latest study in excess

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Just so everyone knows, with my work & show schedule I only get a brief period every couple of days to pop by here, so please don't think I'm losing interest or the project won't occur. In between all of the other things I have going on I find time to compare ideas & see what meets objectives.

Seems to be some confusion about dimensions. Let me clarify.
The space limitations I have to work with are: on the area flanking the stage: 42" wide and 90" tall, the space under the stage is 18" tall and 8 feet deep, but if it sticks out in front by as much as 12" that's fine, as the existing EVs are commonly slid out to put stage monitors on.

Essentially I have been talking about two sightly different variations of fold for the same horn. On one it appears as I pictured. On the other the baffle plate section and mouth would be rotated 90 degrees vertically (relative to their position on that drawing). If that still confuses people, and a clear improvement doesn't surface, I'll post better drawings of both.

On that note, thanks Oliver... I'll look that over tomorrow & see if it gives me a fresh perspective on what I have so far.

just a guy, another amp isn't an option at this time, we have the amp I intially stated to work with: 4 channels, 1350@8ohm, 2100@4ohm, and 2500@2ohm, bridges pairs to 4200@ 8ohm, and it is recommended not to be run below 2.7 ohm/ch. One channel will drive the existing EV subs.

However, you do have a few good points, some of which I've already considered. I agree under 96" would be easier to build; in fact, with the restrictions I have, under 90" would be ideal. I just can't figure out a way to get a fold that short with 20Hz performance and the power restrictions I'm under. This is as close as I could get. You're right, FLH will get me more upper sub grunt. As weltersys pointed out, I'd likely be trading that for thermal considerations. But reducing the power available to match the lowest Xlim on my sim, and then using that as a comparison against your sim that uses an amp that isn't in the spec, and pushes your model past Xlim before 20Hz? Start by reducing your mouth to 4470 cm2 so it'll fit in the available dimensions (assuming 3/4" ply and no obstructions). Then rerun both using the same Eg and space. We're MUCH closer in a level playing field. I haven't tried to fold yours to see if the rest of it will fit within the available space but as it's effectively the same volume I assume it should...btw your model also exceeds your suggested 96" maximum dimension.

As far as why I posted the graphs I did, really they are the ones that make sense to me in this case. Two of the cabinets will be wall loaded between a subterranean concrete wall and a concrete floor, about as 1pi as it gets. If anything, it might be an underestimation as the bar is right against the mouth, and 90" up is a partial roof. The third cab will be 8 feet-ish into the middle of the room and hence half-space loaded. So...the two flanking ones will be doing most of the heavy lifting, and the third one will be used to tweak the way the room fills. Yes, there is discrete processing for that 3rd position. Everything I've looked at for this project I sim as 3 cabs/2pi, and 2 cabs/1pi, figuring that I'm going to end up somewhere in between when the dust starts to rain from the rafters. :D (I learned my lesson about incorrect spacial presentation in modelling the last time I posted something here when I was just learning to use hornresp.)
 
However, you do have a few good points, some of which I've already considered. I agree under 96" would be easier to build; in fact, with the restrictions I have, under 90" would be ideal. I just can't figure out a way to get a fold that short with 20Hz performance and the power restrictions I'm under. This is as close as I could get. You're right, FLH will get me more upper sub grunt. As weltersys pointed out, I'd likely be trading that for thermal considerations. But reducing the power available to match the lowest Xlim on my sim, and then using that as a comparison against your sim that uses an amp that isn't in the spec, and pushes your model past Xlim before 20Hz? Start by reducing your mouth to 4470 cm2 so it'll fit in the available dimensions (assuming 3/4" ply and no obstructions). Then rerun both using the same Eg and space. We're MUCH closer in a level playing field. I haven't tried to fold yours to see if the rest of it will fit within the available space but as it's effectively the same volume I assume it should...btw your model also exceeds your suggested 96" maximum dimension.

OK, but do you really think the tops are going to keep up with the tapped horns in the kick drum range 80-180 hz...? A front loaded horn or BPH will do much better in the kick.
 
just a guy, another amp isn't an option at this time, we have the amp I intially stated to work with: 4 channels, 1350@8ohm, 2100@4ohm, and 2500@2ohm, bridges pairs to 4200@ 8ohm, and it is recommended not to be run below 2.7 ohm/ch. One channel will drive the existing EV subs.

