[lowtech]

Okay.  So the artifacts that I'm referring to are caused by the design of your baffle and cannot be corrected via notch filters.

No, they are not caused by the baffle.

You are at a real disadvantage. You are guessing at what's going on. I have been there and done that.

Yes I am, however you'll need to elaborate since it's general knowledge that baffle diffraction artifacts can generally be identified by a smooth off-axis plot (which is why I asked for them in the first place).

Looking forward to seeing some new (hopefully unsmoothed) plots of your latest on & off-axis measurements.   

[Danny Richie]

Yes I am, however you'll need to elaborate since it's general knowledge that baffle diffraction artifacts can generally be identified by a smooth off-axis plot (which is why I asked for them in the first place).

This is not the case with a design like this.

The off axis response in those two upper octaves are controlled by or limited by the curvature or shape of the woofer that acts like a wave guide. Those upper ranges don't roll past and around the woofers cone and reflect off the surface of the baffle. So the baffle has no effect in those two top octaves.

[Bear]

sooo, errr,  what is an acceptable variation in those frequencies?

happy listening

shane

[Danny Richie]

I'd like to see as smooth and flat of an average as possible across the whole range.

But a narrow band peak or dip in the top two octaves are pretty hard to discern. A rising or falling overall response across the board stands out much more.

Also, looking at the room response, the top two octaves are likely to look just as smooth as any other speaker and will often be more effected by the room than anything.

[lowtech]

sooo, errr,  what is an acceptable variation in those frequencies?

I think that would be an interesting question for Mr. Richie to answer since he's placed a large emphasis on ruler-flat response from all of his previous designs.  Take his new N-1 design for example:



Discounting the rising response, the overall deviation looks like it's on the order of +/- 1db.  Assuming that level of linearity is important, how can the the measured response of the Super-V be considered acceptable?

Anyway, the new, revised (hopefully unsmoothed) curves of the on & off-axis response should be more telling.

[nickd]

With respect to "lowtech"

Don't put so much weight into simple charts. The design work Danny has done on some of his models would take volumes of white paper to explain. That said, I built a pair of his LS-9's with out ever hearing them because I had heard one of his earlier design's (the Ephipany 20/21). The LS-9 is one of the best (most musical and fun) loudspeakers I have ever owned. I just ordered a pair of Super V's without hearing them based on that same faith. Music is more than charts and graphs. A speaker either sounds "lifelike" or sits on your floor (often silent) collecting dust. To my knowledge, no graph has ever been able to explain this "life". They are just a tools for designers to check the quality and direction of their work.

No insult intended, just thoughts. Character and reputation of the designer holds enough weight and value for me. "Good" measurments are just a plus.

[Danny Richie]

Okay Lowtech. You're in luck.

I went back to my other computer (the one with the Clio on it) to look for any other files that I saved that I did not send to this computer. And I did have one saved that was made that same day when I took some off axis responses. This one has on axis, 10 degrees, and 20 degrees off axis. I am glad I saved that.



With a response that is +/-1.5db all the way up to a little above 3kHz and +/-2db all the way up to 6kHz. A little wiggle or so above that smooths out in the off axis.... Considering the size of this driver. I can live with that.

The room response looked great too. The top end averaged out really smooth and the bottom end had a near flat output to 20Hz. In my room it picks up a db or so from 25Hz down to 20Hz. I can live with that too.

The real test is always going to be how it sounds, and it hits a home run in that regard.

Maybe at some point you can quit looking for faults and just listen. You can't not like this speaker.

[sfdoddsy]

Having spent more time than is healthy measuring my OB speakers, I'm not surprised by the 'wriggles'. You simply can't expect the same ruler flat response you get with monopoles with a dipole.

The high treble seems a bit peaky, but in room I'm sure it would be less so.

If Danny is a measuring mood, I'd love to get an objective idea of the bass response. I'm sure it is flat to below 20Hz, but my query still at what volume.

A simple SPL trace based on a 100db or so average would be welcome.

[sl_1800]

Don't put so much weight into simple charts. The design work Danny has done on some of his models would take volumes of white paper to explain. That said, I built a pair of his LS-9's with out ever hearing them because I had heard one of his earlier design's (the Ephipany 20/21). The LS-9 is one of the best (most musical and fun) loudspeakers I have ever owned. I just ordered a pair of Super V's without hearing them based on that same faith. Music is more than charts and graphs. A speaker either sounds "lifelike" or sits on your floor (often silent) collecting dust. To my knowledge, no graph has ever been able to explain this "life". They are just a tools for designers to check the quality and direction of their work.

No insult intended, just thoughts. Character and reputation of the designer holds enough weight and value for me. "Good" measurments are just a plus.

