Dayton 18" HO Ported Build Thread

[Rob Harrison]

Fyi... i am a civil engineer and my dad was a carpenter and cabinet builder. I always have the drive to try and improve things...

Question: is there any data out there with respect to the measured pressures created by sub drivers in ported and closed boxes?
Not spl. Actual pressurization/rarification of the air within the sub cab....

Thanks, rob

 

[Rob Harrison]

OK... did a little research. Gonna throw some science (perhaps flawed, not sure) at you guys on the pressure thing and whether we really need to brace as much as our guts tell us we should (some more than others).... again.. I tend to overbuild and go the extra mile for minor gains.

What levels of SPL could we be reaching at the port of the sub? This is where the pressure wave exits to atmosphere, and would be the highest pressures in the sub (agreed?).

Would it be safe to assume that it's no more than 140 dB? If so.. that converts to a pressure of approximately 200 Pascals.

Dumbing this down because I live in 'Merica.... that's a pressure of approximately... wait for it.... yes... 0.03 pounds per square inch.

Do we agree with this??? This actually sorta surprises me... I was kinda thinking it should be higher.

This surprises me. So much that I question my supposition that the bracing we are all envisioning is to keep the deflections of the enclosure to a minimum - thereby transmitting all acoustic energy out the port to avoid dreaded (insert marketing-hype words like "coloration") due to cabinet deflection.

Looking at this from a statics problem, let's find out what kind of forces are acting on the box.

Across the large side of a sub box.. say 36 inches by 24 inches... that's 864 square inches.
This would be a total force on the side from the 140dB pressure wave of 864 sq inches x .03 pounds/sq inch = 26 pounds force. Recall this is not a point load, but is an evenly-distributed load across the side.

I don't have a piece this size laying around (right now), but I just put a 25 pound dumbbell on a 36" wide by 12" piece of 3/4" MDF supported only at the ends and I measure less than 1/8" deflection. Again.. this is extreme. The loading on a sub sidewall is distributed (point load on my test case) and there is significantly more support on the four sides of a sidewall than my test board. My gut tells me that a distributed 25-pound load on a 36" x 24" wall of a subwoofer cabinet would produce very little deflection.

Put another way... do you think if we made a large MDF box, sealed it, plumbed it with a schraeder valve and used a bicycle pump to increase the pressure in the box to 0.03 PSI, would we see deflection in the sidewalls?

Help me out here... is this sound science or did I go down the escher rabbit hole of physics?

So... if deflection isn't the real problem, maybe it's just a simple mass issue that decreases the minute deflections/vibrations which are the true cause of sonic degradation in a cabinet.

Help me out here....

Thanks,

Rob

 

[fuzz092888]

It's not deflection you're trying to get rid of per se. It's to keep the side walls from resonating. Bracing not only creates a stronger box, which is extremely important in sealed applications with big drivers and lots of power, but it breaks the panels into smaller sections which pushes the resonant frequency beyond what the subwoofer will ever play. In terms of side wall deflection, I estimate it's less of an issue with ported boxes, but with my sealed subs, even at around 250lbs they'll shimmy and shake with the sheer force of the back waves combined with the actual movement of the drivers....and that's with 2.25" thick baffles bracing etc.

 

[TheWarrior]

OK... did a little research. Gonna throw some science (perhaps flawed, not sure) at you guys on the pressure thing and whether we really need to brace as much as our guts tell us we should (some more than others).... again.. I tend to overbuild and go the extra mile for minor gains.

What levels of SPL could we be reaching at the port of the sub? This is where the pressure wave exits to atmosphere, and would be the highest pressures in the sub (agreed?).

Would it be safe to assume that it's no more than 140 dB? If so.. that converts to a pressure of approximately 200 Pascals.

Dumbing this down because I live in 'Merica.... that's a pressure of approximately... wait for it.... yes... 0.03 pounds per square inch.

Do we agree with this??? This actually sorta surprises me... I was kinda thinking it should be higher.

