Tapped Horn - Hornresp vs Reality - AVS Forum
Forum Jump: 
 6Likes
Reply
 
Thread Tools
post #1 of 31 Old 07-08-2014, 05:26 PM - Thread Starter
Newbie
 
Join Date: Jul 2014
Posts: 9
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Tapped Horn - Hornresp vs Reality

I designed a tapped horn using Hornresp and folded it using Sketchup. I feel like something didn't turn out quite right along the way - I mean it sounds decent, but it is fairly far off from what I was expecting.

First, the hardware:
Spidf into Insignia amp/receiver which has a sub out -> dbx 223XL crossover -> Peavey CS400
(The crossover is used to manage the gain on the sub; going straight into Peavey from the Insignia results in extremely quiet bass, even at full sub volume on the receiver.)
Sundown E-15 V2 D2 (wired at 4 ohms; amp does ~200w at 4 ohm)

Room: 15 x 14 with a 3.5 x 4.5 extension with a sink and cabinet; shown in ASCII art (x is the current sub location):
___________
|x..............._|
|................|
|_________|

Enclosure design is attached as an image (Sketchup available if needed). The attached "Tapped.png" shows what I intended to built vs what actually resulted (things I didn't account for in the original design, plus any measurement error). The box should be well sealed as there are screws roughly every 3 inches and liquid nails was used (had some left over from when we built our barn).


So, by the looks of hit, I should be roughly around 127 db between 25-60 hz at 1 meter at 200 watts (which I'm hitting according to my multimeter).

Well, my SPL meter hits about 120-122 right in the mouth of the sub at C weighting and about 105-110 from one meter. Now, I'm already aware that reality seems to yield lower SPL results and less flatness (due to a variety of factors) based on what I've seen in project threads, but this difference is pretty huge. The sub is in a corner as shown in the attached image (I've also tried pointing the throat into the corner and it improved the results, but certainly not to 127 db; I think it was like 110-115?).

The meters accuracy isn't a problem either - I had a Peavey cabinet before (two black widow 15s @ 200w each) pointed into the corner. I mapped the performance in Hornresp (attached) which closely matches the acoustic chart shown in the manual for the Peavey cabinet (can't remember the specific model right now). Playing music, I could hit about 115-120 on the Peavey cabinet - playing similar music on the tapped horn I hit 110; based on the acoustic chart comparison of the black widow enclosure vs the tapped horn, the tapped horn should be mopping the floor with the widows (especially on the low end, and indeed the lows are a lot more powerful, but nothing between 20-60 hz is as loud as it should be).


So - I just threw a lot of information at you (assuming your still reading); to put the issue concisely - my tapped horn should generally be quite a bit louder than the Peavey enclosure, or at the very least compete, but I'm just not feeling the same presence. The Peavey is still at my parents house (I'm still in college; let's not laugh at the cliche of the basshead college student ), so I can use the Peavey in conjunction with the tapped horn, but the problem of the tapped horn's incorrect output is still a problem I want to try and resolve if possible. This is not my first sub build, but it is my first tapped horn (basically I saw a few forums, this one including, talking about how awesome they are and I wanted to take on the challenge). Does anybody with more experience have some suggestions? I don't want all the time and money I put into this build to feel like a waste (I mean, it works, but it is quite underwhelming :/).

Many thanks!

[Apologies for such a long post; I wanted to prevent as much relevant information as possible.]
Attached Images
File Type: png Tapped.png (23.8 KB, 46 views)
File Type: png peavey.png (16.8 KB, 41 views)
File Type: jpg E-15 V2 D2 Folded Version 3.jpg (162.3 KB, 56 views)
File Type: jpg IMG_20140708_184620.jpg (77.5 KB, 54 views)

Last edited by Blast12345; 07-08-2014 at 08:21 PM.
Blast12345 is offline  
Sponsored Links
Advertisement
 
post #2 of 31 Old 07-08-2014, 07:44 PM
Senior Member
 
Gaugster's Avatar
 
Join Date: Feb 2005
Location: Chicago, IL
Posts: 315
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 29 Post(s)
Liked: 16
First I will state that I have never built a bass horn or tapped horn. But have played with hornresp some.

Basic horn idea is to start with high pressure at the front of the speaker and gradually have the pressure reduce as the sound waves move towards the termius/throat. This pressure drop increases the velocity and makes it louder. (Something like that)

This is one reason why horns have high efficiency. The first chamber section may not be small enough to create the high pressure condition. Simulations can be way off if some underlying principals are not followed.

You might be able to modify the box to put more acoustic load ( not sure if that is correct term) and get more output.

Last edited by Gaugster; 07-08-2014 at 07:49 PM.
Gaugster is offline  
post #3 of 31 Old 07-08-2014, 10:21 PM - Thread Starter
Newbie
 
Join Date: Jul 2014
Posts: 9
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Thank you for your reply. When looking into designing a tapped horn, the topic you just addressed was brought up. I do believe it was referred to as compression ratio (S2 size relative to your cone size; I have roughly a 1:1 compression ratio). A higher compression ratio is nice because you stand the possibility of making a much smaller box, but you can quite hit the same lows with decent flatness.

