alternative designs- why go jubilee?

[bodcaw boy]

you guys are close.....hint: look at the curve miketn posted.......

boy

The first flare rate for the Jubilee if I remember is 97Hz I believe so 97Hz x 2 =194Hz and 97Hz x 3 = 291Hz which seems to approximately correspond with some of the peaks we are seeing.

ROY, Am I Cold or Hot with this guess?

mike tn[]

cold....sorry... but you were on to something before when you talked about the active eq settings.

jc (water pipes guy!)

in the full anechoic measurement, ie no boundaries, the vertical flare of the lf horn is not expanding; all expanding is done horizontally.....thus when the wave(think of a bubble) exits the horns, horizontally the wave is okay but in the vertical it goes from no expansion to lots of expansion; so the wave diffracts. in a corner and on the floor one side of the horn in the vertical continues the no expansion but at the top it still diffracts so it is not as bad as the full anechoic curve. that is the dip you see around 200 hz.

some the waviness of the bottom end is the result of not enough mouth and possibly not enough back air chamber (think sealed systems) to support the wavelength.

path length differences in a continously expanding area horn determines the top end limit and will notch and peak harmonically and will never return to a smooth curve.

then another acoustic anomoly is the bifrication of the horn, ie splitting the horn into two mouths. the distance between those two mouths and the angle of splay between those two mouths also add some waviness as the wave goes from supported to not so supported because of the two acoustic sources. it manifests itself in the freq response as a harmonically related peak and dip whose amplitude is dependent on the splay angle. (polar plots should this anomaly very well)

if the horn is continously expanding smoothly and you subtract the effects described above you should see the freq response of the woofer as seen in a sealed box configuration (without the horn; just back air chamber).

trying adding some thick acoustic foam to the front panel and see if that helps to tame some of the peaks and dips. we also did some experiments by adding a nose piece (like the la scala) to make the two mouths mate up as one but it also requires that you extend the top panel out to meet the tip of the nose piece.

or can just add some active eq like miketn and coyotee-o have on their jubs. []

boy!!

[mikebse2a3]

you guys are close.....hint: look at the curve miketn posted.......

boy

The first flare rate for the Jubilee if I remember is 97Hz I believe so 97Hz x 2 =194Hz and 97Hz x 3 = 291Hz which seems to approximately correspond with some of the peaks we are seeing.

ROY, Am I Cold or Hot with this guess?

mike tn[]

cold....sorry... but you were on to something before when you talked about the active eq settings.

jc (water pipes guy!)

in the full anechoic measurement, ie no boundaries, the vertical flare of the lf horn is not expanding; all expanding is done horizontally.....thus when the wave(think of a bubble) exits the horns, horizontally the wave is okay but in the vertical it goes from no expansion to lots of expansion; so the wave diffracts. in a corner and on the floor one side of the horn in the vertical continues the no expansion but at the top it still diffracts so it is not as bad as the full anechoic curve. that is the dip you see around 200 hz.

some the waviness of the bottom end is the result of not enough mouth and possibly not enough back air chamber (think sealed systems) to support the wavelength.

path length differences in a continously expanding area horn determines the top end limit and will notch and peak harmonically and will never return to a smooth curve.

then another acoustic anomoly is the bifrication of the horn, ie splitting the horn into two mouths. the distance between those two mouths and the angle of splay between those two mouths also add some waviness as the wave goes from supported to not so supported because of the two acoustic sources. it manifests itself in the freq response as a harmonically related peak and dip whose amplitude is dependent on the splay angle. (polar plots should this anomaly very well)

if the horn is continously expanding smoothly and you subtract the effects described above you should see the freq response of the woofer as seen in a sealed box configuration (without the horn; just back air chamber).

trying adding some thick acoustic foam to the front panel and see if that helps to tame some of the peaks and dips. we also did some experiments by adding a nose piece (like the la scala) to make the two mouths mate up as one but it also requires that you extend the top panel out to meet the tip of the nose piece.

or can just add some active eq like miketn and coyotee-o have on their jubs. []

boy!!

Thanks for the lesson "Teacher Roy"[]

mike tn

[chops]

So no input about open baffles?! [:|]

[bodcaw boy]

Speaking of alternative designs, has anyone ventured into doing an open baffle design at all with dual 15's per channel?

With the right drivers, one can easily match the sensitivity ( <105dB 1w/1m) of the squawker, get excellent, clean midrange up to 600Hz and natural, ultra-fast and dynamic bass to an extent before implementing active EQ'ing to compensate for the dipole roll-off in the lowest octaves.

This is exactly what I have in the works at the moment. I'm doing some research just to "freshen up" before jumping in head first. This ought to be some fun times coming up soon! []

what kind of directivity (coverage angle) are you going to go with?

