Push Pull Subwoofer Build?

[Primare Knob]

You want the excursion to never exceed Xmax

Cone displacement is the working area of the cone ( SD ) multiplied by Xmax. It is the volume swept by the cone in one direction.

In speaker box light calculator there is a graph for Cone displacement in mm. I assume that it is not using volume but distance because the cone area is a fixed value?

Does that mean that reading this graph, if the xmax is smaller than the cone displacement it won't produce that frequency, or not without distortion?

I read somewhere, that you could use xmax +10% as a maximum limit.

[Primare Knob]

Europower amps are good workhorses but especially on EP4000 the fan is REALLY loud! Even with a fan mod, if willing to void warranty, all fans are still a bit noisy to be in the room. What's the plan? Putting them in a garage and running some cable is a good way to go.
 
If you want a small box, you picked the right version (HO), apparently by accident. These are car audio versions - don't be put off by that. It mostly means they are designed for small boxes.

And what the real benefits are of using 2 X 15" drivers vs 1 x 18". Am I overdoing it again as usual?
 

 
The advantage with dual driver subs is that you can reduce vibration of the enclosure by having them on opposite faces of the box - one on front, the other on the back. You can use this to either create a really inert box, or simplify bracing.
 
2x15 is likely to deliver more midbass headroom. Much of the action in films lies here. A single 18 is likely with more excursion to deliver more deep bass output and often budget will favour this option.
 
Impedance is also a factor, depending on what you are aiming for. With Europower amps, you want to end up with 4 ohms into each channel. 8 is limiting the power, 2 ohms is pushing your luck.
 
Expectations. I find that for home theatre, people expect more output and are more likely to use it. Films are mixed for large commercial systems with more dynamic range. In a music system, people are often more critical of quality. Partly because a movie engages our attention differently.
Is the distortion level how quality is measured in subwoofers?

I plan to put it in the room with a fan modification. I am sure that I could run a quiet 12V fan from an external power supply. Not sure what the amp will do if I leave the fan disconnected from the PCB.

[Primare Knob]

Just to clarify some basics. Is the cone extrusion the key element to low frequencies, and cone area to efficiency?

[Red Spade Audio]

4 hours ago, Primare Knob said:

 I plan to put it in the room with a fan modification. I am sure that I could run a quiet 12V fan from an external power supply. Not sure what the amp will do if I leave the fan disconnected from the PCB.

You may be happy with that but there are a few things to bear in mind. First, quieter fans move less air, meaning they are also less effective at cooling - reliability is compromised. It's not an issue for most people. Second, even if you get a fancy super quiet fan and spend a bit more, the noise isn't just dictated by the fan itself but also by the airflow within the amp itself. In other words, there is a limit to how quiet you can get due to the air moving through the grille and heatsink. It means you might not get it as quiet as you hope.

4 hours ago, Primare Knob said:

Just to clarify some basics. Is the cone extrusion the key element to low frequencies, and cone area to efficiency?

You can't really isolate those two factors, because both of them together along with other factors are needed to determine extension, SPL and sensitivity.

 

I'd put it this way. Displacement (cone piston area x maximum excursion) determines SPL. This is true for all frequencies, however for a given SPL, the displacement increases with decreasing frequency.

 

To get that extra displacement in the bass, you can increase excursion or piston area. More piston area gives you higher sensitivity. More excursion gives you more SPL at the cost of needing more power and compromising linearity.

For a given piston area, the sensitivity of the driver can be changed based on the intended bandwidth. You can maximise sensitivity with a strong motor and low moving mass. You might get an 8" midrange that can give you 100 dB sensitivity but that will only apply above say 500 Hz. On the other hand, a subwoofer will generally have sensitivity in the range of 83 - 90 dB.

 

Hoffman's iron law is a simple concept to keep in mind. It has three parameters - box volume, sensitivity and bass extension. You can choose only two and the third becomes a given. When you have chosen the parameters you have in mind, you can then choose the driver and enclosure design to suit.

[davewantsmoore]

5 hours ago, Primare Knob said:

Is the cone excursion the key element to low frequencies

Yes, although depending on the enclosure, power can also be the limit.

