~20hz Horn Help

[mattmanic]

Hello,

I'm new here but I spend a lot of my time with car audio. Technically this is car audio but as you don't get a lot of big horns in cars I thought it'd be more appropriate to ask this here.

I'm looking to build a new enclosure for 2 Dual 2ohm Fi Bl 12" subs ( BL12 - Fi Car Audio Store )

I'm after maximum efficiency and just to do something different. As I compete in various competitions, SPL is important to me but this is my car too so I want it to sound nice. At the moment I can average 142dB SPL from 20-40hz on 3kw in a 3.5 cubic foot ported box tuned to 30hz. I'm looking to gain at least another 3db from 20-40hz ideally and I'm not bothered about playing anything above 60hz.


A tapped horn was suggested to me by a friend and so I started trying things out in hornresp but I'm struggling to get a flat response with good efficiency down to 20hz or just below. Or maybe I am but I don't know how efficient my current enclosure is to compare it to.


The constraints I'm working to are:

Must be less than 10 cubic feet before displacements for wood, bracing, etc.

Must play 20 - 40hz reasonably flat

Must have 3kw power handling give or take 500w

Must be louder than my current box

Must be foldable in a way in which it will fit in the boot of a hatchback, albeit up to the roof.


Here are some of the designs I've put through hornresp so far:

200L


ID=29.00
Ang=0.5 x Pi
Eg=2.83
Rg=0.00
Fta=2.07
S1=176.00
S2=168.00
Con=28.00
F12=0.00
S2=168.00
S3=328.00
Con=337.90
F23=0.00
S3=328.00
S4=874.00
Con=178.60
F34=0.00
S4=0.00
S5=0.00
L45=0.00
F45=0.00
Sd=480.00
Bl=20.90
Cms=7.54E-03
Rms=11.24
Mmd=212.00
Le=8.60
Re=1.40
TH=2P
Vrc=0.00
Lrc=0.00
Ap1=0.00
Lpt=0.00
Vtc=0.00
Atc=0.00
Pmax=3000
Xmax=18.0
Comment=NEW RECORD - Default horn loudspeaker parameters


210L

ID=29.00
Ang=0.5 x Pi
Eg=2.83
Rg=0.00
Fta=4.46
S1=184.00
S2=184.00
Con=28.00
F12=0.00
S2=184.00
S3=269.00
Con=327.80
F23=0.00
S3=269.00
S4=1513.00
Con=162.70
F34=0.00
S4=0.00
S5=0.00
L45=0.00
F45=0.00
Sd=480.00
Bl=20.90
Cms=7.54E-03
Rms=11.24
Mmd=212.00
Le=8.60
Re=1.40
TH=2P
Vrc=0.00
Lrc=0.00
Ap1=0.00
Lpt=0.00
Vtc=0.00
Atc=0.00
Pmax=100
Xmax=5.0
Comment=NEW RECORD - Default horn loudspeaker parameters


250L

ID=29.00
Ang=0.5 x Pi
Eg=2.83
Rg=0.00
Fta=2.41
S1=117.00
S2=134.00
Con=33.80
F12=0.00
S2=134.00
S3=336.00
Con=311.30
F23=0.00
S3=336.00
S4=1277.00
Con=233.30
F34=0.00
S4=0.00
S5=0.00
L45=0.00
F45=0.00
Sd=480.00
Bl=20.90
Cms=7.54E-03
Rms=11.24
Mmd=212.00
Le=8.60
Re=1.40
TH=2P
Vrc=0.00
Lrc=0.00
Ap1=0.00
Lpt=0.00
Vtc=0.00
Atc=0.00
Pmax=100
Xmax=5.0
Comment=NEW RECORD - Default horn loudspeaker parameters


Copying those into a text file and placing them in hornresps import folder will give you my simulations.




Sorry for the essay and thank you for taking the time out of your day(s) to read this. Just let me know if anything needs clarifying or if I'm asking for the moon...



