Suitable midrange cone, for bandpass mid in Unity horn.

JLH · 2010-12-19 4:18 pm

Few notes about what I’ve found to work, and what doesn’t work too well in a Unity style horn.

1.) You can greatly reduce the effects of the mid range entry ports on the compression driver’s frequency response by making careful engineering choices. When I’m designing the crossover points between the mids and compression driver, I leave them very flexible. What you should shoot for is a situation where the following is true : The expansion rate (flare frequency) that the compression driver “sees” must be lower in frequency than the actual crossover frequency. i.e. – If the flare rate is 900Hz, then your crossover frequency must be above 900Hz.

2.) The compression driver’s crossover point dictates where the mids should tap into the horn. Example -If the compression driver’s crossover is 1500Hz, then the mids need to tap into the horn where the cross-sectional area is 41.54 cm2 or LARGER. It’s not about just trying to get the mids as close to the compression driver as possible. Because the compression driver is low passed at 1500Hz, the cross-sectional area where the circumference is equal to 1 wave length – any area larger than this becomes acoustically invisible to the compression driver. This is because the compression driver is cutoff at 1500Hz – thus its output cannot create acoustical pressure against the horn walls when the cross-sectional area is larger than 41.54 cm2. This pretty much removes the effects of the mid entry ports on the compression driver’s frequency response.

3.) The nice thing about Akabak is you can look at the air velocity of the mid ports while adjusting the applied power. I keep the air velocity below 17 meters/sec for good sound quality. Big mid ranges with small entry ports is generally not a good idea. You can run into air non-linearity with surprisingly low powers. If you want to go small on the mid entry ports, then use small mid ranges.

4.) You must juggle the angle of your horn (flare rates) with the crossover frequencies, mid range entry port location, and mid entry port size. I work all four at the same time. It is very difficult to just pick one and then make all the rest fit into your design.

Rgs, JLH

John Sheerin · 2010-12-19 4:26 pm

On air velocity, this also directly relates to the LF crossover point of the mids. If you run big mids but don't run them low or loud, then you don't need big holes because the volume velocity through the holes is low. If you run small mids low and loud, well, it's probably not a good idea to begin with because the drivers will have to move farther and produce more distortion, but the holes will still have to be larger to get the same volume velocity and limit distortion from turbulence due to the air velocity in the holes.

JLH · 2010-12-19 4:34 pm

pk said:
Hi,

I can't help wondering: Why not use the Beyma TPL-150 as tweeter in an all-out unity speaker?

Best regards
Peter

The opening is about 2.6cm X 12.65cm. So, our starting area is almost 33cm2. Since it is impossible to place the mid range entry ports right at the opening of the TPL-150 let’s say we can get within 5cm. With a 60 degree conical horn the area would have expanded to 118 cm2. This means the crossover for the TPL-150 would have to be 890Hz or lower – which is too low for it. This is one of the reasons why 1” compression drivers are used. The smaller the starting area, the easier it is to place the mid ranges at a practical location.

Rgs, JLH

JLH · 2010-12-19 4:35 pm

John Sheerin said:
On air velocity, this also directly relates to the LF crossover point of the mids. If you run big mids but don't run them low or loud, then you don't need big holes because the volume velocity through the holes is low. If you run small mids low and loud, well, it's probably not a good idea to begin with because the drivers will have to move farther and produce more distortion, but the holes will still have to be larger to get the same volume velocity and limit distortion from turbulence due to the air velocity in the holes.

Yes, exactly. Its always a balancing act. You spelled out some points that I keep in my head, but just failed to mention.

Rgs, JLH

SunRa · 2010-12-19 4:45 pm

The points raised are very interesting, thanks for this info.

1.) You can greatly reduce the effects of the mid range entry ports on the compression driver’s frequency response by making careful engineering choices. When I’m designing the crossover points between the mids and compression driver, I leave them very flexible. What you should shoot for is a situation where the following is true : The expansion rate (flare frequency) that the compression driver “sees” must be lower in frequency than the actual crossover frequency. i.e. – If the flare rate is 900Hz, then your crossover frequency must be above 900Hz.

I would have a question, I am not sure what "flare frequency" means. Is it referring to the frequency at which the horn loads the CD? For example a 15" oblate spheroid waveguide (or conical horn probably) will have a knee frequency of about 700Hz. The advice would be then to make sure the crossover frequency is above that point? I am not sure I got it.

Patrick Bateman · 2010-12-19 7:58 pm

JLH said:
Few notes about what I’ve found to work, and what doesn’t work too well in a Unity style horn.