If that is the case (that is, me needing an increased compression ratio), I would likely need to redo the entire box because of the glue.

Again, I do appreciate your response and the comment about principles vs simulation is something worth considering (regardless of whether or not it is applicable in this case).

Question is still up in the air though
Blast12345 is offline  
post #4 of 31 Old 07-09-2014, 05:05 AM
AVS Special Member
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 9,523
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 1291
You model appears accurate, so assuming there are no air leaks in the cab I question your testing protocol. Elaborate on your testing set up.
Mfusick likes this.

Bill Fitzmaurice Loudspeaker Design

The Laws of Physics aren't swayed by opinion.
Bill Fitzmaurice is online now  
post #5 of 31 Old 07-09-2014, 06:39 AM
AVS Addicted Member
 
Mfusick's Avatar
 
Join Date: Aug 2002
Location: Western MA
Posts: 22,380
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 142 Post(s)
Liked: 771
Could the design be right but the build be a little off ?

-

"Too much is almost enough. Anything in life worth doing is worth overdoing. Moderation is for cowards."
Mfusick is online now  
post #6 of 31 Old 07-09-2014, 08:22 AM
Senior Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 323
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 26
Your model isn't even close to accurate. .5 pi is the theoretical maximum when corner loaded, so if you were in a basement with thick concrete walls it might be appropriate to use .5 pi, but even then it's still safer to use 1 pi and if you get more than that it's a bonus since you will never get the theoretical maximum. It appears your building is more likely upper floor with drywall, vinyl siding with some very large windows. It is very possible that you are getting close to 2 pi loading, that would account for the difference right there.

From the picture it appears the horn mouth is about 4 feet from the corner. As shown in the following picture, that can affect frequencies all the way down to your low knee, reinforcing the idea that you probably aren't getting much more than 2 pi loading across much of your bandwidth.



Moving on, the first thing I usually check with sims don't match measurements is the driver t/s parameters. Did you verify the numbers you are using are correct? The published specs usually are not. The second thing I usually check is the fold. Is it done properly? I don't have time to check but a high percentage of the ones I've looked into usually have sizable errors. The third thing I check is the physical build. Construction errors, especially air leaks, can be devastating. And finally the last thing I would do (but it's still necessary), you should check your equipment, make sure your source and amp are doing what they are supposed to and eliminate the crossover for testing purposes. Personally I'd dump the receiver too and measure at 1 watt, troubleshooting is a lot easier when you eliminate everything that isn't absolutely necessary. There's at least a dozen settings on a receiver that could cause problems and if necessary you can use a DI box to boost the level instead, although you probably won't need one to get to 1 watt. Use a measurement mic instead of an spl meter and test tones (not music) and measure the sub outside well away from any boundary reflections and compare that to a 2 pi sim if you want to make things a lot easier on yourself. And above all, don't compare your memory of a different sub in a different house to this sub in your current location. The room changes everything and aural memory isn't very good.
Mfusick and lilmike like this.

Last edited by diy speaker guy; 07-09-2014 at 08:46 AM.
diy speaker guy is online now  
post #7 of 31 Old 07-09-2014, 10:08 AM
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
Not a lot to add to the previous post. Simply put, 0.5 pi models are not reality.

Measurements would help with understanding, especially groundplane ones taken outdoors.

Also, you built parabolic flares, not conical ones, so it might not hurt to change that parameter so your model reflects what you actually constructed. I don't expect significant changes there though.
lilmike is offline  
post #8 of 31 Old 07-09-2014, 10:32 AM - Thread Starter
Newbie
 
Join Date: Jul 2014
Posts: 9
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Quote:
Originally Posted by Bill Fitzmaurice View Post
You model appears accurate, so assuming there are no air leaks in the cab I question your testing protocol. Elaborate on your testing set up.
I am using a tone/wave generator. I know for a fact I am putting out adequate power from the amp (I am using a multimeter to read ~200 watts/28 volts; I won't turn up past the DDT filter (basically signal compression to prevent clipping) turning on for obvious reasons). I'm measuring sound on my decibel meter with C weighting. My comparison to the 120 db put out by the Peavey sub enclosure is being done with similar music at a volume that I see the DDT light flicker on) Is there any more information you would like to know about my testing?

@Mfusick
The design is a little off without a doubt, but one of the images I posted does contain another hornresp setup that accounts for inaccuracy, at which point I am still significantly off. When I get back to my place I will double check the measurements (as best I can/what I can reach and see) to ensure that they are within a reasonable margin of error.

