WRT hornresp model possibly not reflecting the drawing, I can't explain it as well as a picture can. In the pic below, what is L34? Is is a full 32 cm, as was input into hornresp? The green line shows that the driver center certainly measure a full foot away from the mouth. But I don't think the driver is a foot down the line length (represented by the red line), which is the definition of L34 afaik. In this case, I would have used 10 cm or so as L34.
I could be way off on this but a real world frequency response measurement of the physical sub would clear this right up... Does the response match the hornresp predictions?
I could be way off on this but a real world frequency response measurement of the physical sub would clear this right up... Does the response match the hornresp predictions?
An externally hosted image should be here but it was not working when we last tested it.
This is how I look at L23 and L34.
I will re-read the previous posts and try to answer more questions soon. Short on time right now.
Rgs, JLH
I will re-read the previous posts and try to answer more questions soon. Short on time right now.
Rgs, JLH
I don't think that's right, but again, a frequency response measurement would be the deciding factor.
That short little stubby "chamber" or "path" that the driver fires directly into is not contributing to the line length, it would be seen more like a little flare right before the line terminates.
If this is true then your total line length (or more accurately L23) is off too.
If this is not true then I just talk too much, but I defer to GM's description of the drawing which seems to match mine.
That short little stubby "chamber" or "path" that the driver fires directly into is not contributing to the line length, it would be seen more like a little flare right before the line terminates.
If this is true then your total line length (or more accurately L23) is off too.
If this is not true then I just talk too much, but I defer to GM's description of the drawing which seems to match mine.
post2702 looks more right than the previous.
The path length runs from/to the effective centre of pressure for each of the sections.
The path length runs from/to the effective centre of pressure for each of the sections.
just a guy said:
I agree with AndrewT's description of pressure centers - Big surprise 😀 I might be a little bias 😀
To be more serious - To me the lengths are defined by the distances from one cone side to the other cone side. This is what is pictured in my above graphic. I justify it by knowing that the pressure wave has to cover that distance in order to reach each point from L12 to L23 and from L23 to L34. If it doesn't cover that distance, then how did it get there? There is only one path that it can take, No?
It’s okay if you don't see it the way I do. You are free to do it differently than I did.
Rgs, JLH
Actually now that I think about it, I'm quite certain your interpretation is wrong. You have L23 and L34 overlapping each other in the same airspace. That isn't physically possible, L23 has to end before L34 begins. Therefore the hornresp model and the drawing will differ in line length quite a bit, probably a foot or more. Unfortunately what you modelled is not what you drew. That little stub is not contributing 12+ inches to the line length.
Again, a measurement would speak volumes.
Again, a measurement would speak volumes.
the interference patterns would be set up between the signal coming from the line and that coming direct from the cone.
Those standing waves are what causes the ripples in the frequency response due to reflection etc. This is little different.
Those standing waves are what causes the ripples in the frequency response due to reflection etc. This is little different.
Not sure what you are saying there.
But it's as simple as this.
L23 and L34 CANNOT overlap in the same airspace.
I'm either 100% right or 100% wrong. If I'm wrong I'll say sorry. (Are you watching Mr McBean? Can L23 and L34 overlap?)
Pressure centers and interference patterns have nothing to do with calculating line length. You have a physical shape represented by segments in hornresp. My understanding is that those segments cannot overlap. Period. The small airspace the driver fires into does not add anything to the line length. That little cavity is just a bit of a flare at the end of the line, it does not add any length to the line.
But it's as simple as this.
L23 and L34 CANNOT overlap in the same airspace.
I'm either 100% right or 100% wrong. If I'm wrong I'll say sorry. (Are you watching Mr McBean? Can L23 and L34 overlap?)
Pressure centers and interference patterns have nothing to do with calculating line length. You have a physical shape represented by segments in hornresp. My understanding is that those segments cannot overlap. Period. The small airspace the driver fires into does not add anything to the line length. That little cavity is just a bit of a flare at the end of the line, it does not add any length to the line.
just a guy said:
I guess that is the $10,000 dollar question. I'm not 100% certain, but never the less my horn works all too well. Just ask my wife.
I too would like to get David McBean's opinion whether L23 and L34 can overlap. It is easy to rationalize both ways. David you watching this thread?
Rgs, JLH
A general response to many questions that have ran up to now.
I don’t have a dimensioned drawing that I can post. It is all on paper. Once I get the new improved one’s (W8-740) folding figured out, I’ll post that one. No sense in living in the past.
You can have the mouth exit where ever you want it as long as L34 is correct. The front, the back and sides are all fair game in my opinion. Do what you want; I’m not going to stop you. 😀
Another important detail I have not reviled is explained in the below graphic. The total area of the colored areas are equal to 440cm^2. This is the combined cone area (Sd) of the two woofers. Doing it this way “feels” better to me. I’m not 100% sure it makes a technical difference, but I didn’t think it would hurt.
I speculate that since the short distance of duct of the chamber does not have an expansion to it, it will not have an effect on the taper of the tapped horn. It just provides a feed point near the mouth which has the same area as the cones.
Rgs, JLH
I don’t have a dimensioned drawing that I can post. It is all on paper. Once I get the new improved one’s (W8-740) folding figured out, I’ll post that one. No sense in living in the past.
You can have the mouth exit where ever you want it as long as L34 is correct. The front, the back and sides are all fair game in my opinion. Do what you want; I’m not going to stop you. 😀
Another important detail I have not reviled is explained in the below graphic. The total area of the colored areas are equal to 440cm^2. This is the combined cone area (Sd) of the two woofers. Doing it this way “feels” better to me. I’m not 100% sure it makes a technical difference, but I didn’t think it would hurt.