This surprises me. So much that I question my supposition that the bracing we are all envisioning is to keep the deflections of the enclosure to a minimum - thereby transmitting all acoustic energy out the port to avoid dreaded (insert marketing-hype words like "coloration") due to cabinet deflection.

Looking at this from a statics problem, let's find out what kind of forces are acting on the box.

Across the large side of a sub box.. say 36 inches by 24 inches... that's 864 square inches.
This would be a total force on the side from the 140dB pressure wave of 864 sq inches x .03 pounds/sq inch = 26 pounds force. Recall this is not a point load, but is an evenly-distributed load across the side.

I don't have a piece this size laying around (right now), but I just put a 25 pound dumbbell on a 36" wide by 12" piece of 3/4" MDF supported only at the ends and I measure less than 1/8" deflection. Again.. this is extreme. The loading on a sub sidewall is distributed (point load on my test case) and there is significantly more support on the four sides of a sidewall than my test board. My gut tells me that a distributed 25-pound load on a 36" x 24" wall of a subwoofer cabinet would produce very little deflection.

Put another way... do you think if we made a large MDF box, sealed it, plumbed it with a schraeder valve and used a bicycle pump to increase the pressure in the box to 0.03 PSI, would we see deflection in the sidewalls?

Help me out here... is this sound science or did I go down the escher rabbit hole of physics?

So... if deflection isn't the real problem, maybe it's just a simple mass issue that decreases the minute deflections/vibrations which are the true cause of sonic degradation in a cabinet.

Help me out here....

Thanks,

Rob

In response to your first question, above this one, that 'data' would be the T/S (Thiele-Small) Parameters that are given for every driver. This data tells you what type of box, size, gives necessary info for crossover design, and much more. But, it requires a level of understanding to properly implement.

Do not think of a ported enclosure as a 'squeeze box' for sound. You are not squeezing air out of the port to create sound. Sound is a vibration which propagates as a wave of pressure AND displacement. That port is tuned, to a specific frequency based on that driver's T/S parameters which allows the back wave to exit the enclosure, at just the right speed, extending the low frequency response of a driver. But not all drivers can be ported, effectively.

A brilliant electrical engineer friend of mine commented on my enthusiasm for speakers design as a 'scary art'. There is so much choice to be made that speaker design is as much an engineering feat as it is an art due to the number of choices that can be made, but more so because of how obsessive you can get trying to achieve 'perfection'.

Speaker design is a very deep rabbit hole, but for me, my tastes in speakers have now far exceeded my budget. So DIY will be my solution until we learn how to just beam sound in to peoples minds!

 

[Johnny2Bad]

First of all, excellent build, and excellent build thread. I'm only posting this in the interests of helping someone new to carpentry and woodworking who might want to DIY an enclosure in the future, or follow your example and build a clone of this sub itself.

RE: Haoleb's Post #19
" ...

I have multiple measuring tools but decided to just stick with the good ole' tape measure. Bad thing about a tape is that if you are going for accuracy you can't just hook on and go because I have yet to use a tape measure that is accurate this way. There is always some slop in the end. Have to hold one inch over (or how ever many you want) Then be sure to add on the other end. Lots to remember. And since I am using a guide I also have to subtract the distance from the guide to the sawblade. Hence... Lots of double checking.
..."

----------

A carpenter's tape is designed to accurately measure both inside and outside measurements. That is why the metal lip at the end moves.

In fact, if you look carefully at your photo you can see exactly what is going on.

Outside measurement example: Find the point to mark a cut 18" wide along the length of a 4x8 sheet of MDF. To do so, you hook the lip of the tape on the outside of a panel, pull the tape across the panel and measure 18" from the edge.

You will get an accurate mark if you simply trust the tape and mark at the 18" point.

Inside measurement example: Find the point 18" from the inside corner of your enclosure to place a brace. To do so, you push the lip against the inside wall of your corner, (the metal lip will move inward) and let out the tape beyond your 18" point. * You can then mark where the tape reads 18" and obtain an accurate point exactly 18" from the inside corner of your enclosure.