5 hours ago, Primare Knob said:

and cone area to efficiency?

Yes... larger cone is more efficient, all else being equal....  but it never is so the moving mass (of the cone and coil), and the force of the motor .... and also the sometimes the enclosure, are big factors.

[almikel]

On 02/11/2018 at 10:13 AM, Primare Knob said:

In speaker box light calculator there is a graph for Cone displacement in mm. I assume that it is not using volume but distance because the cone area is a fixed value?

it sounds like they're quoting excursion rather than displacement since it's in mm, but I've never used that tool.

 

On 02/11/2018 at 10:13 AM, Primare Knob said:

Does that mean that reading this graph, if the xmax is smaller than the cone displacement it won't produce that frequency, or not without distortion?

correct - I would regard Xmax as the upper limit of excursion without distortion sky rocketing.

At Xmax most sub drivers will still have quite high levels of distortion, but likely not audible.

 

On 02/11/2018 at 10:13 AM, Primare Knob said:

I read somewhere, that you could use xmax +10% as a maximum limit.

The driver can move beyond Xmax, and runs into mechanical limits at Xmech.

Only a very unsympathetic person would allow their speakers to come close to Xmech.

 

IMHO when designing a sub you need to plan for staying under the Xmax of the driver to reach your low frequency and SPL targets.

If during design/modelling you get close to or exceed Xmax, IMO change your design (more drivers, larger box, different box (eg ported or tapped horn etc), multiple subs etc).

 

With DIY you have the opportunity to "design in" plenty of headroom so your kit just cruises along well within its limits even at elevated volume.

 

 

cheers

Mike

[Primare Knob]

Can someone clarify for me if I can calculate a box size based on a single driver, if I plan to use 2 drivers in opposite polarity?

[Primare Knob]

Are there alternative free programs that produces graphs like speaker box lite?

[almikel]

On 26/10/2018 at 10:25 PM, Primare Knob said:

The reason for thinking about a push pull configuration is that I like the box to stay small.

I'm a simple person and a lazy DIYer - I need my DIY exertions to work at first go - "Oh that didn't work, I'll try something different" is not the sort of DIY I muck with.

Any "push/pull" design carries a risk of "bass suckout", and I just wouldn't go there.

 

I would recommend simple sealed monopole sub designs. If you get 2 drivers build 2 boxes. If you get 4 drivers build 4 boxes.

Use Speaker Box Light to model a single driver in a sealed box. As the box size shrinks you'll get "peaking" before rolloff, and the rolloff will be a bit steeper as the Qtc increases (Hoffman's iron rule at play).

You can use EQ to flatten the peak and extend the low frequency - the EQ "changes" the Qtc, so don't get caught up in the "perfect" Qtc of a driver in a box - muck with the box volume - see how small you can go.

 

My first "go-to" design tool for subs is Rod Elliot's version of the True Audio Linkwitz Transform excel spreadsheet.

http://sound.whsites.net/project71.htm

I've used this spreadsheet for decades as a quick and dirty modelling tool.

This tool won't help with modelling excursion - which is important when designing subs with EQ - but it will provide a quick guide on the dB amplifier gain required to maintain a flat frequency response with EQ using a Linkwitz Transform.

 

Using this tool here's your driver (Dayton RSS390HO-4) in a sealed box of 40 litres with no EQ and a Qtc of 0.73, and F3 of 48Hz

with EQ to hit 20Hz - around 16dB of gain required to hit 20Hz

 

Reduce box size to 30litres, Qtc rises to 0.82 and f3 rises to 49Hz

but eq to hit 20Hz increases to 18dB

 

Remember that a 3dB increase in power is double, and the numbers get big fast with 18dB of gain

• if the driver needed 1W before rolloff then adding 18dB of gain requires 64W
• if the driver needed 10W before rolloff then adding 18dB of gain requires 640W

@davewantsmoore had some numbers in a recent thread for the driver excursion increase to maintain SPL as frequency drops - I can't find it

At least double the excursion for each octave lower?