@diy speaker guy
Regardless of the angle, this should be generally louder (or at least as loud) than the Peavey enclosure in the range I care about. Though this holds true for the bottom end, 40 to 60 does not seem to be the case. I intend to do a side by side comparison (using the wave generator) this weekend to get a more direct comparison then 'similar music' and simulated charts

You are pretty close with your analysis of my building/room location; I am a corner room, first floor (of basement, first floor, second floor home) with two sets of windows (each set is three windows wide; first set is narrow tall, other set is square). It is in fact drywall. Siding may be vinyl by process of deduction (I know it isn't wood or metal, but if it is vinyl, it certainly isn't the vinyl siding I'm used to).

I did not check each individual driver parameter (I haven't the slightest idea how to find most of that myself; I know impedance, cone area, and thats about it for first-hand measuring).

I did try pointing the mouth directly into the corner (at an angle so I had a triangle between the walls and the mouth) and that made a fairly minimal difference (I can't recall the db change precisely, but I suppose this would be useful information for the troubleshooting).

I have checked for air leaks using a lighter along any seems and it didn't appear that I had any (at least from the parts I had quick access to last night; tonight I will do a more in-depth test). So, this is still a possibility.

I'll eliminate most of the equipment and test at 1 watt tonight. On a side note, the Peavey enclosure was in this room in the same place (I brought it home because I had expected I wouldn't need it).

I did test the tapped horn in the middle of the room after my dad and I unloaded it from the pickup and I believe it was around 100 db, but I never did any test tones. I will be sure to do that tonight.

I totally understand your cautioning about comparing via memory, but I also remember being able to watch things vibrate. For instance, I could watch the plexiglass window in my server rack vibrate significantly (based on reflection) with the previous subs, but it doesn't seem like this tapped horn is doing that at all.
Blast12345 is offline  
post #9 of 31 Old 07-09-2014, 10:47 AM
AVS Addicted Member
 
Mfusick's Avatar
 
Join Date: Aug 2002
Location: Western MA
Posts: 22,380
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 142 Post(s)
Liked: 771
Build on of lilmike's or Bill fitz designs for horn and transplant the driver.. see if you like it more. I think either could tell you if your drive would work or not.

-

"Too much is almost enough. Anything in life worth doing is worth overdoing. Moderation is for cowards."
Mfusick is online now  
post #10 of 31 Old 07-09-2014, 11:45 AM
Senior Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 323
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 26
Quote:
Originally Posted by Blast12345 View Post
Regardless of the angle, this should be generally louder (or at least as loud) than the Peavey enclosure in the range I care about. Though this holds true for the bottom end, 40 to 60 does not seem to be the case.
These two subs are nothing alike, the Peavey has no output below 70 hz and is very strong above 70 hz, and the tapped horn has a massive hole in response between 70 and 100 hz but is very strong below that. So it stands to reason that they would sound COMPLETELY different. And 40 to 60 hz is right in the middle of the modal range so it's completely possible that the two subs are stirring up modes in different ways which is completely possible due to a number of factors.

Quote:
Originally Posted by Blast12345 View Post
You are pretty close with your analysis of my building/room location; I am a corner room, first floor (of basement, first floor, second floor home) with two sets of windows (each set is three windows wide; first set is narrow tall, other set is square). It is in fact drywall. Siding may be vinyl by process of deduction (I know it isn't wood or metal, but if it is vinyl, it certainly isn't the vinyl siding I'm used to).
My house is very similar. I have almost no room gain at all at low frequencies but I do have a few very annoying room modes. A .5 pi sim would not accurately describe my situation in any corner of my house, even 1 pi would be pushing it.

Quote:
Originally Posted by Blast12345 View Post
I did not check each individual driver parameter (I haven't the slightest idea how to find most of that myself; I know impedance, cone area, and thats about it for first-hand measuring).
You can make a jig with $2 worth of parts or you can buy one for a couple hundred. (Parts Express is out of stock so the only option is the WT2 right now.) It's almost always worth it to measure the parameters. Furthermore, I'm not familiar with this driver at all but seeing the word "Sundown" brings thoughts of ultra high excursion which USUALLY goes hand in hand with ultra high inductance. This can play havoc with what you think you should be seeing based on a simulation. Check data-bass.com measurements of ANY of their high excursion drivers (with the possible exception of tc sounds drivers) and compare that to a sim of the driver in the same box and you can see how much these high inductance drivers refuse to conform to a simulation. You need a MUCH more complex system than Hornresp is capable of for accurately simulating high inductance drivers.

Quote:
Originally Posted by Blast12345 View Post
I have checked for air leaks using a lighter along any seems and it didn't appear that I had any (at least from the parts I had quick access to last night; tonight I will do a more in-depth test). So, this is still a possibility.
Half the joints are inside the box and you can't check anything inside the box with a lighter.

Quote:
Originally Posted by Blast12345 View Post
I totally understand your cautioning about comparing via memory, but I also remember being able to watch things vibrate. For instance, I could watch the plexiglass window in my server rack vibrate significantly (based on reflection) with the previous subs, but it doesn't seem like this tapped horn is doing that at all.
This is not surprising. As mentioned, these subs are COMPLETELY different, the Peavey is VERY strong above 70 hz and the tapped horn is VERY strong below 70 hz and the areas where they are very strong do not overlap at all. Everything has a specific resonant frequency. The stuff that rattles when the Peavey is playing is NOT the same stuff that will rattle when the tapped horn is playing. But give the tapped horn 2000 watts and it will probably rattle the entire house structure pretty effectively (assuming there's no air leaks).