<snip>

2.) The compression driver’s crossover point dictates where the mids should tap into the horn. Example -If the compression driver’s crossover is 1500Hz, then the mids need to tap into the horn where the cross-sectional area is 41.54 cm2 or LARGER. It’s not about just trying to get the mids as close to the compression driver as possible. Because the compression driver is low passed at 1500Hz, the cross-sectional area where the circumference is equal to 1 wave length – any area larger than this becomes acoustically invisible to the compression driver. This is because the compression driver is cutoff at 1500Hz – thus its output cannot create acoustical pressure against the horn walls when the cross-sectional area is larger than 41.54 cm2. This pretty much removes the effects of the mid entry ports on the compression driver’s frequency response.

I wonder if this is why the original Unity horn was sixty degrees, but the first synergy horn was fifty degrees? In other words, a horn with a narrower angle would allow one to locate the midrange holes further from the throat.

Patrick Bateman · 2010-12-19 8:02 pm

JLH said:
Few notes about what I’ve found to work, and what doesn’t work too well in a Unity style horn.

Also - I've built five Unity horns now, John's built at least three, and Cowan has built at least two. It's time for you to make some sawdust!

JLH · 2010-12-20 3:07 am

SunRa said:
The points raised are very interesting, thanks for this info.

I would have a question, I am not sure what "flare frequency" means. Is it referring to the frequency at which the horn loads the CD? For example a 15" oblate spheroid waveguide (or conical horn probably) will have a knee frequency of about 700Hz. The advice would be then to make sure the crossover frequency is above that point? I am not sure I got it.

Start by reading Tom Danley's white paper on tapped horns.

http://www.danleysoundlabs.com/pdf/danley_tapped.pdf

On page 4, it explains why conical horns have a variable flare rate. At the throat the flare rate is high, as you move to the mouth the flare rate lowers. The local expansion rate determines the flare rate. To find the local flare rate you can use Horn Response. Input the values for your conical horn and then change the horn type from Conical to Exponential and Horn Response will calculate the flare rate for you. What the compression driver "sees" is the throat and the distance required to double in area.

Read the Synergy Horn patent to get a better understanding of what I'm talking about.

http://www.goodsoundclub.com/PDF/Synergy_Patent.pdf

I'm not Tom, but if I were a betting man, I'd say Yes to Patrick's assumption as to why the Unity was 60 degrees and the Synergy horns tend to run more narrow angles.

I've made enough sawdust. You just need to use a bigger stick to get it out of me.

Before I release my verison, I want to make sure everything is available and people can build their own. No sense in designing an off the wall one of a kind that will never see the light of day. That really doesn't help anyone here.

Rgs, JLH

dwk123 · 2010-12-20 4:58 am

Thought I'd throw out an interest check.......

I think I'm getting close to admitting that my Yorkville U15s are unlikely to make it back into a functioning system as long as we're in our current house. I had them sounding spectacular with a PC based system, but then the PC died and I lost the EQ filters, and the room they were in got overrun with other requirements. I haven't played them in ~2 years, and it's not looking like I'll get back to it in the foreseeable future.

I know most here are diyers, but is anyone interested in either the U15's or else just the Unity horn assembly and xover? I have a number that I think I'd have to get for the full speakers if I sold them, but since I'm not sure of the market value of the 18Sound 15's I'm not entirely sure what I'd need to get for the horn assembly by itself.

I'm not committed to selling at this point, but if the price is right they'd be better put to use by someone rather than sitting collecting dust.

pk · 2010-12-20 8:19 am

JLH said:
The opening is about 2.6cm X 12.65cm. So, our starting area is almost 33cm2. Since it is impossible to place the mid range entry ports right at the opening of the TPL-150 let’s say we can get within 5cm. With a 60 degree conical horn the area would have expanded to 118 cm2. This means the crossover for the TPL-150 would have to be 890Hz or lower – which is too low for it. This is one of the reasons why 1” compression drivers are used. The smaller the starting area, the easier it is to place the mid ranges at a practical location.

Rgs, JLH

Hi JLH,

Thanks a lot for the explanation!

Best regards
Peter

John Sheerin · 2010-12-20 4:05 pm

Patrick Bateman said:
Also - I've built five Unity horns now, John's built at least three, and Cowan has built at least two. It's time for you to make some sawdust!

I've built more like 8, actually, but some of those were the same horn body completely redone for different purposes. I've been enjoying the one I just set up in the garage. I don't think I posted pics, so here are a few. The bass cabinet is 4x Eminence HL10's in a vented cabinet. It's flat to ~25hz in half space with a bit of eq to flatten it out.

Setup pointing at the ground getting ready for a ground plane measurement: