wiring error and confirming the build as well!

 

[Rick Sykora]

Very interested to see these measured!

Recognizing the challenges with subjective comparisons of speakers (particularly when you are only building one and when you have them at different times), I’d be curious to hear thoughts of this versus the S2000.

We will have the measurements in time, but I’m impatient!

As promised, here is a measurement comparison between the Mandolin and the S2000. Note that the Mandolin is much flatter and, along with the stronger bass, seems a better overall speaker to me. Best I can do until Amir tests. May be a while as he does not have the Mandolin yet and his lead time is running 4-6 weeks.

This measurement was done at my usual settings - on-axis, 1m, gated at 3ms. The signal level was not changed...

 

[mtmpenn]

Thank you!

 

[Kustomize]

Thankyou!

that dip around 3khz seems bigger than what Jeffs measurements show :S his is like 1 to 1.5 db.

 

[Rick Sykora]

Thankyou!

that dip around 3khz seems bigger than what Jeffs measurements show :S his is like 1 to 1.5 db.

Yes, I noticed that too, but he also appears to have a more major dip around 1k. These drivers have been updated since his measurements, so may explain the diff. May bear some more scrutiny, but my scaling is also more revealing. Would be.nice if there were some other build measurements to compare but have not found any. Otherwise, may just have to wait for Amir and see what his fancy rig bears out.

 

[andreasmaaan]

The discrepancy at 3kHz could also be due to differences in vertical measurement angle. Are you sure you measured on Jeff’s intended design axis?

 

[Rick Sykora]

The discrepancy at 3kHz could also be due to differences in vertical measurement angle. Are you sure you measured on Jeff’s intended design axis?

As usual, I measure on tweeter axis. Would not expect Jeff to do otherwise but his doc does say measured on design axis. So, you may be right!

 

[andreasmaaan]

As usual, I measure on tweeter axis. Would not expect Jeff to do otherwise but his doc does say measured on design axis. So, you may be right!

Yeh, I would also be surprised if it weren't the tweeter axis. And actually, I'm just looking at Jeff's Designer's Comments doc now. The crossover point is 1.8kHz, which surprises me as it's quite low to run that SB29RDC! But it does mean that the little dip you have at 3kHz is unlikely to be crossover integration-related.

 

[Rick Sykora]

@andreasmaaan design axis post reminded me of something. Have said before, but think it is worthwhile to remind those who might obsess over these measurements. While they can be useful in a relative sense, would caution against overemphasis on any small deviations. If you move the mic slightly off axis, the results can shift significantly. This is particularly true for vertical displacement for multi-driver speakers. I typically use a laser pointer to make 2 or more checks that am measuring on the tweeter axis.

Finally, given what I just mentioned, will also add that the Mandolin/S2000 comparison was done keeping the tweeter vertical axis at the same height. Just as with any valid scientific measurement, I tried to ensure that the only change in the test setup was swapping the different speaker. So, the comparison should be valid useful one.

 

[Mattcc22]

Any updates on testing by Amir? Hoping to see the full measurements soon! I know he's been a busy cranking stuff out.

 

[Rick Sykora]

Any updates on testing by Amir? Hoping to see the full measurements soon! I know he's been a busy cranking stuff out.

Between the holidays and fixing the Purifi SPK5 that got busted up, have not shipped it to Amir. The SPK5 is shipping to him today.

The Mandolin needs a permanent port and may take another look at that 3k dip before I ship. Am very interested in seeing it tested, but there have been higher priorities all around lately.

 

[Mattcc22]

Between the holidays and fixing the Purifi SPK5 that got busted up, have not shipped it to Amir. The SPK5 is shipping to him today.

The Mandolin needs a permanent port and may take another look at that 3k dip before I ship. Am very interested in seeing it tested, but there have been higher priorities all around lately.

Gotcha, thanks for the update.

 

[MusicNBeer]

That 3kHz dip may fill in quickly as you go off horizontal axis with the tweeter's wide dispersion.

My DIYs are the same SB woofer in MTM configuration with Scanspeak HDS tweeter. Mine are crossed low at about 1.7KHz. The MFC is a great sounding woofer at an excellent price.

 

[Wolf]

Try again, the ring radiating dimple domes do not have great off axis response. Back when Jeff was designing these and the vented Piccolo, I was not a fan of the SB29 dimples. Every build to that point sounded dull with them. Jeff left a little rise to yield a flatter power response, where others before had not, and this made all the difference in their treble tonality.

 

[Rick Sykora]

Just finished some more analysis of this speaker. The source of the 3k bump is either the woofer or the crossover. All drivers have been updated since Jeff designed and might account for the variation. The woofer nearfield response has a dip around 1k Hz but the tweeter response is smooth over its bandwidth. The new version of the woofer (used here) is supposed to be flatter around 3k Hz than its predecessor, but is close to the high end of its range that I suspect it might benefit from lower crossover point. Btw, the bump is still prominent off-axis too.

A quick check of the crossover confirms it is built correctly and major components are well within tolerances.

At this point, time to send off to Amir. When I get them back, may see if I can improve, but they are close enough to Jeff’s original measurements that am confident they represent his design.

 

[Rick Sykora]

Hurray! Yes, this one finally shipped to Amir. He should have in about a week...