I'm still going to recommend measuring outside, that's the only way to get any real data that you can trust in a troubleshooting sense. Comparing in room is fine but it won't tell if there's anything wrong with either sub. And I don't think your testing methodology is even vaguely capable of giving any useful information. Using "similar" music and reading spl from a meter when the light flashes is not going to cut it if you really want to know what's going on.

Last edited by diy speaker guy; 07-09-2014 at 11:56 AM.
diy speaker guy is online now  
post #11 of 31 Old 07-09-2014, 12:08 PM
AVS Special Member
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 9,523
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 1291
Quote:
Originally Posted by Blast12345 View Post
I am using a tone/wave generator. I know for a fact I am putting out adequate power from the amp (I am using a multimeter to read ~200 watts/28 volts; I won't turn up past the DDT filter (basically signal compression to prevent clipping) turning on for obvious reasons). I'm measuring sound on my decibel meter with C weighting. My comparison to the 120 db put out by the Peavey sub enclosure is being done with similar music at a volume that I see the DDT light flicker on) Is there any more information you would like to know about my testing?
This is how you should be testing. Use a measurement program like HolmImpulse and a USB mic, like this one:
http://www.parts-express.com/dayton-...phone--390-808
Test the cab outdoors, the mic an inch off the ground, 2 meters from the horn mouth. The level doesn't matter, you just want it well above ambient, so 80dB or more should suffice. Then run a 100Hz sine wave at 4.0v. Measure that result with the sound meter. That result is the equivalent of 1m/1w with a 4 ohm cab. Use that result to calibrate the chart. You now have a 1w/1m half-space result. Compare that to the HornResp sim in 2pi space. If they aren't quite close something's off with the cab or driver.

Bill Fitzmaurice Loudspeaker Design

The Laws of Physics aren't swayed by opinion.
Bill Fitzmaurice is online now  
post #12 of 31 Old 07-09-2014, 01:08 PM
AVS Special Member
 
Oklahoma Wolf's Avatar
 
Join Date: Nov 2006
Location: Eastend, SK, Canada
Posts: 1,745
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 26
Check again for leaks. In a horn design, it's entirely possible to have something check out fine at low volume and then leak like crazy once you crank it. I found that out the hard way when one my SDX10s started clanking well below the point it should have, and I discovered the rubber gaskets that came with the drivers weren't quite up to the job of sealing off a horn design
Oklahoma Wolf is offline  
post #13 of 31 Old 07-09-2014, 04:20 PM
Senior Member
 
Gaugster's Avatar
 
Join Date: Feb 2005
Location: Chicago, IL
Posts: 315
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 29 Post(s)
Liked: 16
Quote:
Originally Posted by Blast12345 View Post
I did not check each individual driver parameter (I haven't the slightest idea how to find most of that myself; I know impedance, cone area, and thats about it for first-hand measuring).
Chapter 2 of "Bullock on Boxes" describes how to measure TS parameters for a speaker However you need at least a waveform generator, DMM and some resistor loads. Both Amazon and Parts Express sell this paperback book. Worth checking out as I am sure your driver is well broken in by now. Like others mentioned the measured results for Fs etc... can be significantly different.

Gaugster

Last edited by Gaugster; 07-09-2014 at 04:23 PM.
Gaugster is offline  
post #14 of 31 Old 07-09-2014, 05:01 PM
Senior Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 323
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 26
You don't need a book, there are a bunch of websites that describe how to do this. Like this one - http://sound.westhost.com/tsp.htm
But that's the hard way. The really hard way.

If you want to diy the easy way is to use REW or LIMP (from the ARTA suite) and a simple jig like this - Lilmike’s DIY Impedance Measurement Jig or even a much much simpler jig like this - http://zobsky.blogspot.ca/2008/01/si...t-jig-for.html
You can even remove the switch from the simple jig, technically all you really need is a resistor and a couple of 3.5 mm jacks.

If you don't want to diy anything there's this - http://www.woofertester.com/wt2product.htm
By all accounts that's the best product on the market by far.

But if you want to save a few bucks and don't mind waiting there's this - http://www.parts-express.com/dayton-...ystem--390-806
Previously called WT3 and currently out of stock, here's the story on this one -

Quote:
Originally Posted by PARTS EXPRESS View Post
Note: The Dayton Audio DATS is temporarily out of stock due to a hardware update. This hardware will upgrade the DATS to the DATS V2. An estimated time of arrival will be provided when Dayton Audio is further along in development. The DATS V2 upgrade improves the stability, reliability, and virtually eliminates the startup time associated with past iterations of the DATS and WT3 systems.
It remains to be seen if the hardware upgrade will solve the well known issues with this device. Until proven, my recommendation is the S&L WT2.