Thanks in advance

Matt

 

[just a guy]

I can take a shot at this but before we begin I have a few questions and concerns.

1. As far as your current setup goes, how are you getting response to 20 hz? You should have a fairly steep high pass filter at or near tuning (30 hz) so you shouldn't be getting anything at 20 hz. If you are not using a steep hpf at or near tuning you are not doing it right.
2. From your Hornresp files I can see that you want flat reponse to 20 hz. You can forget about getting extra output from a horn since you are tuning the horn significantly lower than your current ported box. If you want gains you need to tune to the same frequency or higher.
3. 10 cubic feet isn't very large for a horn for dual 12 inch drivers. I'll see what I can do but still...
4. Do you have measured t/s parameters for your drivers? I can use the published specs but it's not ideal.
5. Absolutely the first thing that needs to happen is a sim of your current box. There's no way to compare without that. Do you have all dimensional info so I can do that?
6. What type of amp are you using? What model? What class is it? What fuse does it use?
7. Do you know how to fold a horn? Folding is by far the most time consuming part and I don't really want to do that.

Get back to me and let me know about these things and I'll see what I can do. I can simulate front loaded horns and all types of transmission line as well as ported and tapped horn so I should be able to come up with something that does what you want as long as your goals are realistic.

 

[sine143]

I would first measure the response of a simple sealed box inside and outside of your car to see what type of cabin gain response your vehicle has.

a horn designed to play "flat" to 20hz in honresp, will likely be a 20 to 50hz monster, lacking upper bass due to cabin gain when installed in a vehicle.

 

[just a guy]

I would first measure the response of a simple sealed box inside and outside of your car to see what type of cabin gain response your vehicle has.

a horn designed to play "flat" to 20hz in honresp, will likely be a 20 to 50hz monster, lacking upper bass due to cabin gain when installed in a vehicle.

I'm assuming he likes the response he has now which is one of the reasons I want to sim the current box. A lot can be inferred from this info.

Your advice is good if flat response is desired but in my experience with car audio, flat response is rarely desired.

 

[mattmanic]

Thanks for the reply.


1. Purely playing below tuning with reduced volume/power and using the air pressure in the car to load the subs, ie. If I open a door too much at close to full volume, the subs will bottom out. I have no hpf on the subs.

2. The response of my current box drops off gradually after tuning. 143 at 40 and 30hz, 141 at 25hz and 137 at 20hz. I was looking to get that response flatter and louder so perhaps with the cabin gain taken into effect, I wouldn't need to tune as low as 20?

3. Thank you, size of ported boxes in cars is normally taken care of by adding more power. I'm not sure whether this is the same for horns

4. I'm afraid I don't, is it simple to measure them?

5. I will measure up in an hour or so.

6. It's a Loudest.com tl1021 3kw rms. They have a very bad reputation but mines been upgraded and actually does about 2900w into a 1ohm load, I have tested it myself with a true rms clamp ammeter and voltmeter. The amp doesn't have any fuses but it's popped a 200a fuse on the cabling before. I now use 2 runs of 0awg and 400a of fusing.

7. As far as I know it's simply a case of keeping the line length the same as the design by following the centre of the path. If there's more to it than that, I'll research and it shouldn't be a problem.



I have 3 8" midbass drivers per front door so meeting the subs at 50 or 60hz isn't an issue

 

[mattmanic]

I am looking for a flat response, as the competition I take part in looks for as high an average spl from 20 - 40hz as possible and I'd also like to be able to listen to music with it turned down a little.

 

[just a guy]

Thanks for the reply.


1. Purely playing below tuning with reduced volume/power and using the air pressure in the car to load the subs, ie. If I open a door too much at close to full volume, the subs will bottom out. I have no hpf on the subs.

