Hi, fellas, to preface, I've already read a couple threads on the subject and also skimmed the Salvatti et. al. paper on the subject of ports.
This will be a long story but the overall topic shall be will it be worth the effort to increase my port area by another 20%, and at what point are my woofers unable to properly control the air mass in that port?
Some background. I'm working with a (mostly) SPL oriented enclosure in a 97 Jeep. Subs are 4 15" Shocker neo-Sigs (rare TC sounds build SPL subs)... very strong motor force, very stiff suspension, and rather heavy cone and coil (to handle loads of power).
Box volume is roughly 24-26 cubic feet and for practical purposes cannot be changed. Going larger would require a major tear-down and I recently experimented with reducing box volume to raise tuning and it was a colossal failure, I lost over 2dB, so I don't think that's getting me anywhere.
My peak frequency has hovered between 43 and 45hz regardless of what I've done in this vehicle (a dozen different boxes and a different wall setup).
My initial port was roughly 300 square inches and effectively tuned far too low. I quickly knocked what I could off of port length and my port "depth" is pretty much the thickness of the baffle plus half the port width. This configuration got me about 155.3dB @ 43hz. Thinking I could raise tuning I performed a bit of surgery on the port wall and opened it up to about 420 square inches. This gained me up to 156.8dB @ 45hz. I did some other minor tricks and am currently hovering between 157.0 and 157.4 depending on the day/weather.
In any event, my largest single gain so far has been increasing port area, and increasing port area overall gives me more options for rounding the mouth or tuning lower again if I miss the mark too high.
The catch being opening the port will again require some pretty major surgery on the box, replacing the baffle, and a couple hundred bucks in materials, so if there's a point where more port area won't help I'd just as soon save myself the money and heartache. It's pretty loud as it sits but I'd like to squeeze another full dB out of it before resorting to doubling power which opens the huge can of worms which is batteries right away.
Anyway, I'm interested in any input on port area vs. efficiency in this application. Consider the ONLY goal is gaining on the SPL meter (I'll also try just about anything else that'll gain). I have TS parameters of the subs and impedance sweep graph of the subs in the box in the vehicle if that'll help.
This will be a long story but the overall topic shall be will it be worth the effort to increase my port area by another 20%, and at what point are my woofers unable to properly control the air mass in that port?
Some background. I'm working with a (mostly) SPL oriented enclosure in a 97 Jeep. Subs are 4 15" Shocker neo-Sigs (rare TC sounds build SPL subs)... very strong motor force, very stiff suspension, and rather heavy cone and coil (to handle loads of power).
Box volume is roughly 24-26 cubic feet and for practical purposes cannot be changed. Going larger would require a major tear-down and I recently experimented with reducing box volume to raise tuning and it was a colossal failure, I lost over 2dB, so I don't think that's getting me anywhere.
My peak frequency has hovered between 43 and 45hz regardless of what I've done in this vehicle (a dozen different boxes and a different wall setup).
My initial port was roughly 300 square inches and effectively tuned far too low. I quickly knocked what I could off of port length and my port "depth" is pretty much the thickness of the baffle plus half the port width. This configuration got me about 155.3dB @ 43hz. Thinking I could raise tuning I performed a bit of surgery on the port wall and opened it up to about 420 square inches. This gained me up to 156.8dB @ 45hz. I did some other minor tricks and am currently hovering between 157.0 and 157.4 depending on the day/weather.
In any event, my largest single gain so far has been increasing port area, and increasing port area overall gives me more options for rounding the mouth or tuning lower again if I miss the mark too high.
The catch being opening the port will again require some pretty major surgery on the box, replacing the baffle, and a couple hundred bucks in materials, so if there's a point where more port area won't help I'd just as soon save myself the money and heartache. It's pretty loud as it sits but I'd like to squeeze another full dB out of it before resorting to doubling power which opens the huge can of worms which is batteries right away.
Anyway, I'm interested in any input on port area vs. efficiency in this application. Consider the ONLY goal is gaining on the SPL meter (I'll also try just about anything else that'll gain). I have TS parameters of the subs and impedance sweep graph of the subs in the box in the vehicle if that'll help.