How is this possible? Look at your photo. Notice that in the first inch, the first inch of the tape is not one inch long. The "1 inch area" is actually a little shorter than one inch, you can see it in your photo ... go ahead, count the 8ths; there are only 7 of them (plus a bit of a 16th).

The gap between the L-shaped lip and the beginning of the tape markings is ... you guessed it ... the "missing" sixteenth-ish marking on the tape. So you can put the lip on the edge of your material and get an exact inch at the marked inch, or any value further down the tape.

For the inside measurement, when you push the "L" against your inside corner, it moves and takes up that gap. Now look again ... the "L" will be exactly the thickness of the portion missing from the marked inch on the tape.

The tape is very accurately manufactured so that the thickness of the "L" lip and the dimensions the lip can move in and out are correlated to an accurate inside and outside measurement. All you need to do is make sure you take up or let out the slack by pushing or pulling on the lip as required.

There is no additional accuracy gained if you are measuring your desired 18" as 19" less 1".

* If the 18" point is not visible on the tape because an obstruction prevents you from fully extending the tape, you can determine where the 18" point is because the length of the body of your measuring tape will always be indicated on the unit.

On my Stanley Leverlock 25', for example, it's 3 inches or 76 mm, and that value is embossed in the plastic so wearing of the label will not make that information unavailable to me. So if I extended it to 15", I know that the body will end at exactly 18", and I can make my mark at the far end of my measuring tape's body.

 

[Johnny2Bad]

It's not deflection you're trying to get rid of per se. It's to keep the side walls from resonating. Bracing not only creates a stronger box, which is extremely important in sealed applications with big drivers and lots of power, but it breaks the panels into smaller sections which pushes the resonant frequency beyond what the subwoofer will ever play. In terms of side wall deflection, I estimate it's less of an issue with ported boxes, but with my sealed subs, even at around 250lbs they'll shimmy and shake with the sheer force of the back waves combined with the actual movement of the drivers....and that's with 2.25" thick baffles bracing etc.

Yes you want to move the resonance frequency of any panel (or object for that matter; such as the driver basket) beyond (higher in frequency) the frequencies the sub is expected to generate.

Bracing is an example of one method that can move a sympathetic resonance higher up in the audio band.

This makes the resonance unlikely to be excited by something the sub is playing.

Also the program information falls in energy as frequency rises. So in a typical music performance there might be 100 dB of bass at 60 Hz but probably not 100 dB of harmonics at 10 Khz. This makes the resonance less likely to be excited by other drivers playing in the room, like a full range speaker that is part of a 2.1 system (for example).

This is because almost all instruments (and the human voice) have the highest fundamental at or below 1Khz, the very, very few that can go higher are done by about 4 Khz. So all 10KHz energy (excepting possible synthesized notes) is a harmonic, which will be at a much lower level than a fundamental. This means the energy required to excite the resonance is less likely to be experienced in playback.

Note that I'm just using 10Khz as an illustrative example, you probably would have difficulty moving a 60 Hz sympathetic resonance up to 10Khz, but regardless, higher frequency equals less liklihood of excitation as the energy present in the program material falls as frequency rises, regardless of exactly where you are able to move that resonance upward.

Even moving a 60Hz resonance to 120Hz would mean the overall SPL would have to be higher to excite that resonance than it would have to be had you did nothing to address it.

 

[TLS Guy]

You have some misconceptions about resonant systems.

In resonant systems pressure and displacement are out of phase. The high pressure areas are where there is no displacement that is the antinode, there is a node of displacement. At the antinode of displacement, (the port mouth) there is an antinode of pressure.

There is a lot of data showing the type of bracing you intend to use is not only useless, but often harmful.

Bracing of the type you contemplate is like placing your finger on the violin string and increases the pitch of the resoance and adds harmonics..

Figure of 8 bracing, it bracing of that type is best. You need to dado the braces over the entire length of the board and interlock the panels.

 

[Haoleb]

A carpenter's tape is designed to accurately measure both inside and outside measurements. That is why the metal lip at the end moves.

In fact, if you look carefully at your photo you can see exactly what is going on.