2. The response of my current box drops off gradually after tuning. 143 at 40 and 30hz, 141 at 25hz and 137 at 20hz. I was looking to get that response flatter and louder so perhaps with the cabin gain taken into effect, I wouldn't need to tune as low as 20?

Excursion goes up exponentially below tuning so ideally it shouldn't be played below tuning at all. If you are burping a single frequency and it is above tuning you don't really need a hpf at all but if you are truly sweeping 20 - 40 hz as you imply, you definitely should have a 20 hz tuning. You can't rely on cabin gain to reduce excursion requirements since excursion skyrockets right below tuning.

3. Thank you, size of ported boxes in cars is normally taken care of by adding more power. I'm not sure whether this is the same for horns

In both ported boxes and horns it can be difficult to get anything resembling flat response out of an undersized box. You probably have enough room to get decent response with ported boxes but a horn will probably be a challenge. Since you only want a single octave of flat response though, it might be ok. We'll see.

4. I'm afraid I don't, is it simple to measure them?

It's quite simple if you have the equipment but impossible if you don't. You can make a simple jig with $2 of parts and use ARTA or REW to measure or you can measure with a woofer tester (WT2 or WT3) but they are expensive.

5. I will measure up in an hour or so.

This first sim of your current box will tell us a lot.

6. It's a Loudest.com tl1021 3kw rms. They have a very bad reputation but mines been upgraded and actually does about 2900w into a 1ohm load, I have tested it myself with a true rms clamp ammeter and voltmeter. The amp doesn't have any fuses but it's popped a 200a fuse on the cabling before. I now use 2 runs of 0awg and 400a of fusing.

Ok, sounds like you know what you are doing, I'll trust you that you have the power that you say you do. This is not common with car amps though, that's why I asked.

7. As far as I know it's simply a case of keeping the line length the same as the design by following the centre of the path. If there's more to it than that, I'll research and it shouldn't be a problem.

Well... kind of. There are more accurate methods than a simple center line measurement and it can be frustrating trying to make the final product fit into a perfect rectangular shape but if you are willing to try and realize that it might take a while you will probably do fine. Let's see if we can get a decent horn in this small size requirement first.

And it might help if you could post up the rules of your competition. That will give me a more clear idea of what you are actually doing. I don't know if you are burping single frequencies, sweeping frequencies or playing music and if so, how you decide what music to play.

 

[just a guy]

Are you able to measure frequency response? If so, the previous suggestion of measuring a sub in a sealed box inside and then outside the vehicle is a good one since you want flat response.

This is problematic in that response will change dramatically depending on where you put the sealed sub inside the vehicle but it will give some kind of idea what kind of vehicle response you are working with.

Without this info I can just go by your stated measured response and the sim of your current box and/or I can try to look up your specific vehicle to see if anyone else has measured cabin gain.

 

[mattmanic]

Untitled Page

Those are the rules for the competition format, generally the loudest cars will be tuned to 27-30hz but designed not to run xmax too high, either through loading or enclosure size, and the subs have a decent amount of travel anyway. 18mm each way xmax on these iirc but its more than an inch before they hit the back plate.


The current box measures 40 x 16 x 12" with 18mm wood and the port is 2" x 14.5", 19" long. Net volume is 3.2cuft and I've just worked it out, tuning is actually 32hz.

The only accurate measurement data I have is from when I was running it isobaricly with 4 of the same subs. I can only remember rough numbers for its current state, but when in iso, 20hz - 135.9, 25hz - 139.5, 33hz - 141, 40hz - 140.2, peak 35hz - 141.3. A fair bit quieter as I couldn't load the amp low enough with all of them. My amp makes 3k at 2ohms sorry, I was mistaken before.

The car is a Peugeot 206.

I can measure frequency response, I have a termlab sensor as used in competition. If I have time over the weekend I'll build a quick sealed box and take some measurements inside and out of the car.

 

[GM]

Copying those into a text file and placing them in hornresps import folder will give you my simulations.

Greets!