If SPL is the main concern, you can add quite a bit of port area to your current setup. This would trade SQ for more SPL at a centred frequency. Keeping port length the same obviously would raise the tuning frequency, of the enclosure.
Personally I would take one of the 15" drivers, put it in a new 6 - 6.5 cubic feet test enclosure and start adding port area until measurements show it isn't worth it any more. Then apply that to your main enclosure.
Personally I would take one of the 15" drivers, put it in a new 6 - 6.5 cubic feet test enclosure and start adding port area until measurements show it isn't worth it any more. Then apply that to your main enclosure.
That depends on the current box and port. Once the port is sufficiently large, there is nothing to be gained anymore. Do you know the air speed inside the port? Does the tuning frequency decrease at maximum drive level? If so, the port is too small.
That never happens. Think of a passive radiator, which behaves like a giant port. It can be 15" in diameter and have a mass of 1 kg and still behave properly.
I can easily lengthen the port to tune back down to 43hz if higher tuning hurts me. Most big-league SPL guys have a much higher peak frequency so I suspect raising tuning will help, then again much of this is vehicle dependent and this Jeep has always peaked around 43 with a dozen different boxes so that may be optimum?
Sadly the whole using a test box in the car is right out. I'm past the point of being able to put a box in this vehicle. Measuring near field outside is dead too since the pressure sensors we use for competitions are worthless below about 130dB so I'd have no precise way to measure slight gains or losses.
Besides which, a solid 6 cube test box with say 120 square inches of port is going to eat some time and a hundred bucks worth of materials. By the time I built and tested two I could just have re-worked the baffle and opened up the existing port.
I'm more looking for the theory or conventional wisdom on how big you can go with the port.
The original port of 300 square inches was sufficiently large to avoid chuffing, but I did gain a good 1.5dB by increasing it to 420"^2.
Would airspeed in the port be measured or calculated/predicted? That said, air speed is pretty violent everywhere in there. Low power testing shows me the same peak frequency as max power, even removing the seats only lowers my peak frequency by 1-2hz (though gains about 2dB) and the driver's seat blocks off much of the port.
Browsing this thread here:
Understanding - and exploiting the impedance curve?
I found the following comment: " After a certain point, the driver suspension can no longer continue to keep up with the mass of air moving out of the port, and the drivers start to lose control over it. Your efficiency starts to fall at this point."
Is this true? What is that point? Is it at a point where the port is the size of a doorway or something so extreme as to be a non-issue? Is this dependent on the woofers?
I'm certain that it isn't "too small" by conventional wisdom on box design, but practical testing shows that bigger port is gaining me output. My main concern is if there's a "too large" point where I'll start to lose.
Don't laugh at the gorilla tape, it gains 2-3 tenths of a dB for me and is a common SPL competition trick.
Thanks for the replies, guys.
Hey. I'm still learning certain details about ports behavior, and more often than not, I read a lot of misleading stuff or even clear BS. So more often, I try to experiment on my own, and it pays off. The problem with your situation is, that in your car, things and behavior change. So my knowledge about ports can easily break down in a car.
The idea about trying your findings in a box outside car is then only half good. And I actually did this. If you want to squeeze most SPL from your ports, you need to find best size/volume and stiffness qualities ratio. If you are volume limited, your larger port will not add as much SPL as your smaller box will take, because you took the space by the port. If you're not limited, it's practical to have up to twice as much port area, as your speaker cones have. Then it's really diminishing returns. But you propably cannot afford it in your car, and this can change in your car, as not only port SPL does your final measurement, but all that volumes, car shaping etc.... Anyway if you can, don't hesitate to add on your current port area. It should still add some SPL.
The idea about trying your findings in a box outside car is then only half good. And I actually did this. If you want to squeeze most SPL from your ports, you need to find best size/volume and stiffness qualities ratio. If you are volume limited, your larger port will not add as much SPL as your smaller box will take, because you took the space by the port. If you're not limited, it's practical to have up to twice as much port area, as your speaker cones have. Then it's really diminishing returns. But you propably cannot afford it in your car, and this can change in your car, as not only port SPL does your final measurement, but all that volumes, car shaping etc.... Anyway if you can, don't hesitate to add on your current port area. It should still add some SPL.