I didn't go into detail about any of this the first time because realistically nobody ever bothers. Out of the dozens of times I've mentioned that t/s parameters should be measured I don't believe I've ever talked anyone into doing anything about it.
diy speaker guy is online now  
post #15 of 31 Old 07-09-2014, 06:30 PM - Thread Starter
Newbie
 
Join Date: Jul 2014
Posts: 9
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
I know that the Peaveys are stronger at those higher frequencies, but I had the crossover set at 70 hz anyways (my tops are JBL TR 225s; they each have 2 15s that are internally crossovered to 50 or 60hz+). Ableit, the crossover does do a roll off, so I still get a (weaker) 80 hz even though it is set at 70.



Alrighty, here is what I've found out tonight.

It does about 3 decibels better being fired into the corner as opposed to the picture I posted earlier.

I checked again for air leaks (at least what I could get to) as best I could (any suggestions btw?) and could find none on the enclusure, but I needed to tighten my subwoofer a bit (to allow the rubber ring to make a tight seal). It fixed the buzz I was getting at 60 hz, but it didn't increase performance by a huge margin (at first I was thinking the buzz was being created as air rush over my wires which are hanging in the mouth right now, but it seems that wasn't the case; turns out it was the seal around the sub).

It is a bit late to measure the TS parameters for this project, but I will probably build the DIY one for any future builds.

Lastly, I crawled inside as far as I could (I'm really thin :P) and measured some stuff. S1 and S4 are perfect. But (look at attached picture for help), at intersection of the center pieces the distances are off (red circles). Instead of 10.462 it is 10 1/4 and instead of 10.718 it is ~10 1/2; could this have a substantial impact on my result considering this changes the slopes of those boards?

[Forgot to mention this earlier, don't trust the 7.385 measurement on the front; that was a labeling mistake in sketchup. The proper length is 8 1/4", which I used. Additionally, the angle pieces are not present.]

@Bill Fitzmaurice
I will certainly consider looking into that because then I can isolate the problem to the driver+box as opposed to room placement (and if the driver checks out ok, at least based on measuring parameters as per diy's suggestion, then I can conclude it is the box).

As a general question, what are the benefits of the dayton mic (or similar) over a SPL meter?
Attached Images
File Type: jpg Errors.jpg (109.6 KB, 12 views)
Blast12345 is offline  
post #16 of 31 Old 07-09-2014, 07:23 PM
AVS Special Member
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 9,523
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 1291
Quote:
Originally Posted by Blast12345 View Post
@Bill Fitzmaurice
As a general question, what are the benefits of the dayton mic (or similar) over a SPL meter?
Using an SPL meter, along with the required correction factors (C weighting isn't flat) you can measure one frequency at a time, and eventually construct an SPL chart. It will take about an hour, depending how much resolution you go for. Using measurement software you can take a fully charted measurement in about three seconds. With HolmImpulse you get frequency response, impulse response, phase and THD, with resolution up to 1/50 octave. And HolmImpulse is freeware.

Bill Fitzmaurice Loudspeaker Design

The Laws of Physics aren't swayed by opinion.
Bill Fitzmaurice is online now  
post #17 of 31 Old 07-09-2014, 07:36 PM
Senior Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 323
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 26
Quote:
Originally Posted by Blast12345 View Post
I know that the Peaveys are stronger at those higher frequencies, but I had the crossover set at 70 hz anyways (my tops are JBL TR 225s; they each have 2 15s that are internally crossovered to 50 or 60hz+). Ableit, the crossover does do a roll off, so I still get a (weaker) 80 hz even though it is set at 70.
The tapped horn has a 15 db hole in response at 90 hz (assuming the sim is correct) so even with the crossover set at 70 hz these subs are going to sound completely different even above 70 hz unless you use a brickwall filter or very close to it.

The spec sheet for your mains says they have a passive crossover so it is extremely unlikely that they are internally crossed at 50 hz. The spec sheet says they are down 10 db at 50 hz, that's very likely just where they roll off naturally.

And just to be clear, when measuring the subs the mains should be off. Properly matching the mains to the subs is a whole other topic and since it requires delay with the tapped horn but no delay with the Peavey and likely different xo frequencies and slopes for the different subs it can be a source of problems in itself.

Quote:
Originally Posted by Blast12345 View Post
Instead of 10.462 it is 10 1/4 and instead of 10.718 it is ~10 1/2; could this have a substantial impact on my result considering this changes the slopes of those boards?
You would have to do an accurate sim to find out. If you can draw it up accurately in Sketchup you can use Sketchup to find all the measurements and make things easier.

Quote:
Originally Posted by Blast12345 View Post
As a general question, what are the benefits of the dayton mic (or similar) over a SPL meter?
The mic uses specific software to play specific sine wave tones at a specific power level and measure those specific tones and the time it takes them to play and decay and then provides a series of graphs of spl vs frequency vs time and including distortion. There's a lot you can learn from that info. The procedure you are using now uses a meter to measure tones of unknown frequency and unknown waveform shapes at very likely wildly fluctuating power levels and doesn't take time into account at all. And then you are using this to make judgments by ear using different subs with different but "similar" music. There's not much to be learned from the info provided by the procedure you are using.