Or you could save them in HR’s ‘export’ folder, then attach the files to your post, so that we could import them to view immediately.

Anyway, got to get back to rearranging leaves for now……………

GM

 

[sine143]

try to take your outside of car measurments in as open of a space as you can get. In the middle of a parking lot with no cars would be ideal.

 

[just a guy]

If I have time over the weekend I'll build a quick sealed box and take some measurements inside and out of the car.

Just use the box you already have and plug the port. That's a lot easier than building a sealed box just for a 2 minute measurement.

I skimmed the rules but I need to read closer to see if there's any loopholes that could be exploited.

I'll get to work on the sims. Should I assume you won't be measuring t/s parameters? The reason I ask is because I have no idea how good this company is at providing accurate t/s parameters. Also, I believe they hand build their products which can lead to inconsistancies. Also, they offer several levels of upgrades, all of which could potentially skew the specs of the finished product, possibly by quite a bit.

 

[just a guy]

One last question before I start. Are you allowed to adjust the volume knob between frequency tests? For example, test 20 hz and then turn it up before running the 33 hz test or is that against the rules?

 

[sine143]

could he not achieve said result by applying eq? lol.

 

[just a guy]

could he not achieve said result by applying eq? lol.

No. What I am talking about is the problem with a 30 hz tuning playing a 20 hz tone. So he plays 20 hz and he is limited by excursion since it's below tuning. Then he plays 33 hz and he is no longer limited by excursion so he could turn the power up dramatically and get a higher spl. But only if it's ok by the rules.

 

[bjorno]

FYI....

b

 

[just a guy]

So here's what the numbers say. Please check all my inputs. It is up to YOU to catch any stupid mistakes I might make.

Shown at 3000 watts into 1.4 ohms end load (dual coils in series and then drivers in parallel) in .5 pi space since this is for a car.

This looks about right once you figure in a bit of cabin gain and smooth out the peak at tuning. That peak will be reduced due to power compression, higher driver q (large vs small signal t/s) and MASSIVE port compression.

I had to export the file, email the file, fire up my virtual machine, receive the email, load it into the Akabak script folder, fire up Akabak, run the sim, and finally check air velocity at the port output.

I knew velocity was going to be high with such a small port but I was expecting to see 110 m/s. This is a huge problem and indicates that your port isn't really acting like a port at 3000 watts. So basically that big spike at tuning won't be there when measured at high power for a bunch of reasons.

We can definitely come up with something better, even a ported box with a bigger port would be much better than this.

EDIT - I noticed the same t/s inconsistency problem bjorno did. It looks like we ended up with the same t/s so I hope this is all correct. It's hard to know what to do when a manufacturer publishes bad info. When the electrical and mechanical parameters don't match I usually just use the electrical.

An externally hosted image should be here but it was not working when we last tested it.

 

[sine143]

No. What I am talking about is the problem with a 30 hz tuning playing a 20 hz tone. So he plays 20 hz and he is limited by excursion since it's below tuning. Then he plays 33 hz and he is no longer limited by excursion so he could turn the power up dramatically and get a higher spl. But only if it's ok by the rules.

so, using a low shelf filter and lowering 20hz by say, 6 db (or using a 12 db per oct hipass set to 35hz or so) and keeping the amp at max tilt wont achieve what you are suggesting?

 

[just a guy]

Yeah, since you put it that way I guess it would. I didn't understand what you meant right away, sometimes I'm a bit dense. Also I assumed if OP had the ability to do this he would already be using a hpf.

 

[mattmanic]

Thanks, that does look pretty close to the measured results with my current box, is that with 2 drivers or just one?

I can adjust volume and open/close doors to change loading and you've got up to 30 seconds per tone generally, but as things heat up quick you usually just do your best, swing the door til the meter reads loudest and leave it.

I'm on my phone at the moment so I'm having a little trouble keeping up. I'll have a proper look at your results in a minute