On the efficiency note - ports are not really efficient, and if you wanted compete at efficiency, you would not want to measure at port frequency anyway. Larger port will give you more SPL, but it will also give you lower impedance, thus more loading capability. The speaker takes more power with larger ports.
300 square inch isn't exactly large for 4 x 15" if SPL is your main concern. Similar sized ports can be found in cabinets that aim at 130 - 140 dB. If a 40% larger port gains you 1.5 dB, it's reasonably safe to assume that another increase of that magnitude will increase SPL further, though by a lesser degree than the previous step.
This can easily be simulated in a program such as Hornresp, which will tell you the velocity of the air in the port, as well as the expected gain and resulting tuning. Hornresp will likely tell you that you can easily double or triple the port area.
I was assuming that you wanted to find the maximum available port gain, in which case, in my experience simulations fall short of actual measurements. In that case I would build a single test cabinet, that doesn't necessarily has to be located inside the same vehicle. I wouldn't build several but rather a single cabinet that can be adjusted (for say 3 different port surface areas). In my experience this shouldn't have to cost more than 30 - 50 bucks.
A quadrupling of drivers (and power) will gain you 12 dB on average, meaning that a single cabinet will still give you output in the 144 dB range (when in a similar car).
It's also true that most vehicles have a say in what frequency you should tune at, which in this case appears to be 43 Hz.
Best regards Johan
This can easily be simulated in a program such as Hornresp, which will tell you the velocity of the air in the port, as well as the expected gain and resulting tuning. Hornresp will likely tell you that you can easily double or triple the port area.
I was assuming that you wanted to find the maximum available port gain, in which case, in my experience simulations fall short of actual measurements. In that case I would build a single test cabinet, that doesn't necessarily has to be located inside the same vehicle. I wouldn't build several but rather a single cabinet that can be adjusted (for say 3 different port surface areas). In my experience this shouldn't have to cost more than 30 - 50 bucks.
A quadrupling of drivers (and power) will gain you 12 dB on average, meaning that a single cabinet will still give you output in the 144 dB range (when in a similar car).
It's also true that most vehicles have a say in what frequency you should tune at, which in this case appears to be 43 Hz.
Best regards Johan
if the primary goal is in-car response,
and you've obviously done a lot of work optimizing
the boxes themselves, to the point where you are only finding limited gains right now,
why not take a look at optimizing the volumetrics, dimensions and other characteristics
of the vehicular interior itself?
WRT "in-house" loudspeaker characteristics,
a lot of the MLTL, BIB, and Synergy-oriented folks here at DIYA
seem to have paid a lot of attention to
in-room [edited rest of this line] location, orientation, and other treatments to good effect.
in an even smaller space,
payoffs may be even more likely to be found,
at lower material cost and fabrication times?
and you've obviously done a lot of work optimizing
the boxes themselves, to the point where you are only finding limited gains right now,
why not take a look at optimizing the volumetrics, dimensions and other characteristics
of the vehicular interior itself?
WRT "in-house" loudspeaker characteristics,
a lot of the MLTL, BIB, and Synergy-oriented folks here at DIYA
seem to have paid a lot of attention to
in-room [edited rest of this line] location, orientation, and other treatments to good effect.
in an even smaller space,
payoffs may be even more likely to be found,
at lower material cost and fabrication times?
Excellent news. The modification to increase port area will add about as much volume to the box as the increase in port volume so I'll maintain 24-26 cube box volume.
As you say, in a vehicle there's limits and mine are about 33" X 44" on the baffle so out of that there's only so much cone and port area to be had.
Can you elaborate please? Impedance as it sits is around 2 ohm at 45hz. My amp will safely drive 1 ohm (probably less if I felt daring). Are you saying I can lower impedance and get more power or that I'm getting more power for more power's sake with lower impedance and I want higher impedance to be more efficient? At the end of the day a higher number on the meter is the primary goal here.