If you really want to use the spl meter and get some kind of valid results you need to take the sub outside and play test tones at a measured power level at frequency steps. For example calibrate the power level and then play a 10 hz tone, measure. Play a 11 hz tone, measure. Play a 12 hz tone, measure. Continue until you cover the passband you want to study. Write down all the results and then manually draw a graph. Or just get a measurement mic and it will do the measurements and draw the graph for you in a couple of seconds.

Last edited by diy speaker guy; 07-09-2014 at 07:43 PM.
diy speaker guy is online now  
post #18 of 31 Old 07-09-2014, 07:58 PM - Thread Starter
Newbie
 
Join Date: Jul 2014
Posts: 9
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Quote:
Originally Posted by Bill Fitzmaurice View Post
Using an SPL meter, along with the required correction factors (C weighting isn't flat) you can measure one frequency at a time, and eventually construct an SPL chart. It will take about an hour, depending how much resolution you go for. Using measurement software you can take a fully charted measurement in about three seconds. With HolmImpulse you get frequency response, impulse response, phase and THD, with resolution up to 1/50 octave. And HolmImpulse is freeware.
Many thanks good sir.

Ahh - I stand corrected on the "internal" crossover of the TR 225s; I must have misinterpreted the information when I looked at the documentation last.

The mains have been off during testing.

Music is the best I have to go off of at this given point, and though I don't disagree with you on it being a bad comparison, it isn't that I'm judging only by ear. I'm more focusing on the music in my chest (granted, that isn't a whole lot better) and how it registers on the meter. I should elaborate on "similar" - there is a playlist I frequently use for bass testing or showcasing the system to friends and neighbors, so it is roughly the same set of music I am comparing the two enclosures to. I know, far from ideal, but until I get the other set back here I do not have any better information to give.

Also, I was getting a significant voltage drop (50-70% easy) around 37 hertz, but I assumed that was some eq setting that needed to be adjusted. Well, skipped over the insignia (straight into dbx crossover and into sub) and now I have a very powerful 37 hertz. In addition, I ended up with about a 3-5 decibel gain while playing music (test results obviously didn't change much, with the exception of the area around 37 hertz). Firing it into the corner I can hit about 118 at 60 hz. So, that matches 1 pi really closely if I assume my decibel meter to be accurate (sorry; its all I have to work with right now ) I will measure tones with the Peavey this weekend so I can know how the tapped horn compares.

Perhaps my worries about messing this cabinet up in some important manner was actually the result of a few issues (position of mouth, 37 hz dead zone, the inductance with high excursion drivers that was mentioned, mild errors in building the cabinet, and the general difference in simulated results vs reality).

Based on this new information, do any of you have any new comments or conclusions that could be drawn?

Last edited by Blast12345; 07-09-2014 at 08:12 PM.
Blast12345 is offline  
post #19 of 31 Old 07-09-2014, 08:01 PM
AVS Special Member
 
Ricci's Avatar
 
Join Date: Mar 2007
Location: Louisville KY
Posts: 5,092
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 204
Good points have been made by diyspeaker guy. This is most likely a combo of the horn built being slightly different than the simulation for the reasons mentioned. I'm not familiar with this particular Sundown but the effect that inductance has on the behavior of most long throw high inductance drivers can not be underestimated. For a simple sealed or IB build it doesn't matter much but by the time you get to a much higher order alignment like a tapped horn the system performance is very sensitive to the driver behavior. You may have something that performs nothing like what was modeled simply due to this. For future reference if you do more modeling throw out the 0.5 and 1.0 space simulations entirely. In my experience those are only even remotely applicable if you are in a car cabin. Otherwise use 2.0. I actually use 4.0.
Mfusick likes this.
Ricci is offline  
post #20 of 31 Old 07-09-2014, 08:15 PM - Thread Starter
Newbie
 
Join Date: Jul 2014
Posts: 9
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Quote:
Originally Posted by Ricci View Post
Good points have been made by diyspeaker guy. This is most likely a combo of the horn built being slightly different than the simulation for the reasons mentioned. I'm not familiar with this particular Sundown but the effect that inductance has on the behavior of most long throw high inductance drivers can not be underestimated. For a simple sealed or IB build it doesn't matter much but by the time you get to a much higher order alignment like a tapped horn the system performance is very sensitive to the driver behavior. You may have something that performs nothing like what was modeled simply due to this. For future reference if you do more modeling throw out the 0.5 and 1.0 space simulations entirely. In my experience those are only even remotely applicable if you are in a car cabin. Otherwise use 2.0. I actually use 4.0.
Dually noted.