Outside measurement example: Find the point to mark a cut 18" wide along the length of a 4x8 sheet of MDF. To do so, you hook the lip of the tape on the outside of a panel, pull the tape across the panel and measure 18" from the edge.

You will get an accurate mark if you simply trust the tape and mark at the 18" point.

Inside measurement example: Find the point 18" from the inside corner of your enclosure to place a brace. To do so, you push the lip against the inside wall of your corner, (the metal lip will move inward) and let out the tape beyond your 18" point. * You can then mark where the tape reads 18" and obtain an accurate point exactly 18" from the inside corner of your enclosure.

How is this possible? Look at your photo. Notice that in the first inch, the first inch of the tape is not one inch long. The "1 inch area" is actually a little shorter than one inch, you can see it in your photo ... go ahead, count the 8ths; there are only 7 of them (plus a bit of a 16th).

The gap between the L-shaped lip and the beginning of the tape markings is ... you guessed it ... the "missing" sixteenth-ish marking on the tape. So you can put the lip on the edge of your material and get an exact inch at the marked inch, or any value further down the tape.

For the inside measurement, when you push the "L" against your inside corner, it moves and takes up that gap. Now look again ... the "L" will be exactly the thickness of the portion missing from the marked inch on the tape.

The tape is very accurately manufactured so that the thickness of the "L" lip and the dimensions the lip can move in and out are correlated to an accurate inside and outside measurement. All you need to do is make sure you take up or let out the slack by pushing or pulling on the lip as required.

There is no additional accuracy gained if you are measuring your desired 18" as 19" less 1".

* If the 18" point is not visible on the tape because an obstruction prevents you from fully extending the tape, you can determine where the 18" point is because the length of the body of your measuring tape will always be indicated on the unit.

On my Stanley Leverlock 25', for example, it's 3 inches or 76 mm, and that value is embossed in the plastic so wearing of the label will not make that information unavailable to me. So if I extended it to 15", I know that the body will end at exactly 18", and I can make my mark at the far end of my measuring tape's body.

Good tip, I got out the calipers to see just how far out it was on each dimension and it is only off about 20 thousandths if the end is kept square. Not bad considering the abuse I have put that tape through over a few years of use.

 

[Johnny2Bad]

You have some misconceptions about resonant systems.

In resonant systems pressure and displacement are out of phase. The high pressure areas are where there is no displacement that is the antinode, there is a node of displacement. At the antinode of displacement, (the port mouth) there is an antinode of pressure.

There is a lot of data showing the type of bracing you intend to use is not only useless, but often harmful.

Bracing of the type you contemplate is like placing your finger on the violin string and increases the pitch of the resoance and adds harmonics..

Figure of 8 bracing, it bracing of that type is best. You need to dado the braces over the entire length of the board and interlock the panels.

Talking to me? You don't say whom.

Let's assume you are.

This is trivially easy to demonstrate. Take any loudspeaker with an enclosure. With your right hand, do the "knuckle rap test" on the side wall and listen to the tone (frequency) and loudness (SPL) of the resulting sound.

Now place the index finger of your left hand somewhere near (ideally, not exactly at) the centre of that same panel. With your (free) right hand, repeat the knuckle test. Note the tone and loudness of the resulting sound.

The frequency will rise (for a sympathetic vibration the frequency in the program material must activate the resonant frequency of the item [panel, etc] under test, which will then self-resonate until it dies out naturally). Also the SPL will fall, meaning it will add less to the overall sonic picture even as it does resonate sympathetically.

By the way, best practice is to utilize a number of damping and resonance control methods; constrained layer, your choice of thin ("BBC" method*, many papers to read on their subject but in essence, use a thin panel that resonates at a high frequency) or a thick and dense panel that resists being excited but will resonate after whatever SPL and Program Material limit is met or exceeded.

There is also bracing, placement of smaller panels (eg a 4"x4" [inch] panel glued with Epoxy on a side wall that is larger than 4"x4" [inches]), applying various polymers and materials of varying density, eg cork, etc.