I'm still a bit unconvinced that I could build a box so cheap. I'd figure 2 sheets of MDF/Ply at 35-40$ a whack plus some glue and screws.
Of course, if the consensus is I can open up this port up to say 600 square inches without fear of losing I'll just go straight for that for the cost of 3 sheets of MDF/ply, glue, screws, and a can of spray paint.
First of all, "limited" gains may be a bit of a misnomer here. Above 150dB gains become hard to find, and harder still above 155. Once you're at about 160dB many folks are happy to gain only 1dB from doubling power, so at this point I'd be elated at a full dB gain from my efforts and pretty pleased with .5.
You are very correct eluding to "the 'room' is the most important factor" the caveat being modifications ahead of the B-pillar are strictly limited in competition organizations (for this very reason)! There's loads of things that can be done to get some significant gains, many of which would make the car un-driveable and pretty much everything I haven't done would bump me into "extreme" classes (competing against concrete filled doors, 3" thick bullet proof glass windows, etc.).
I've taken care of the fundamentals. The wall is flat (V shape is proven inferior) and port on driver's side is known to be optimum for passenger side on the dash metering. I've done about as much as is allowed for my class and is practical for me to still use this as a daily driver. If you look at the world record SPL vehicles you'll see they indeed take sculpting the room to the acoustic goal to the extreme with impressive results. The few things remaining that I could improve in front of the wall are going to be very costly.
Thanks so much for the info, guys. I'm definitely going to see how much more port I can get into here in the next couple weeks. I'm eager for any other wisdom or suggestions on the topic.
How about some pics of this rig? Also, how are you bracing this box? I ask because I see competition setups all the time with very little bracing. I would think controlling the box from flexing would reduce losses and increase spl.
Hispls: correlation between impedance, power, SPL and efficiency is quite tricky one, as you handle four variables at once, which is hard to do at once anyway. It depends on what you need. I don't know that. If only SPL, you can kind of leave most part of the efficiency.
Making larger ports will cause impedance to go lower, so power will go higher, and SPL will go higher too, using two factors. 1. More power, 2. Less power compression at the port side. I'm not sure about efficiency though, it might go lower at the same time.
As the port has maximum SPL at lowest impedance point, it's obvious, that the speaker draws most power there. On the other hand, on an imaginary flat response alignment, you have two peaks at impedance curve, while SPL doesn't follow these. It's about flat. Therefore at the right impedance peak, which can get in 100+ Ohm ranges, the speaker still plays loud, but takes very little power. That means It's very efficient. Usually speakers with high Re,very high Bl, and very high Qms, are efficient.
Making larger ports will cause impedance to go lower, so power will go higher, and SPL will go higher too, using two factors. 1. More power, 2. Less power compression at the port side. I'm not sure about efficiency though, it might go lower at the same time.
As the port has maximum SPL at lowest impedance point, it's obvious, that the speaker draws most power there. On the other hand, on an imaginary flat response alignment, you have two peaks at impedance curve, while SPL doesn't follow these. It's about flat. Therefore at the right impedance peak, which can get in 100+ Ohm ranges, the speaker still plays loud, but takes very little power. That means It's very efficient. Usually speakers with high Re,very high Bl, and very high Qms, are efficient.
Almost all my pictures are on Photobucket which is now pay-per-view. I'll throw a few into imgur and post later though, perhaps I'll get some useful input.
Indeed ANYTHING flexing is a huge loss for SPL (and any SQ as well). Conventional wisdom is that for straight SPL purposes turbulence in the box hurts so you'll see many are shaped smooth, rounded everywhere, and lack any sort of internal bracing. I did 3 layer sides with some external bracing which meets with the sides of the vehicle, 3 to 4 layers at the top glued to the ceiling, 3 layer bottom glued and screwed to the floor, 4 layer back, and 5 layer baffle. In retrospect the 5 layer baffle was inadequate and I had a local welder build me up some brace pieces. I did some testing and the most onerous piece in or out gives me the same number so at a guess losing some from flexing but gaining it back from better air flow? Never enough data to be sure...