That's fine, my sim still beats yours within the limitations of the existing amp. (See attached pic.) Besides, if they have a budget to replace a blown sub every month they should be able to afford a new amp in 1 or 2 months. (14000 watt rms amp costs under $1000 delivered to your door.)

I just can't figure out a way to get a fold that short with 20Hz performance and the power restrictions I'm under. This is as close as I could get.

I haven't tried to fold yours to see if the rest of it will fit within the available space but as it's effectively the same volume I assume it should...btw your model also exceeds your suggested 96" maximum dimension.

Use your imagination. From the pic you posted it seems you think that lengthwise folds are the only possibility. There are literally dozens of ways to fold that thing.

Of course mine will fold in the same size package, it's the same size. Mine absolutely does not exceed 96 inches since it's not folded yet. If you really need it, I can show rough sketches of a dozen ways to fold it that will easily fit. I'd prefer that you do some of this work though.

You're right, FLH will get me more upper sub grunt. As weltersys pointed out, I'd likely be trading that for thermal considerations.

Not just upper frequency grunt, 3 db more on the low end is not trivial. Take the time to see how much it would take to get an extra 3 db at tuning with your sim and then get back to me.

As I pointed out, I don't see a thermal issue. I spent a lot of time posting info on why I feel that way, I'm sorry you didn't read it.

But reducing the power available to match the lowest Xlim on my sim, and then using that as a comparison against your sim that uses an amp that isn't in the spec, and pushes your model past Xlim before 20Hz?

First, xlim has nothing to do with this. I don't even know xlim of that driver. Xmax is listed at 18 mm. I redid your sim so that xmax was not exceeded, that is the normal way to do this. My model does not exceed xmax either, assuming you use a proper high pass filter. You are planning to use a high pass filter, right? If so I already addressed this issue and it's not a problem.

Start by reducing your mouth to 4470 cm2 so it'll fit in the available dimensions (assuming 3/4" ply and no obstructions). Then rerun both using the same Eg and space. We're MUCH closer in a level playing field.

Reducing the mouth size to 4470 cm2 will make little difference. Why would I want to run both sims with the same Eg? One can handle MUCH more power within xmax than the other. Both sims I showed were in the same space, both volume space and boundary space (2 pi).

Below I've attached a pic. Since you don't want to consider a larger amp, this is a VERY fair comparison. Your sim is in light grey, shown at xmax with 1400 watts. Mine is in black, shown with 2100 watts. Mine is still nowhere near xmax (assuming you use a hpf). And mine still beats yours. Power compression is not an issue at all here since the drivers are not exceeding pmax. (Both designs shown with 3 cabs.)

To be fair, the required hpf would make things ~ equal on the low end (and you can only get that back with more power) but mine still has lots more on the high end. Room gain will fill in the low end on mine while making yours sound boomy and quieter in comparison. And yours will be at the ragged edge of xmax while mine will be loafing along at very low distortion.
 

Attachments

  • trackzilla.png
    trackzilla.png
    105 KB · Views: 262
Last edited:
Below I've attached a pic. Since you don't want to consider a larger amp, this is a VERY fair comparison. Your sim is in light grey, shown at xmax with 1400 watts. Mine is in black, shown with 2100 watts. Mine is still nowhere near xmax (assuming you use a hpf). And mine still beats yours. Power compression is not an issue at all here since the drivers are not exceeding pmax. (Both designs shown with 3 cabs.)

To be fair, the required hpf would make things ~ equal on the low end (and you can only get that back with more power) but mine still has lots more on the high end. Room gain will fill in the low end on mine while making yours sound boomy and quieter in comparison. And yours will be at the ragged edge of xmax while mine will be loafing along at very low distortion.

Where would you low pass this, and do you feel it would be a good (better) match to the existing tops?
 