Basically, you just "go for it". Commercial manufacturers have to worry about material cost, manufacturing complexity (and therefore cost), etc but a DIY'er can do whatever on a 1-time build, plus he gets to listen and sort out his results vs his effort which is the always valuable learning of lessons that can be re-used in the next build.

* Used on the LS3/5A plus many UK Stand mounted ("Bookshelf") and Floorstanding monitors of the 1970's and later, right to this day.

 

[Rob Harrison]

TLS guy and Haoleb: I humbly ask for your help with regards to port length.

I have tried modelling this design so I may learn. I do not get the output/results TLS posted. In running several online models, my own spreadsheet using the vent length formula, etc. I am unable to reproduce the vent length of 37.29" using the following: Vb = 7.672 cft, Fb = 19.05 Hz, vent area 3" x 14".

I was using a k value of 0.732.
In order to get the length, the k needs to be 0.45ish.

I tried to educate myself some more and found several online articles regarding the use of the end-correction factor k. In a series of tests conducted and documented by Troels Graveson (you can search it easily) wherein he kept all variables the same while varying a port length and measuring response. The conclusion is that the calculated vent lengths/fb are off from reality and that the actual vent lengths to hit the design box tunings were anywhere from 50 to 80% of the calculated lengths.

I've also found some other posts that present a figure that appears to be annotated in German that presents end correction factors for various configurations. In the case of this RS446HO-4 design, the end-correction factor would be 2.227. Plugging this in as k in the vent calculators results in a vent length of 24.31 inches, keeping all other factors the same. This is 65% of 37.29".

I reviewed the response graphs measured by Haoleb on his sub with the hopes of identifying the Fb in there.... I don't see it. Could be some other things going on to mask it, such as room gain.

So.. after the long-winded setup, what are your thoughts on the vent-length calculations and end-correction factors?

I have read that the best approach is to get a suggested vent length, build it... test it... tweak it.... then finalize it. That's much easier said than done.

Appreciation,

Rob

 

[Sebastian]

I'd like to bump this thread because I'm building one. I have all the mdf parts of this design but I'd like to know how will hst18 perform in this enclosure. I think it needs over 300 liter enclosure and this is just 217 liters. But that hst18 looks like one badass sub :0 over 100mm total cone travel. I'd like to hear your opinions about the hst18 and this enclosure

Thanks!

 

[TLS Guy]

I'd like to bump this thread because I'm building one. I have all the mdf parts of this design but I'd like to know how will hst18 perform in this enclosure. I think it needs over 300 liter enclosure and this is just 217 liters. But that hst18 looks like one badass sub :0 over 100mm total cone travel. I'd like to hear your opinions about the hst18 and this enclosure

Thanks!

I did not design that sub for your driver.

You can not just swap drivers.

Changing drivers requires a total redesign.

If you use the hst 18 in that box it will sound awful. You have to use the driver I designed that box for, not ifs or buts.

I'm in the UK now and do not have the time to design a box for that driver now.

You have two choices.

Use the driver the box was designed for.

Wait until I get back in a little over three weeks.

 

[Sebastian]

I did not design that sub for your driver.

You can not just swap drivers.

Changing drivers requires a total redesign.

If you use the hst 18 in that box it will sound awful. You have to use the driver I designed that box for, not ifs or buts.

I'm in the UK now and do not have the time to design a box for that driver now.

You have two choices.

Use the driver the box was designed for.

Wait until I get back in a little over three weeks.

Aalright, I'm not in hurry. I'm new in DIY stuff so I'm trying to learn as much as possible. I could build that box with Dayton driver. (because I already have those parts slit) and then build second sub with HST18. I think behringer Inuke 6000 would be more than enough for those.

 

[ARES24]

More than enough, the dayton is spec'ed at 900rms/1800max. 3000 watts is ... wow. I run 500w/ch, my house shakes.....

 

[TLS Guy]

Aalright, I'm not in hurry. I'm new in DIY stuff so I'm trying to learn as much as possible. I could build that box with Dayton driver. (because I already have those parts slit) and then build second sub with HST18. I think behringer Inuke 6000 would be more than enough for those.

