SynPA Project: midrange port advice for 6" Eminence

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hey guys. I'm kinda falling asleep here. I've been staying up late for the past week reading tons of info on the synergy and unity DIYs.

I'm not an experienced builder, but damn handy. I've started working out some drawings of what I want to do.

Drivers:
BMS 4555 - 1.5-inch compression driver
Eminence LA6-CBMR - 6-inch sealed midrange (x2)

So here's my drawing of the side view. This is not going to be a large horn. More compact but still for PA use. My trouble is the midrange aperture size and location. I can't seem to pinpoint how others have determined these parameters. I've just finished reading the Synergy Tripp post and still can't seem to find the info I need. From what I understand this is the most crucial part of the build. Can someone take a look at my drawing and give me some advice so I can guesstimate? Also is this a crucial measurement that needs to feed into horn response?
An externally hosted image should be here but it was not working when we last tested it.
 
The way to determine the aperture is to model it in Horn Response. You don't really need to model the tweeter, just know how low it can be used.

The midrange you want to model to optimize:

1) the distance of the ports from the throat of the horn (actually, from the diaphragm of the tweeter, but close enough). If this distance is too far from the throat, the midrange won't be able to reach the lowest frequency of the tweeter. If it is too close, you might not get much loading of the midrange by the horn.
2) the cross sectional area of the ports
3) the length of the ports through the wall.
2 and 3) can be adjusted some by making the holes conical shaped ("frustrums") if needed to push the midrange higher in frequency after you build it.​
4) VERY IMPORTANT: the sealed volume Vrc behind the midranges/woofers, assuming the horn is large enough to do anything at low frequencies.
5) The size of the horn mouth will determine how low it might be able to act like a horn (for efficiency or directivity control). To get down to under 500Hz, plan on at least a 24" wide horn. You can have woofers provide output below that, into the bass even, but the horn won't be anything more than a baffle that holds the woofer located close to the tweeter in that case.

Play with the values above till you get the midrange covering the range you need it to.

You might take a look at my SynergyCalc spreadsheet (and the pdf file directions for it, see my signature below) for help in modeling the horn and setting up HornResponse to find out how it will behave. It isn't really very hard to do. If you are only going to have a single-expansion horn as shown, just set the value in Cell C28 to 1.0. The spreadsheet will even calculate the board cuts and angles for you. If single expansion only, you can ignore panel types 3 and 4, and make angles y and w whatever you want (90 degrees is convenient).
 
Last edited:
Ooop, I see that you will be using a sealed midrange. In that case, when you model with horn response, enter the drivers' values into but with a very large volume (Vrc) behind the midrange (since you don't want the model to change the fixed compliance provided by the midrange's already-sealed back).
 
Downloaded the SynergyCalc spreadsheet. Thank you for making this tool! :)

Since this is a midrange horn I played around with adjusting the lowest frequency for horizontal pattern control value until it closely matched the dimensions of my horn. Which is pretty small and still not exact, but close. Made other adjustments and went to Horn Response. There are still some parameters I don't get. Vrc is difficult with a sealed midrange. I can try and calculate the volume of the sealed basket if that would help. Here i put 600 for each driver which comes to 1200. If I measure the volume of the sealed area it would be about 1.56 liters. For Vtc I put 0. I'm still not sure how I'll get that value. Is this the volume of the driver cone? Here's a plot using the values I have.
An externally hosted image should be here but it was not working when we last tested it.
 
I just realized that the above values for Ap1 and Lp are incorrect. I've also made some corrections to Vrc and Lrc and adjusted for the closed-back driver.

Ap1 = Offset driver throat chamber port cross-sectional area (sq cm)

@Bill I've read thru your spreadsheet and PDF a few times, but I'm still not getting it. Can you give some more insight as to how I can get an accurate value for Ap1?
 
Since this is a midrange horn I played around with adjusting the lowest frequency for horizontal pattern control value until it closely matched the dimensions of my horn. Which is pretty small and still not exact, but close. Made other adjustments and went to Horn Response.

Suggest you use the Hornresp Multiple Entry Horn Loudspeaker Wizard tool rather than the OD option. That way you can see the total system response not just the MF response.
 

Attachments

  • Attach_1.png
    Attach_1.png
    50.5 KB · Views: 283
You will also need to take into account the acoustic path length of the compression driver.

For the 4555 it is 6.46cm and that number is directly from BMS themselves so it is accurate.

That means that you will effectively need to add 6.46cm to the con12 value.
I have simmed something similar and you will have to cross very low to the compression driver probably around 800Hz.

For Ap1 Area=pi x (r squared) 1/2 inch port = 0.635cm radius Area = 1.27cm squared

For two 1/2 inch ports Ap1 will be 2.54
 
Last edited:
Paul spencer did some hornrep simulation of the sealed pyle pdmr5 driver where he used a rear chamber of zero which is the opposite of what bill suggested.

His simulation measured fairly closely to the prototype so I would be inclined to keep the rear chamber small.

Some information here if you haven't seen it
DIY Synergy horn - AVS Forum | Home Theater Discussions And Reviews

As you are intending to use this for a PA output will be important. 60x60 horns usually allow the mid drivers to have more output lower down, the wider you make the horn pattern horizontally the more drivers are needed to keep the same output.