I speculate that since the short distance of duct of the chamber does not have an expansion to it, it will not have an effect on the taper of the tapped horn. It just provides a feed point near the mouth which has the same area as the cones.
Rgs, JLH
just a guy said:
JLH said:
Hi just a guy and JLH,
The Hornresp tapped horn simulation model assumes that L23 and L34 do not overlap, as shown in the schematic diagram.
Just a guy - if you are still using Version 17.nn of Hornresp then I strongly recommend that you upgrade to the latest release, Version 20.20. The Tapped Horn Wizard has been significantly enhanced since Version 17.nn.
Kind regards,
David
Attachments
I know, I've seen the screenshots. All those sliders scare me a bit, lol. I'll be downloading the new one soon to optomize whatever design I end up choosing. The first time I used hornresp to do a tapped horn there was no sliders at all. I thought version 17.whatever was pretty darn fancy.
David McBean said:
David,
Thank you for your reply. Now here is my question: While Hornresp performs calculations based on L23 and L34 not overlapping, doesn’t my construction method result in having them overlap? In other words, my Hornresp input data is correct for the construction method I used. If you haven’t been following this discussion between Just-A-Guy and me, please refer to the below posts for clarification.
This is the Hornresp input data and Google SketchUp renderings
http://www.diyaudio.com/forums/showthread.php?postid=1627597#post1627597
This is how I interpret L23 and L34 based on my construction method
http://www.diyaudio.com/forums/showthread.php?postid=1712919#post1712919
I rationalize this interpretation by saying the pressure wave must still cover the overlapping distances twice in my design. This is very much like pulling your car into a driveway. In order to get back to the street you must recover the length of the driveway to get back out.
Rgs, JLH
No.
No.
Just because your driveway "taps" into the street does not mean your driveway is adding length to the street. If you measure your street from one end to the other your driveway does not enter into the calculation of the loverall ength of the street in any way.
Study the hornresp schematic. It's pretty easy to see that L23 and L34 CANNOT overlap. Hornresp needs L23 to end before L34 can begin. Hornresp is worrying about pressure wave distances specifically so you don't have to. Forget about pressure centers.
just a guy said:
We will have to agree to disagree on this. I think we are at a impass on our interpretations. We should probably move onto something more interesting to discuss. I do not want this thread to degrade just because two guys have different opinions.
Rgs, JLH
That's fine, but remember that lots of people are showing interest in building your designs, so it's probably just polite to make sure it's correct.
Anyway, if you don't believe GM's interpretation of your drawing OR McBean's interpretation of how to model with hornresp (which is incredibly unwise on both counts imo) the only thing that will clear up the confusion is a frequency response measurement.
Anyway, if you don't believe GM's interpretation of your drawing OR McBean's interpretation of how to model with hornresp (which is incredibly unwise on both counts imo) the only thing that will clear up the confusion is a frequency response measurement.
just a guy said:
Fair enough. I don’t disagree with that assessment. Unfortunately, I do not have any gated measuring equipment to do proper indoor testing. As you know, indoor measurements would be pretty useless due to room effects. Additionally, it is far too cold (toady’s high temp will only be 10F / -12C) to do outside measurements. If everyone can be patient and wait for Spring, I will do the proper measurements then.
Rgs, JLH
AKaback question
I'm a little confused.
How do I model a th in 7 sections with the offset driver in akback?
The horn resp (if possible equivalent) would look like this:
s1 -s2 (driver entry) s3 -s4 -s5- s6 -s7(other side of driver)
-s8 (mouth)
I'm guessing:
Waveguide 1, duct for driver entry, Waveguides 2-7, enter driver rear here?, Horn 1
I assume I need some kind of definition for the entry angle?
How do I connect the duct between the wave guide sections?
Lastly, is there a schematic diagram like hornresp?
Thanks
I'm a little confused.
How do I model a th in 7 sections with the offset driver in akback?
The horn resp (if possible equivalent) would look like this:
s1 -s2 (driver entry) s3 -s4 -s5- s6 -s7(other side of driver)
-s8 (mouth)
I'm guessing:
Waveguide 1, duct for driver entry, Waveguides 2-7, enter driver rear here?, Horn 1
I assume I need some kind of definition for the entry angle?
How do I connect the duct between the wave guide sections?
Lastly, is there a schematic diagram like hornresp?
Thanks
just a guy said:
Good analogy, the pink/whatever colored slots are the driver's effective acoustic centers feeding the path-length closest to the terminus, so whatever the distance from their vertical center-lines parallel or vertical to the terminus (depending on where it is) is the most accurate dim to input for L34 (or L45).
If the layout was such that the driver's were offset in a 'bulge' as I first thought, then the L34/L45 becomes a bit harder to define, but they're still close enough to the terminus as to have a minimal effect on performance WRT HF response and driver damping, i.e. driver(s) close to terminus = driver-in-mouth BLH and a 1/4 WL or further away of the gain BW's HF corner there's enough damping to say it's mimicking a compression loaded horn of 'X' rear chamber Vb, which depending on the desired performance goals may not be desirable, especially for a HIFI/HT sub app and apparently in prosound too considering DSL's offerings.
GM
GM said:
Hmm, lest I get nit-picked, this is an oversimplification, to find the acoustic center you have to at least calc each duct's end correction which will be different for each side. This means it will theoretically be down in the horn's oscillating ~uniform particle density jet-stream, but complex wall/duct boundary shear forces will wreak havoc with it, so my SWAG is that it will have no well defined frequency point. This would also hold true for just the driver 'swinging in the breeze' and I assume why quite a bit of the HF 'ugliness' predicted in sims is either much reduced or not there in actuality.
GM