[Guest " "]

jwcullison

So the drone is back in the picture.

Saw your graph.

Do you think 2 drivers and a drone is the best of both worlds (dip correction as well as bandwidth to 1khz)?

[chops]

Speaking of alternative designs, has anyone ventured into doing an open baffle design at all with dual 15's per channel?

With the right drivers, one can easily match the sensitivity ( <105dB 1w/1m) of the squawker, get excellent, clean midrange up to 600Hz and natural, ultra-fast and dynamic bass to an extent before implementing active EQ'ing to compensate for the dipole roll-off in the lowest octaves.

This is exactly what I have in the works at the moment. I'm doing some research just to "freshen up" before jumping in head first. This ought to be some fun times coming up soon! []

what kind of directivity (coverage angle) are you going to go with?

I'm not sure I follow. If you're referring to the "sweet spot", dipole loudspeakers generally have a broader area rather than just a "spot".

[jwc]

Roy. Wow. Unbelievably helpful. Thanks.

I thought of the "nose" as you call it and bring the side walls out slightly further. Possibly to make the physical horn 10-12 cm longer. However, the nose has a very rapid flare unless you bring the side panels of the last turn more inward. This would work but then the side panel are moving away from the walls of the room and then the concern of the "non-physical horn" mouth being affected. I'm not too concerned about the loss of high end as the horn really is doing well out there so I guess the shortening concern shouldn't be an issue.

The Dips or peaks aren't really that big of a concern, the point was brought up out of principle. Educational...thought I would take a stab at the possibility of a certain fold of the horn being a potential issue. Sure you would have addressed it if so.

I was thinking the back chamber volume was right on......mine came out to be 37.7 liters per driver. Mine intentionally slightly more as I think the with of my baffle is .25 (.125/side) inches wider than the real deal.....I assume from calculation. Not sure how many liters the drivers take up and how that comes into play with the calculated needed Vb.

You made a suggestion that you have something up your sleeve ....upcoming option for the jub maybe????? I know you can't answer.

If the drone was there....oooohhhhh. I'm totally lost in this department. I bought a drone just to put as the bottom driver to see what it will do as for tuning the cabinet. An adjustable drone in which you can add weight to help tune. Oh my....I look like an idiot....went from a naked drone to one with loads of metal washers.....RTA curve looks the same every time. You would have loved to have been a third party to hear in on the conversation I had with partsexpress for advice on how to tune this drone in this test cabinet. That was a goooood one. I guess I would have to get fancy and get an impedance curve of some how to tune it correctly. Oh well. Hope you get the drone working in real jub one day.......I think that project hasn't died as the "home Jub" as you said from Hope..........................under that sleeve isn't a pack of cigarettes is it?

jc

[jwc]

jwcullison

So the drone is back in the picture.

Saw your graph.

Do you think 2 drivers and a drone is the best of both worlds (dip correction as well as bandwidth to 1khz)?

That is hard to answer. My original idea was to do the two active with one drone for a second test box....but the two actives did better than I anticipated. I have the cuts made for the second jub cabinet just like the first.

Now the actual subjective sound of the drone with no weight is pretty good. I've lost some SPL......I would stabe at a 101-102 dB maybe.

Yes it is interesting how those dips aren't as noticeable with the one driver-one drone.

[mikebse2a3]

path length differences in a continously expanding area horn determines the top end limit and will notch and peak harmonically and will never return to a smooth curve.

------------------------------------------------------------------------

Roy can you explain this in more detail? Exactly which path length differences are you referring to?

mike tn[]

[bodcaw boy]

Speaking of alternative designs, has anyone ventured into doing an open baffle design at all with dual 15's per channel?

With the right drivers, one can easily match the sensitivity ( <105dB 1w/1m) of the squawker, get excellent, clean midrange up to 600Hz and natural, ultra-fast and dynamic bass to an extent before implementing active EQ'ing to compensate for the dipole roll-off in the lowest octaves.

This is exactly what I have in the works at the moment. I'm doing some research just to "freshen up" before jumping in head first. This ought to be some fun times coming up soon! []

what kind of directivity (coverage angle) are you going to go with?