I haven't tried to include a high pass (I usually do that in Akabak and don't have the latest Hornresp yet). But from experience I know that a flh with the same rolloff frequency and q need a highpass a few hz higher than a tapped horn. That will take a little bite out of the flh output down near tuning but it can be gained back by adding more power - assuming the drivers can handle it. If there was more size to play with the flh could be designed with a little bump (1 or 2 db) at the knee so that it would remain flat with the hpf in place but that's not an option here since the size is limited.

To be honest I skimmed over the info on the tops, I don't even know what he's using and I don't care. I'm interested in subs, and I think my quick design is better in every metric that I've shown. To give a real answer I'd have to know all kinds of info about the tops and the desired crossover point and I really don't care about any of that.

EDIT - I just realized you were asking about low pass not high pass. I guess I have high pass on the brain.

Anyway, this model is not complete enough to answer that question. It would have to be changed into something that could actually be built first, an offset driver model with PAR segments. That would change things a bit, especially in relation to low pass considerations.
 
Last edited:
Not just upper frequency grunt, 3 db more on the low end is not trivial. Take the time to see how much it would take to get an extra 3 db at tuning with your sim and then get back to me.

First, xlim has nothing to do with this. I don't even know xlim of that driver. Xmax is listed at 18 mm. I redid your sim so that xmax was not exceeded, that is the normal way to do this. My model does not exceed xmax either, assuming you use a proper high pass filter.

Below I've attached a pic. Since you don't want to consider a larger amp, this is a VERY fair comparison. Your sim is in light grey, shown at xmax with 1400 watts. Mine is in black, shown with 2100 watts. Mine is still nowhere near xmax (assuming you use a hpf). And mine still beats yours.

Room gain will fill in the low end on mine while making yours sound boomy and quieter in comparison. And yours will be at the ragged edge of xmax while mine will be loafing along at very low distortion.
As an objective observer, both designs have merit.

I would not count on room gain, room mode cancellation could easily be greater than gain.
Also, a sub that goes lower sounds louder, it only takes a 5 dB increase at 20 Hz to sound twice as loud, while it takes 10 dB at 1000 Hz for the same apparent difference.

Back to one "beating" the other, the TH has an impedance minima at 20 Hz, the driver could be burned up before exceeding Xmax there, while the FLH needs a HP around 25 Hz.

Whether response that low is needed is another question, but the FLH excursion maxima is at 40 Hz, right where the heavy demands in much EDM lie, while the TH has another minima at around 47 Hz.
Depending on the program, the FLH may be near Xmax while the TH will be "loafing along at very low distortion".

Up above about 55 Hz, the sensitivity advantage of the FLH is apparent, but excursion is not a problem for either sub up there, so boosting the TH (if needed) is not a problem until the amp runs out of headroom, at which point the tops probably will also.

Art
 

Attachments

  • FLH,TH.png
    FLH,TH.png
    153.1 KB · Views: 177
I don't want to nitpick too much here but I have to address a few points.

I would not count on room gain, room mode cancellation could easily be greater than gain.

I don't usually count on LOTS of room gain but I've never seen a sub become quieter by placing it in a room. There is always a bit of room gain. There can be cancellations at very certain frequencies but there is always an overall trend of gain.

Also, a sub that goes lower sounds louder, it only takes a 5 dB increase at 20 Hz to sound twice as loud, while it takes 10 dB at 1000 Hz for the same apparent difference.

Both designs have the same rolloff frequency and q by design (I did that on purpose). The hpf will change that a bit but a bit of eq and extra power will level them out again.

Also remember that not much music contains anything below 20 hz anyway, so it's likely that both of these could be used without a hpf at all for a good long time before encountering a song that would give them any trouble at all.

... the FLH excursion maxima is at 40 Hz, right where the heavy demands in much EDM lie ...

The FLH also has a 5 db advantage at 40 hz (and even more if a larger amp is used) so that excursion max at 40 hz doesn't really matter much in a direct comparison. For the tapped horn to get 5 db louder would take almost 4x more power. On one hand you have an excursion maxima but +5db, on the other hand you have an excursion minima but -5db.

Depending on the program, the FLH may be near Xmax while the TH will be "loafing along at very low distortion".