Aalright, I'm not in hurry. I'm new in DIY stuff so I'm trying to learn as much as possible. I could build that box with Dayton driver. (because I already have those parts slit) and then build second sub with HST18. I think behringer Inuke 6000 would be more than enough for those.

Unless you are in the demolition business the Dayton driver will be more than adequate. You have to remember that that sub design is very efficient and actually takes very little power. People think subs need to have gobs of power. Good ones do not. The fact is that most subs are terminally inefficient.

The amps driving my bass lines barely get warm even if playing a movie threatening damage to the house.

The other issue is that tuned boxes, pipes and horns are acoustic transformers. The pressure is very high behind the driver, especially in pipes and horns, and so the driver does not need to move much. Most of the output from is the port, or in a pipe the pipe mouth and for a horn the horn mouth.

Sealed enclosures do require a lot of displacement from the driver, as a loudspeaker cone is a very inefficient coupler to the room.

I would just go ahead and build that sub with the Dayton driver. Just remind me to do the design for the other diver and we will see what works out best for it.

 

[Sebastian]

Unless you are in the demolition business the Dayton driver will be more than adequate. You have to remember that that sub design is very efficient and actually takes very little power. People think subs need to have gobs of power. Good ones do not. The fact is that most subs are terminally inefficient.

The amps driving my bass lines barely get warm even if playing a movie threatening damage to the house.

The other issue is that tuned boxes, pipes and horns are acoustic transformers. The pressure is very high behind the driver, especially in pipes and horns, and so the driver does not need to move much. Most of the output from is the port, or in a pipe the pipe mouth and for a horn the horn mouth.

Sealed enclosures do require a lot of displacement from the driver, as a loudspeaker cone is a very inefficient coupler to the room.

I would just go ahead and build that sub with the Dayton driver. Just remind me to do the design for the other diver and we will see what works out best for it.

thanks! :3 Yeah I need to do some pencil work this weekend for drawing out the bracings. Lol Have these subwoofer served you good? I have had Klipsch rw-12d for about over 3 years or so and it just didn't have enough power for my use, for under 30hz that 12 driver travels a lot so I think it hits xmax too easily. And bass port was loose when I got it.

 

[Haoleb]

The subs have served me extremely well so far. Going on about a year and a half now. Now in my own home as well and I can crank it up as loud as I can stand. The subs are a very efficient design. I used to have a pair of the kappa 12VQ subs (see sticky at the top of DIY forum for the design) I was able to get the clip lights to come on with the same amplifier setup I am running now which is a single bridged QSC RMX1450a amplifier per sub. With these dayton subs I have never once seen the clip light come on.

Box design is still extremely ridgid. No cracks in the paint on any joints. No buzzes or rattles.. etc. And no issues what so ever!

 

[Rob Harrison]

I've had my 18" HO running for several months and love it. I feed it with a Carvin DCM1400 pro audio power amp, bridged. I believe the amp is rated at 1400 watts for a load of 4 ohms.

The 18" is 4 ohms.

I have the front volume controls on the amp at less than 1/4. The sub shakes my entire house. So, yeah.... it's super efficient.

Absolutely fun experience to watch movies....

When a heavy bass event occurs... it puts a smile on your face!

One of my favorites... it's extremely quick and short lived, but in Man from UNCLE... there is a scene when the guys get in an elevator and it starts going down..... there's a low note that just kills!

BTW... this HO build replaces a servo-controlled velodyne 15. The Dayton is way nicer.

I'm embarking on building a new sub for my son's Subaru... one 12" Dayton HF DVC sub will be replacing the current sub I made a few years ago housing a pair of 10" JL Audio W0 woofers. The JL box is 2.1 cubic feet. Will build a sealed sub around .5 cubic feet first for fun.... but I plan on making a vented sub in the near future.

 

[Rob Harrison]

BTW.... I stand on many shoulders in this group. Thanks everyone for the advice and posting your experiences and knowledge!

I am very grateful.

 

[Rob Harrison]

Sorry... made a mistake above. I'm using the 12" DVC HO sub from Dayton, not the HF.