I'm not sure I follow. If you're referring to the "sweet spot", dipole loudspeakers generally have a broader area rather than just a "spot".

well every direct radiator has a characteristic coverage pattern. usually about from low freq to high, it starts off wide dispersion that narrows down to about 120 degree for a certain bandwidth then beams down to about 90 degrees over a certain bandwidth and then starts to collapse. where it starts to collapse is determined by the voice coil diameter. so say you are going to use 15". it is 120 degree from about 300 to about 600 then 90 degrees from 600 to about 1400 hz for a 3" voice coil. this is all approximate since i have slept since then. putting the woofers one on top of another with literally not much space between the frames means that the woofers will create a bubble (d'appolito) that resembles a coverage pattern of about 90 to 70 degrees at about 900 hz. so you could essentially have 90 degrees to about 300 hz.

now take a cabinet without a back and whose sides, top and bottom are about 40" deep and you should be able to start dipole action about 300-400 hz. now you could extend the 90 degree coverage pattern well below 100 hz as long as you can keep the amplitude up.

is that what you meant about open baffle? or are you going to mount dual 15" on the back as well to make a dipole in an box?

boy!!

[DrWho]

in the full anechoic measurement, ie no boundaries, the vertical flare of the lf horn is not expanding; all expanding is done horizontally.....thus when the wave(think of a bubble) exits the horns, horizontally the wave is okay but in the vertical it goes from no expansion to lots of expansion; so the wave diffracts. in a corner and on the floor one side of the horn in the vertical continues the no expansion but at the top it still diffracts so it is not as bad as the full anechoic curve. that is the dip you see around 200 hz.

trying adding some thick acoustic foam to the front panel and see if that helps to tame some of the peaks and dips. we also did some experiments by adding a nose piece (like the la scala) to make the two mouths mate up as one but it also requires that you extend the top panel out to meet the tip of the nose piece.

Hmmm, so why does it happen more at 200Hz than at any of the other frequencies?

[chops]

edit...

[bodcaw boy]

path length differences in a continously expanding area horn determines the top end limit and will notch and peak harmonically and will never return to a smooth curve.

------------------------------------------------------------------------

Roy can you explain this in more detail? Exactly which path length differences are you referring to?

mike tn[]

sure let me try. let's say you are driving down a street and your tires are leaving marks where they have tread. go ten feet and measure both the right and left tire tread marks. they should both be about 10 feet. 10 feet of the right - 10 feet of the left = 0 feet= 0 feet path length difference. now go another ten or so feet but this time make a left 45 degree left turn. stop and get out and measure the left and right tire tracks. the left tire track will be shorter than the right tire track. subtract the two and you have a path length difference. that path length difference divided by 1132 feet /sec and then inverted and divided by two will give you the freq at which the wave traveling down the horn will wrap on itself harmonically and go in and out of phase with itself. this causes peak and nulls in the freq response and the wave will never recover and become flat again.

hope that helped.

boy!!

[chops]

Speaking of alternative designs, has anyone ventured into doing an open baffle design at all with dual 15's per channel?

With the right drivers, one can easily match the sensitivity ( <105dB 1w/1m) of the squawker, get excellent, clean midrange up to 600Hz and natural, ultra-fast and dynamic bass to an extent before implementing active EQ'ing to compensate for the dipole roll-off in the lowest octaves.

This is exactly what I have in the works at the moment. I'm doing some research just to "freshen up" before jumping in head first. This ought to be some fun times coming up soon! []

what kind of directivity (coverage angle) are you going to go with?

I'm not sure I follow. If you're referring to the "sweet spot", dipole loudspeakers generally have a broader area rather than just a "spot".

well every direct radiator has a characteristic coverage pattern. usually about from low freq to high, it starts off wide dispersion that narrows down to about 120 degree for a certain bandwidth then beams down to about 90 degrees over a certain bandwidth and then starts to collapse. where it starts to collapse is determined by the voice coil diameter. so say you are going to use 15". it is 120 degree from about 300 to about 600 then 90 degrees from 600 to about 1400 hz for a 3" voice coil. this is all approximate since i have slept since then. putting the woofers one on top of another with literally not much space between the frames means that the woofers will create a bubble (d'appolito) that resembles a coverage pattern of about 90 to 70 degrees at about 900 hz. so you could essentially have 90 degrees to about 300 hz.

now take a cabinet without a back and whose sides, top and bottom are about 40" deep and you should be able to start dipole action about 300-400 hz. now you could extend the 90 degree coverage pattern well below 100 hz as long as you can keep the amplitude up.

is that what you meant about open baffle? or are you going to mount dual 15" on the back as well to make a dipole in an box?

boy!!

Hi Roy,

This is what I mean by dipole (open baffle)...

The drivers being used will most likely have a 2.5" VC. They will be crossed over at 600Hz to the 511B/902-8B and will be augmented on the bottom by an EQ to compensate for the dipole roll-off.

[jwc]

So Roy, I could make a folded bass horn in which the top and bottom of the horn is constant and the two sides are "bent plywood" throughout to the perfect expansion and I could still get the phase "in and out" you describe because of the different wall lengths? If so........dissapointing reality for the folded bass horn concept and loss of the high end.