As shown with 2100 watts rms my sim never gets anywhere near xmax, so even in the worst case scenario my design will be loafing - even at the 40 hz excursion maxima. The TH has xmax minima at different frequences than the flh, but don't forget that my design has the spl advantage at ALL frequencies.

Up above about 55 Hz, the sensitivity advantage of the FLH is apparent, but excursion is not a problem for either sub up there, so boosting the TH (if needed) is not a problem until the amp runs out of headroom, at which point the tops probably will also.

Art

And that's the whole point. The amp only has a limited amount of voltage. My design can do a lot more with that voltage. Also, the amp's power supply is limited, so assuming there is NO room gain at all, my design can be eq'ed down in the top end making life much easier on the amp's power supply.

Since the TH can only take 1400 watts down low (excursion limited) it can certainly be boosted on the top end but as far as I'm concerned that should never be necessary if it can be accomplished at no cost within the design itself. As far as I'm concerned, needing to boost the top end is a design defect. Why not use the rising response to your advantage if it doesn't cost anything?
 
Last edited:
AndrewT: The design is 2 parallel sets of two drivers in series. to get an idea of what would happen, just go to driver arrangement in Hornresp & change it to 2 drivers in series. That won't be entirely accurate as the idle drivers will be passive radiators now, you'd have to do it in Akabak to get the real picture. Regardless It ain't a real pretty curve, so it will yell ouch so you know it's in pain, but it is unlikely to destroy the other 3 drivers quickly.

Just a guy:
We won't be getting a new amp for that club any time soon. No matter how much any sub model cries out for it. No budget, no physical space for one. Not my choice, just reality.

I'm not ignoring the impedance maxima and Xlim problem that my design has around 26Hz. I do maintain however that it is counterproductive to limit the whole available power handling to address one frequency set. There are many other ways to solve that. Two quick ones that come to mind are a band limit Zobel, or a tidy filter in an Eminence D-Fend. Dramatically less expensive in either case than an amp, and both leave us with a far more level playing field for the designs. The D-Fend has a bonus of allowing me to bulletproof the rest the subs' spectrum and are likely to be annexed to the plan regardless of what design we end up building. Yes, it is solving a design flaw with additional components, but the TH would still be a cooler running cab with a simpler build.

My point about 96" is one of very basic math. Yes your model has close to the same volume mine does; no yours still does NOT end up under 96" in length, which is all I was trying to say. The total internal volume of 40.5 x 16.5 x 95.25 (which is allowing external 3/4" ply and the space available for the box that goes under the stage before any internal panels) is 1043L. My model is 1036L volume, yours is 1058L. If you have a way to fold your cab, or mine, so that all dimensions are less than 96" and it fits into the given restrictions, you are wasting your time in audio and should design tesseract buildings ;) That is all an aside IMHO, an as any model over 42 x 18 x 90 externally will require two variations of the build, both limited to 42" in one dimension, one with a second fixed limit of 18" and the other two cabs with a second fixed limit of 90".

To address the relevant concerns about low-mid support, we plan to cross the subs a little high, likely around 100-110, to give the top boxes a bit more breathing room. Band shows in there are absolutely nowhere near the volume of the EDM shows. The room simply cannot support that SPL with several open mics. Initially at the EDM shows we were also renting them additional top boxes, but immediately realized they were overkill. The venue isn't after additional full range support, they want a thorough massage added to their existing rig. We will set up an alternate processing profile for live music events so the subs will be under dramatically less stress and the tops will have no problem matching them. I feel the extreme lows are far more relevant to our scenario than a traditional rising response.
 
To address the relevant concerns about low-mid support, we plan to cross the subs a little high, likely around 100-110, to give the top boxes a bit more breathing room. Band shows in there are absolutely nowhere near the volume of the EDM shows. The room simply cannot support that SPL with several open mics. Initially at the EDM shows we were also renting them additional top boxes, but immediately realized they were overkill. The venue isn't after additional full range support, they want a thorough massage added to their existing rig. We will set up an alternate processing profile for live music events so the subs will be under dramatically less stress and the tops will have no problem matching them. I feel the extreme lows are far more relevant to our scenario than a traditional rising response.