When I do longer sweeps from 20-70 I begin to see what you mean by the inductance. There is mild unevenness in sound (whether it be from the room acoustic, inductance, whatever), but I think I might be able to pull a good portion of that out with an equalizer.
Blast12345 is offline  
post #21 of 31 Old 07-10-2014, 05:16 AM
AVS Special Member
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 9,523
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 1291
Quote:
Originally Posted by Blast12345 View Post
When I do longer sweeps from 20-70 I begin to see what you mean by the inductance. There is mild unevenness in sound (whether it be from the room acoustic, inductance, whatever), but I think I might be able to pull a good portion of that out with an equalizer.
You can see the effect of inductance in HornResp, just run the sim, then change Le to 4mH, run it again and compare the two. There's very little difference in the passband, a major difference above 100Hz. OTOH the room will have a huge impact on response. That's why measuring outdoors is critical, lest you drive yourself nuts trying to fix something that isn't broken, or trying to fix something so badly broken that it's un-fixable.
Mfusick and ejh2854 like this.

Bill Fitzmaurice Loudspeaker Design

The Laws of Physics aren't swayed by opinion.
Bill Fitzmaurice is online now  
post #22 of 31 Old 07-10-2014, 08:14 AM
Senior Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 323
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 26
Quote:
Originally Posted by Bill Fitzmaurice View Post
You can see the effect of inductance in HornResp, just run the sim, then change Le to 4mH, run it again and compare the two. There's very little difference in the passband, a major difference above 100Hz.
As I mentioned before, Hornresp can't simulate inductance correctly. A very high inductance driver absolutely will have huge effects below 100 hz in reality. The entire response shape can change and you can see it in the measurements. But Hornresp won't show anything like that.

Data-bass.com is probably the best resource for this. Pick a high inductance driver from the list. Simulate it in the same box they measured it in. Look on in shock and horror while you realize your sim doesn't look ANYTHING like the measurement. In the ultra high inductance drivers the inductance effects can affect response almost as much as problems caused by the room.

Last edited by diy speaker guy; 07-10-2014 at 08:18 AM.
diy speaker guy is online now  
post #23 of 31 Old 07-10-2014, 07:03 PM
AVS Special Member
 
Ricci's Avatar
 
Join Date: Mar 2007
Location: Louisville KY
Posts: 5,092
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 204
Correct. You need complex inductance modeling to even get in the ballpark with those types of drivers and HR can't do it. You can fool with the single generic LE parameter all day and you cannot make the simulation act like these drivers actually behave in a basic sealed enclosure.
Ricci is offline  
post #24 of 31 Old 07-10-2014, 07:36 PM
AVS Special Member
 
Bill Fitzmaurice's Avatar
 
Join Date: Nov 2009
Location: New Hampshire
Posts: 9,523
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 1291
Quote:
Originally Posted by Ricci View Post
Correct. You need complex inductance modeling to even get in the ballpark with those types of drivers and HR can't do it. You can fool with the single generic LE parameter all day and you cannot make the simulation act like these drivers actually behave in a basic sealed enclosure.
That very well may be, but this isn't a basic sealed enclosure, nor would one call 1.8mH high inductance. I can't say what a truly high inductance driver would do in a horn, as I don't use them.

Bill Fitzmaurice Loudspeaker Design

The Laws of Physics aren't swayed by opinion.
Bill Fitzmaurice is online now  
post #25 of 31 Old 07-10-2014, 09:37 PM
Senior Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 323
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 26
Being familiar with the name Sundown but not this particular driver, when I looked it up tonight it wasn't really what I expected. 2 inch VC, 400 watt power handling, 12 mm xmax, I wasn't even aware that they made anything cheap with low excursion and a tiny VC like this. That Le seems really low for a low end driver like this (which is why I didn't pay much attention to the Le spec in the sim) and now that I've looked it up I'm not sure I trust the published spec for a few different reasons (even though Sundown's more recent and higher end drivers seem to fairly accurate published specs based on third party testing) but it is possible that inductance isn't an issue with this one.
diy speaker guy is online now  
post #26 of 31 Old 07-11-2014, 07:59 AM
AVS Special Member
 
Ricci's Avatar
 
Join Date: Mar 2007
Location: Louisville KY
Posts: 5,092
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 204
I wasn't actually talking about this particular driver or the OP's situation with that last post but in general about using the LE parameter in HR. I'm not familiar with the OP's driver. It is likely much better behaved than drivers with heavier 3" coils and no shorting rings.

The point about a sealed enclosure is that it is the simplest alignment and if the simulated response cannot be made to match the real response of even a simple sealed system then there is no hope for a much higher order system like a tapped horn.

Basically before spending a lot of time and effort developing a horn for a driver based on simple or incomplete manufacturer specs it is advisable to take measurements of the driver to be used and learn some things about its behavior. In general what I have found is that the more xmax the driver claims the more important it is to do this...