Guess that further proves the genius of the jubilee's great performance on the "high end"

[bodcaw boy]

in the full anechoic measurement, ie no boundaries, the vertical flare of the lf horn is not expanding; all expanding is done horizontally.....thus when the wave(think of a bubble) exits the horns, horizontally the wave is okay but in the vertical it goes from no expansion to lots of expansion; so the wave diffracts. in a corner and on the floor one side of the horn in the vertical continues the no expansion but at the top it still diffracts so it is not as bad as the full anechoic curve. that is the dip you see around 200 hz.

trying adding some thick acoustic foam to the front panel and see if that helps to tame some of the peaks and dips. we also did some experiments by adding a nose piece (like the la scala) to make the two mouths mate up as one but it also requires that you extend the top panel out to meet the tip of the nose piece.

Hmmm, so why does it happen more at 200Hz than at any of the other frequencies?

because it can........

no just joking doc, because that just happens to correspond to the inside height of the jub lf horn (half wavelength) and that causes the first reflection back in the horn. now it does ripple out but with diminishing amplitude and affect at other freqs.

[rigma]

Roy,

Can you clarify where you would add the foam, are you talking about on the front that faces into the room?

Rigma

[bodcaw boy]

Speaking of alternative designs, has anyone ventured into doing an open baffle design at all with dual 15's per channel?

With the right drivers, one can easily match the sensitivity ( <105dB 1w/1m) of the squawker, get excellent, clean midrange up to 600Hz and natural, ultra-fast and dynamic bass to an extent before implementing active EQ'ing to compensate for the dipole roll-off in the lowest octaves.

This is exactly what I have in the works at the moment. I'm doing some research just to "freshen up" before jumping in head first. This ought to be some fun times coming up soon! []

what kind of directivity (coverage angle) are you going to go with?

I'm not sure I follow. If you're referring to the "sweet spot", dipole loudspeakers generally have a broader area rather than just a "spot".

well every direct radiator has a characteristic coverage pattern. usually about from low freq to high, it starts off wide dispersion that narrows down to about 120 degree for a certain bandwidth then beams down to about 90 degrees over a certain bandwidth and then starts to collapse. where it starts to collapse is determined by the voice coil diameter. so say you are going to use 15". it is 120 degree from about 300 to about 600 then 90 degrees from 600 to about 1400 hz for a 3" voice coil. this is all approximate since i have slept since then. putting the woofers one on top of another with literally not much space between the frames means that the woofers will create a bubble (d'appolito) that resembles a coverage pattern of about 90 to 70 degrees at about 900 hz. so you could essentially have 90 degrees to about 300 hz.

now take a cabinet without a back and whose sides, top and bottom are about 40" deep and you should be able to start dipole action about 300-400 hz. now you could extend the 90 degree coverage pattern well below 100 hz as long as you can keep the amplitude up.

is that what you meant about open baffle? or are you going to mount dual 15" on the back as well to make a dipole in an box?

boy!!

Hi Roy,

This is what I mean by dipole (open baffle)...

The drivers being used will most likely have a 2.5" VC. They will be crossed over at 600Hz to the 511B/902-8B and will be augmented on the bottom by an EQ to compensate for the dipole roll-off.

it is open back... okay here is my guess with the yellow speaker above. the woof coverage pattern is probably the equivalent of 50 degrees vertical from about 900 hz down to about 500 or 400 Hz and in the horizontal, the dipole effect will probably start at about 400 hz and get to 90 degrees. do you know how much boost will be required at about 50 Hz? just curious because of the amount of displacement the woofers will have to experience.

boy!!

[DrWho]

Hmmm, how observably simple [A]

Thanks Boy! []

[D-MAN]

As I said previously, add to that the possibility of the frame reflection at 200-250 and there you go! Attached is an example.

It is a far different matter when listening to music. You will never hear the dips in reality - or perhaps better stated as - you will never recognize the dips as being there!

To take it a step further, a totally flat response would probably be "boring" - technically correct, but dissatisfying.

I think PWK was right with "dirty curves". They give "life" to the performance.

As to altering it in the Jubilee, add some absorptive material to the back chamber behind the driver to suppress reflections from coming back through the cone.

Add some fibrous absorptive material to the bottom of the back chamber to suppress standing waves from developing between the top and bottom (the height of the enclosure) being the only parallel walls in the enclosure (from submultiples of frequency wavelength that would fit between the two parallel surfaces).

Lastly add "wings" to the outside of the mouth channels, ala "Little Bast@rd" to lessen difractive effects and slightly increase your LF response.

DM