OK. I'm pretty sure you won't get convincing kick drum for EDM with the phase issues the TH has in that range.
 
I'm not ignoring the impedance maxima and Xlim problem that my design has around 26Hz.

I wouldn't worry about that at all. This is music, not HT. I've seen some tracks with a strong 27 hz note but not many, and very few if any have anything useful below that.

Anyway, this is why we show our graphs at xmax and keep the volume knob at or below that point. Then nothing bad is going to happen.

Yes, it is solving a design flaw with additional components, but the TH would still be a cooler running cab with a simpler build.

Assuming the same power input they will get up to about the same temp. on average. Yours would have a slight advantage with no drivers encased in a chamber but I don't think thermal problems are an issue, especially under 2100 watts.

My point about 96" is one of very basic math.

If none of your other dimensions are flexible at all then I guess you are right. (I'll trust you, I didn't do the math myself.) I just assumed some of your dimensions would be at least a bit flexible. Maybe I need to read a bit more closely.

I feel the extreme lows are far more relevant to our scenario than a traditional rising response.

My design still has as much low end spl as yours, even with a hpf, and even when limited to 2100 watts. The rising response is just bonus. And assuming the room is concrete and mostly sealed you could see as much as 20 db of room gain at the low end. (Probably closer to 12 db though.) Have you measured the room response?

Anyway, it's more than clear by now that you aren't really looking for an alternative so I won't push it any further. I'm certain that my design would be quite a bit louder (even with the existing amp) and fit the room gain curve much better but I'll leave it at that. I won't mention this again unless someone makes any claims that I need to counterpoint.
 
Last edited:
Anyway, it's more than clear by now that you aren't really looking for an alternative so I won't push it any further.

Actually it's pretty clear that around here it's pretty much tapped horns or the highway. People are way too tied up in using them when it's just not the right thing given the design goals.

And while we're at it....a club around me has two TH-118's.......that sound like complete crap. Could be they aren't setup right, but who knows, I've yet to hear a good tapped horn so I don't buy into them.
 
Last edited:
Well... we have Danley and tapped horns to thank for reviving interest in horns in general so it's not such a bad thing.

Most people don't understand how horns of any kind work, and since tapped horns are the new big thing it's not surprising they are favored, especially given Danley's marketing hype and the ease of calculation with Hornresp.

I agree they are rarely the best tool for the job since unless you are playing outside with no boundaries, a dead flat response is rarely going to sound good when room gain is factored in.

BUT I have to admit the best sounding sub I own is a tapped horn. I've heard plenty of very bad ones, but I've got one that sounds very good within it's bandwidth limitations. It sounds better than a flh using the same driver, but the two horns are completely different in style and construction, and they cover different bandwidths so it's hard to make any conclusions from that. This is purely subjective but the tapped horn sounds "tighter". If I could compare it to getting hit, it would be more like a nice 16 oz construction hammer and less like a rubber mallet. But on the other hand, I've heard a tapped horn that sounded so "loose" it sounded like the bass was half a beat behind the music. I haven't looked much into the reasons for all this, so it's all purely subjective and impossible to profile an alignment without much more in depth research than I have ever done.

I think there are good and bad examples of all alignments, and as you mentioned, setup and associated electronics are very important.
 
I admit I'm a bit TH-centric. I've used lots of FLH and vented box subs over the years. I currently own 16 of those CV T36 FLH and a variety of other boxes. The T36 serve as my defacto benchmark as a result & they are quite decent. I've never heard a TH in person other than the ones I built personally. But I've built 8 completely different TH designs, a total of 28 TH cabs. Some for testing & others for customer installs, with tunings from 40Hz down to 15Hz, and the sonic character of all of them was just nicer to my ears than the T36 or the other FLH I've used in the past. I'm trying not to simply discount an FLH design altogether, but I don't know what exactly in the graphs would account for the character differences I'm so fond of, so I associate the sonic character with the design variety and perhaps that is unwarranted predjudice. By comparison, the TH sound open & relaxed...adroit mighty be a good word, the T36 have impact & volume but they sound like a wad of wet cotton when A/B against the TH. An analogy: A friend of mine is a 5th degree black belt in 3 different arts, once he helped me load-in a PA. We stacked 3 cabs high with the 90Lb top cab inverted. The rest of my crew do this with 2 guys. I do it alone, but slowly & with carefully applied brute force. He walked toward the stack carrying the cab, spun around while inverting the cabinet, and set it exactly in position. I'm the T36, he is the TH.