Sorry OP...I'm not really directing this at you but for others who may be reading and thinking about DIYing a horn.
Ricci is offline  
post #27 of 31 Old 07-11-2014, 01:39 PM - Thread Starter
Newbie
 
Join Date: Jul 2014
Posts: 9
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
@diy speaker guy
This actually is newer than the higher end SA series (I own both; from what I can tell, everything is the same except for the ever-important motor and consequential specifications, not that there similarities mean much). That being said, there is no saying that this lower end model has measurements as accurate as the SA series. My friend is needing to trade vehicles with his dad, so he is needing a new box for his sub, so it seems I'm going to be doing TS measurements sooner than I thought, at which point it becomes worthwhile to measure the E-15 while I'm at it.

@Ricci
No worries - the discussion is much appreciated.
Blast12345 is offline  
post #28 of 31 Old 07-12-2014, 01:50 PM
AVS Special Member
 
BassThatHz's Avatar
 
Join Date: Apr 2008
Location: East side of NW Cascades
Posts: 2,776
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 171 Post(s)
Liked: 331
C-weighting is -3db@40hz, -6db@20hz and -12db@10hz; but between 4khz to 40hz it should be almost deadflat (+- 3db or so).

Quote:
Originally Posted by Blast12345 View Post
200 watts according to my multimeter.
How did you measure it? Is that music-power or sinewave power? The cheaper multimeters tend to give inaccurate results, especially below 40hz.
Play a 60hz sinewave and measure the volts and amps, it should give a realistic number on that. If you don't have an ammeter, that could be why your readings are off (you are not taking the ACTUAL impedance into account, which could be as high as 20-30ohms).

Quote:
Originally Posted by Blast12345 View Post
Well, my SPL meter hits about 120-122 right in the mouth of the sub at C weighting and about 105-110 from one meter. Now, I'm already aware that reality seems to yield lower SPL results and less flatness (due to a variety of factors) based on what I've seen in project threads, but this difference is pretty huge.
I'm currently amplifier-limited, but I get 115 to 121db continuous @ 10ft with 4 to 8 sealed 18's on my termlab meter with bass-music; if I had more watts it would probably go even louder.



jbrown plays his system at around ~105-113db at his 10ft LP, at which point we had to stop going louder for fear of breaking his house or upsetting the wife / cops.


We measured a ported HO-15 doing 113db @ 50hz with ease @ 3ft.


I love my termlab because it is specifically designed for measuring loud bass (105 to 183db) and there is no way to cheat the device, short of putting it inside the port and/or blowing air directly on the sensor hole, but the turbulence would likely show as wideband spectral-noise so you probably couldn't get away with it.
Attached Images
File Type: png uxl18 HTTYD_2.png (279.3 KB, 62 views)
File Type: png Ho-15 50hz2.png (131.1 KB, 62 views)

Last edited by BassThatHz; 07-12-2014 at 01:53 PM.
BassThatHz is online now  
post #29 of 31 Old 07-12-2014, 03:37 PM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 16,154
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 212 Post(s)
Liked: 925
I don't suspect that much of anything is wrong here.


the horn probably achieves its target, but the reality is a 2pi space model in a room with reflections (that cancel) and modes (the may do either).


the dual 15 black widow are loaded with top end punch and the tapped horn has a huge suckout there. so bass vs. no bass. bass wins, no surprise. :-)


the inductance may affect the response a little, but my guess is that isn't the big factor here.


the expansion rate in the model may also affect things a bit, but again, probably not the primary culprit here.

Listen. It's All Good.
LTD02 is online now  
post #30 of 31 Old 07-12-2014, 06:58 PM - Thread Starter
Newbie
 
Join Date: Jul 2014
Posts: 9
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Quote:
Originally Posted by BassThatHz View Post
How did you measure it? Is that music-power or sinewave power? The cheaper multimeters tend to give inaccurate results, especially below 40hz.
Play a 60hz sinewave and measure the volts and amps, it should give a realistic number on that. If you don't have an ammeter, that could be why your readings are off (you are not taking the ACTUAL impedance into account, which could be as high as 20-30ohms).
I'm using a sinewave to measure the power. Now that you mention it, I do have access to an ammeter, but I don't think it is necessary to be very precise with power output. When the DDT turns on, I know that I'm pretty close to my power limit. Plus, I'm not worried too much about power output because this is the same amp that the previous subs were using (though they used two channels, each channel was providing less power considering the higher impedance; so based on amplifier rating - the widow enclosure put out 240w and the tapped horn is putting out 200w; yes, I know these numbers aren't going to be perfectly represented in reality, but it provides ballpark numbers). I do thank you for the information you have provided.

@LTD02
It is pretty unfortunate I wasn't able to make that range less awful, but I did want a strong low end. I knew I could make up that range with something else easily if needed, but until I actually built the tapped horn I was guessing I probably wouldn't need anything to compensate.

I'll post some test results tomorrow involving a direct comparison of the widows vs the horn. I'll also gladly try them together as I have enough power (I have a CS800X in my server rack right now because I had been testing the Sundown @ 400w in the Peavey enclosure; on a side note, it made little to no difference, hence why I built a new enclosure )
Blast12345 is offline  
Reply DIY Speakers and Subs

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off