When I rebuild I plan to go 8 layers on the baffle, two of which to flush/indent mount the subs which generally gains. I'd like to stay away from internal bracing to avoid the issue of turbulence spots inside and also to keep the volume inside the box more usable as cargo area when needed and easier for me to climb into and work when and if I need to fiddle with things.
In short, most folks rely on multiple layers and external bracing of wood or often steel cages (or filling with concrete) for these sorts of applications. Some apply internal bracing and use turnbuckles, pipe, or threaded rod to minimize disrupting the airflow.
Here's my measured impedance curve (coils are dual .7 ohm), though I'm a bit off from flat response. These subs as I built them to survive high power/pressure just don't do wide bandwidth flat and trade a fair bit of efficiency for power handling, though they have motor force to spare. Yet another audio-compromise. About as good as I can do with these motors to get a musical driver is building them up 18" with rather soft suspension but they require such a huge box I've never had much luck getting 18s loud.
External bracing to avoid air turbulence in the box, smart!. Having a diy home theater background, we never consider this. Thanks for sharing that info.
Hey. The curve looks like if the port was very undersized (doesn´t go low in the impedance dip between two peaks, but it can also be caused by the car "enclosure".
I´am aware of tuning for SPL competition. My The flat response was just easier to use for my point about efficiency, as curved response brings even more variables to the situation. (But still works in the same manner). Yes. Large driver with soft suspension will most often work best in huge enclosure. Not everybody can afford these sizes. Me neither...
I´am aware of tuning for SPL competition. My The flat response was just easier to use for my point about efficiency, as curved response brings even more variables to the situation. (But still works in the same manner). Yes. Large driver with soft suspension will most often work best in huge enclosure. Not everybody can afford these sizes. Me neither...
Bear in mind in addition to being very ugly on a nicely finished home theater cabinet, we have completely different goals.
According to a local guy who has been doing big builds for 20+ years and a world record tier competitor internal bracing is preferable if you're building for music since it prevents too much peak around tuning in such a build. The theory may well be hokum, but he is a voice of experience and his builds speak for themselves that he knows both how to get loud and be musical.
Interesting. Do we go by multiples of the DC resistance of the coil when examining the impedance curve or some other metric?
For the record FS of these subs is 44hz broken in and 47hz fresh (had to replace a couple broken ones recently so they are mismatched). Z-max is 26 ohms with coils in series (2.8 ohm DC)
Pic dump commencing
We used scraps of stuff I had laying around to build up to the edges of the vehicle. Doesn't need to be pretty there, just functional and it is easily hidden.
First attempt at port wound up tuned very low. Despite your best efforts to guess it's difficult to predict airspace. Best method is to go as big as you can and shrink where needed.
Batteries up front are great at managing the weight but definitely not optimum for sound. Ideally the wall is dead flat. Also as you'll see later I carpeted the top of the battery box initially. More recently I replaced the carpet with painted MDF and gained a couple tenths.
Another sub-optimum feature is that this "window" can be removed off the back of the box for maintenance access and to keep it available for cargo space if I do a road trip and need some suitcases or whatever. Assuming I can get the port opened up wide enough I will likely seal this completely.
Amps have been in a state of flux. The 6 channel moved up out of the way and I replaced the 8 smaller amps (one per coil) with one large one which simplified greatly and gained half a dB though it did require adding more batteries.
The Zed amps use overbuilt and regulated power supply and didn't care about input voltage, the new Korean class D is far more sensitive and current hungry.
This was opening up the side of the box to gain port area (my big gain), and what I intend to do to the other side to move the subs over to the passenger side another 3-4 inches and gain another 100 or so square inches of port.
Don't be fooled by the "rated" power of this amp, it's a monstrosity and should make about 100V at 14.8V, more at 18V (if I ever opted to go 18V charging system).
I'm buying a couple more of these amps broken (for a very good price) and getting them repaired locally they'll be bench/load tested whenever that gets done.
- Status
- Not open for further replies.