My very first TH was the venerable Volvotreter 38Hz 6.5 Tang Band, and from there I lept into other designs, mostly my own although I am building 2 this month for an outdoor patio that are a variation of JBells Dayton 18".

BTW, I'm in the club this thread is for ATM babysitting a DJ...at least this one has a clue about levels & gain architecture, I've worked with him before.
 
Last edited:
Hi,
:2c: Does this (The picture) comply or contain any of the 'numbers' that's been presented in this thread?

b:)

I don't know. I've been staring at this pic for 10 minutes and I don't really know what's going on.

When I look at the schematic drawings and the spl response (last 3 pics in the bottom right corner) it looks like you are making some kind of comparison between Trackzilla's TH and my FLH designs. BUT the schematic of my design in my Hornresp says 1056 liters (compared to 1063 in your pic) and the schematic of the TH says 1036 liters (compared to 976 in your pic). So if this pic is actually a comparison of our respective designs, either you or I have made a somewhat large error in basic inputs.

Next note - in the schematic pics, you list the TH at 1350 watts and the FLH at 2940 watts. But in other places (spl and sound pressure graphs) you list power as 2000 watts into each design. In MY Hornresp, the TH design passes xmax at around 1400 watts so I'm not sure how you conclude it could handle 2000 watts. Also, you've shown my design at xmax (in the excursion graph) but it takes 3500 watts for my design to hit xmax.

Also, I'm not sure why you chose the low pass frequency and slope that you did. It would seem to me that using a 12 db/oct low pass would give a lot more flat(ish) bandwidth from the FLH so I'm not sure why you are choking it off with such a low, steep filter that doesn't even afford flat response.

I would be interested to know exactly what you simulated here and what your personal conclusions are. But from the info I see, I have no idea what you are up to or how you feel about any of it. Neither of the designs are the correct internal volume and it seems like you simulated with different amounts of power in different places, and somehow showed my design hitting xmax at way less power than it should while suggesting the TH design can handle way more power than it should. It would be nice if you labeled your graphs better and included the input screens so I could get some kind of clue of what I'm looking at.
 
Last edited:
I was struggling with much the same & trying prep a response...

Bjorno is a smart guy and usually full of useful info but I can't make any sense of this pic at all.

you are not just faster at creating models than I, you're faster at cohesive replies too

The only reason I was able to do the model so quickly is because it's dead simple to do hyp/ex models. If I did a proper offset driver model with PAR segments from scratch it would take much much longer, probably at least 10 or 15 minutes to get a useful result and then another 10 or 15 to make sure it's good enough to present.

Starting with a hyp/ex model is a very quick way to get a theoretical result in a couple of minutes but it's not something that can actually be built. Don't feel bad for spending time on sims, there's nothing wrong with that.
 
AndrewT: The design is 2 parallel sets of two drivers in series. to get an idea of what would happen, just go to driver arrangement in Hornresp & change it to 2 drivers in series. That won't be entirely accurate as the idle drivers will be passive radiators now, you'd have to do it in Akabak to get the real picture. Regardless It ain't a real pretty curve, so it will yell ouch so you know it's in pain,
thank you
but it is unlikely to destroy the other 3 drivers quickly............................
that's what would be important to me.
I would NOT want three other drivers damaged because one driver failed.
the fact that the performance would crumble is unfortunate.

The reason I ask is that for PA work, redundancy in the SPL delivery system is paramount to keeping the (paying) customers happy.
I would always go a for single amplifiers driving single drivers. I would choose to use 4 of them. Now one amplifier and/or one driver can fail, the performance carries on and with luck 95% of the audience will not know there has been a failure. Hopefully your roadies get the broken channel running again in a few tens of minutes. With planning the "downtime" could be less than 10minutes for that one dead channel.

The multi driver arrangement going dead is a calamity that is avoidable.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.