If you haven't already got the drivers then 4 x pyle pdmr5 will get you to over 120dB each horn and cost very little. Being only 5 inch will make it easier to fit them closer to the throat and the smaller volume of air under the cone will likely allow the crossover to be closer to 1K.
 
@David
Thanks for that suggestion. I will try that today.

@fluid
I'm kinda going at this backwards. For PA the size, weight and other factors are important and so I'm trying to build a horn in a predefined space. I'd like to experiment with an arrangement similar to the Yorkville U15. My horn size is roughly 15" wide by 10" high (90x45). Eh not very large. Using two of the 6-inch Eminence drivers seemed like the way to go. I've also considered the Celestion 4-inch drivers. Your suggestions are helpful.
 
If you are able to get hold of the 4" Celestion that Tom uses in the real Synergies then that is the best driver to use.

The problem with the closed back drivers is it is hard to make them go low. When you make them larger like the six inch versions it is also then harder to make them play high enough, which is not a great combination.

Art's version finds a nice middle ground with a big CD that is happy to cross quite low and a woofer that is made to cross quite high by the use of the volume plug.

The extra path length in the BMS drivers makes the reflection notch lower. The 1" exit 4550 is probably the best option as it has the most xmax and is happier crossing below 1kHz.

The 4550 and four or more of the 4" Celestion's would make a very good combination. There is a reason why this combination in used by Danley.
 
Port size, length and Vtc act together to set the upper limit of the bandpass. Less Vtc and smaller port all tend to make for a higher crossover.

With port diameter this has an effect on the compression ratio the driver is firing into. The smaller the port the higher the compression ratio and the more strain and pressure you put on the driver. So consider the ratio of Atc to Ap1 and keep it under 10:1 if possible.

Bill has used 2 x 7/16" port holes in his small syns
http://www.diyaudio.com/forums/multi-way/292379-small-syns-10.html#post4788726

That gives an Ap1 of 1.9cm squared for each driver.

Vtc for a bare driver can be approximated with a truncated cone calculation if you can get the dimensions you need from the driver. If you put a rough number into hornresp then you can see if you will need to fill in some volume to get the response you want.

A few people have used this driver in various model so if you look around you might find a suitable figure to use.
 
The reasoning for making Vrc very large is that a limited Vrc effectively tightens the compliance of the driver (like making the spring more stiff). You ALREADY KNOW the compliance of the sealed driver, it's right there on the Eminence data sheet, you want the simulator to use that. You don't have the option (or desire) to make the suspension more stiff than Eminence already tells you it is. So, to make it not reduce the compliance.... you can make the volume "behind the cone" (if it were open back, though it's not) large enough to not make it any stiffer. Make sense?

(There might be another way to make the sealed volume's effect go away, but I did have good match to simulation doing it this way. David McBean, is there another approach?).

Does Tom use Celestion drivers? I thought he used Misco midrange drivers?

A good way to figure the "volume inside the speaker cone" (at front) is to fill one with rice and pour it off into a measuring cup!
 
The reasoning for making Vrc very large is that a limited Vrc effectively tightens the compliance of the driver (like making the spring more stiff). You ALREADY KNOW the compliance of the sealed driver, it's right there on the Eminence data sheet, you want the simulator to use that. You don't have the option (or desire) to make the suspension more stiff than Eminence already tells you it is. So, to make it not reduce the compliance.... you can make the volume "behind the cone" (if it were open back, though it's not) large enough to not make it any stiffer. Make sense?

My logic would have been that by leaving it at zero you are using the parameter as is. I see the logic behind the idea of removing it's effect by making it really big. I haven't measured either way so I don't want to argue the point with you!

Tom use Celestion drivers? I thought he used Misco midrange drivers?

The misco was used in the unity, the celestion 4" in the SH50, seems to be confirmed from pictures of the inside of both.
 
The reasoning for making Vrc very large is that a limited Vrc effectively tightens the compliance of the driver (like making the spring more stiff). You ALREADY KNOW the compliance of the sealed driver, it's right there on the Eminence data sheet, you want the simulator to use that. You don't have the option (or desire) to make the suspension more stiff than Eminence already tells you it is. So, to make it not reduce the compliance.... you can make the volume "behind the cone" (if it were open back, though it's not) large enough to not make it any stiffer. Make sense?

(There might be another way to make the sealed volume's effect go away, but I did have good match to simulation doing it this way. David McBean, is there another approach?).

Setting Vrc and/or Lrc to zero means that no rear chamber has been specified (open back). The rear side of the driver diaphragm radiates into the specified Ang space, which may or may not be a problem depending upon the system being simulated.

Making Vrc very large and Lrc small is the best way to remove the sealed volume's effect without having to contend with rear radiation.
 
I've been reworking my plans using the Celestion 4-inch drivers. These babies take up some room when you have 4. So redrawing is necessary. Will post later today.

Include a very small throat chamber in front of the compression driver diaphragm and use Ap1 and Lp in the Nd record to specify the compression driver exit path rather than adding the length to L12.

David, maybe some of us have been inputting values incorrectly? To make things a little more clear, especially for me, I thought Ap1 had everything to do with the port for the midrange driver. If we're to compensate for the compression driver, how do we estimate the additional throat chamber value and where is that value added in the parameters. Yes I am confused.

fluid mentions the acoustic path length needs to be accounted for and that value comes from the manufacturer. Not something I could find on the spec sheet. Do I need to request this from BMS?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.