 

Well said.

[TRADERXFAN]

Having spent more time than is healthy measuring my OB speakers, I'm not surprised by the 'wriggles'. You simply can't expect the same ruler flat response you get with monopoles with a dipole.

The high treble seems a bit peaky, but in room I'm sure it would be less so.

If Danny is a measuring mood, I'd love to get an objective idea of the bass response. I'm sure it is flat to below 20Hz, but my query still at what volume.

A simple SPL trace based on a 100db or so average would be welcome.

There will be more room effect than speaker for the measured bass response, even being a dipole bottom end. -Unless Danny goes to extremes to measure it by taking it outside or something...

[Danny Richie]

If Danny is a measuring mood, I'd love to get an objective idea of the bass response. I'm sure it is flat to below 20Hz, but my query still at what volume.

The response is linear regardless of volume.

[rajacat]

Having spent more time than is healthy measuring my OB speakers, I'm not surprised by the 'wriggles'. You simply can't expect the same ruler flat response you get with monopoles with a dipole.

The high treble seems a bit peaky, but in room I'm sure it would be less so.

If Danny is a measuring mood, I'd love to get an objective idea of the bass response. I'm sure it is flat to below 20Hz, but my query still at what volume.

A simple SPL trace based on a 100db or so average would be welcome.

There will be more room effect than speaker for the measured bass response, even being a dipole bottom end. -Unless Danny goes to extremes to measure it by taking it outside or something...

It would seem that an anechoic chamber would be handy or almost a necessity for designing speakers and drivers.

-Roy

[TRADERXFAN]

I think for everything above the schroeder frequency of his room, it's fine. Danny uses that Clio which grabs measurements before reflections. Just for bass, you can't really do that. So that's why he posts frequency response only down to 200hz.

[face]

I found that measuring in a well damped room and with a large bass trap on the floor between the speaker and mic really helps when measuring around and below 200hz.

[Danny Richie]

Been there done that. I had an anechoic chamber 22 feet long.

It did allow good accurate measurements down to about 60 to 80Hz. Below that was a little iffy still as I just can't get far enough away from the speaker with the mic to get concreate levels of accuarcy down there. I could get a really good idea or close approximation to within several db.

Still anything below that range is going to vary depending on the room that they are in.

I can use gated time windows with my Clio and get good accuracy down to 200Hz. And I can back away a little further and get far enough away without reflection to get a resonably good response down to about 100Hz.

Modeling the predicted roll off below 200Hz is pretty easy and tuning can be confirmed with an impedance sweap. So I really don't miss the chamber too much.

[lowtech]

Don't put so much weight into simple charts. The design work Danny has done on some of his models would take volumes of white paper to explain.

 

Well said.

Hallelujah!  Danny should be praised for being so green.  Damn the white papers and save those trees!   

[lowtech]

Having spent more time than is healthy measuring my OB speakers, I'm not surprised by the 'wriggles'. You simply can't expect the same ruler flat response you get with monopoles with a dipole.

Steve, with all due respect, at the wavelengths where the 'wriggles' are seen are too short to be effected by any dipole effects.  What we're seeing is due to diffraction.

Danny, thanks for posting the revised measurements.  They look better.

[Danny Richie]

Again, it is not possible that the wiggles in the top end are caused by diffraction.

Also, those are the same measurements I posted before with the addition of the 10 and 20 degrees off axis responses.

[gprro]

If Danny is a measuring mood, I'd love to get an objective idea of the bass response. I'm sure it is flat to below 20Hz, but my query still at what volume.

The response is linear regardless of volume.

I've been curious about this too. Regardless of volume, or regardless of most sane volumes  ? Love to see a plot for response vs. max input power. For example, the woofer amps puts out how much power into that load? Then plot max output per frequency? If the low end is maxed at 95db (hypothetical #) at 20hz I would need to at least double the woofs.

Which brings me to my second question, and one of the deciders for me between these and the LS9's I think. Given that I can be a volume monster at times, what's the max clean or safe output of the V's tweeter at crossover? There's the only spot I could think of that the 9's could have an advantage...maybe not though, but the tweeter too runs out of excursion at some point.

In the other thread you asked if 105 db's is enough for the Dodd battery amps and the Super V's...no the occasional peaks in the 115-120 range in a large room is enough  ! Hmmm open back line arrays with multiple servo subs...when they comming   Custom open back BG's?

[TRADERXFAN]

Again, it is not possible that the wiggles in the top end are caused by diffraction.

Also, those are the same measurements I posted before with the addition of the 10 and 20 degrees off axis responses.

Do you mean not caused by baffle diffraction?  The horn on the coaxial tweeter is most likely diffracting to some degree.