 

mike

[Primare Knob]

Can someone clarify for me if I can calculate a box size based on a single driver, if I plan to use 2 drivers in opposite polarity?

I think I can answer that for myself. The answer seems yes, although I don't fully understand why besides the simple fact of using 2 drivers. Does it have to do with the pressure?

It also seems that I have used the term Push Pull wrong as I am looking to do a dual opposite design.

[Primare Knob]

I am not sure how reliable speaker box lite is.

I am looking to use the Dayton RSS390HO,

An Qtc of 0.707 uses 33.35 liter of internal volume.
Using 2 drivers doubles the volume to 66.7 liters but lowers the Qtc. Is that correct?

The SPL graph isn't showing an increase of +6dB when using 2 drivers as I would expect.

The web version no longer is showing the SPL graph, but the app is. And the web version yesterday showed 3 more graphs that aren't there today.

Power and SPL seem more than adequate for my preferences. My preferred listen volume is about -12.5 dBFS, which would mean a peak SPL of 102.5dB.

[almikel]

19 hours ago, Primare Knob said:

Are there alternative free programs that produces graphs like speaker box lite?

WinISD

[almikel]

19 hours ago, Primare Knob said:

Can someone clarify for me if I can calculate a box size based on a single driver, if I plan to use 2 drivers in opposite polarity?

a single driver in a box is a monopole alignment.

Add a 2nd driver with opposite polarity you have a type of 'isobaric" alignment where the pressure inside the box remains the same as 1 driver moves out and the other driver moves in...highly likely to produce bass suckouts in the room - try it if you want - my lazy DIY gene says don't bother.

You want all sub drivers moving in the same direction at the same time - increasing or decreasing pressure in the room.

"Funky" designs of push/pull may work - but will need experimentation/prototyping.

 

I had a large woofer box with 2x15" drivers in the same cabinet but separated inside. When I replaced the drivers and fired it up I couldn't believe how poor the bass was, until I realised one driver was out of phase creating massive suckout. 

 

Monopole subs are easy to design and will produce predictable results - IMHO stick with the well traveled road.

Sealed monopole enclosures have lots of flexibility in size - you can go with very small enclosures if you have EQ/amp power/driver excursion/multiple subs.

 

mike

[almikel]

20 hours ago, Primare Knob said:

An Qtc of 0.707 uses 33.35 liter of internal volume.
Using 2 drivers doubles the volume to 66.7 liters but lowers the Qtc. Is that correct?

No - using 2 drivers and doubling the volume will keep Qtc the same

edit - assuming a "normal" box alignment - 2 drivers on the front baffle.

Edited November 4, 2018 by almikel

[davewantsmoore]

22 hours ago, Primare Knob said:

Can someone clarify for me if I can calculate a box size based on a single driver, if I plan to use 2 drivers in opposite polarity?

There are no hard and fast rules like this.    Sliding the box size up and down will adjust the ratio of SPL to input power.... and will also change the frequency response (only relevant if you won't be adjusting the frequency response with some type of filter / crossover)

 

Quote

The reason for thinking about a push pull configuration is that I like the box to stay small.

I wouldn't really suggest that as a relationship you should focus on from the outset.

 

 

It helps to think about some rough acceptable ranges for size, frequency range, and SPL ..... and then see what designs can hit it (and then adjust your ideas of size, frequency, and SPL, as you realise there is no free-lunch.

 

20 hours ago, Primare Knob said:

An Qtc of 0.707 uses 33.35 liter of internal volume.

What Qtc tells you, is "this is the shape of the frequency response" around the resonance frequency of the system.

 

....  If you are going to apply some type of EQ to change the shape of the frequency response, then Qtc is irrelevant.

 

You will see people focus on Qtc sometimes because they don't plan on EQ'ing that part of the frequency range......   For example a 2-way 6" woofer and 1" tweeter in a box with a passive crossover.    There is no EQ applied to the speaker below 100Hz.... and so the frequency response will be whatever the box tells it to be (ie. Qtc).

 

Edited November 4, 2018 by davewantsmoore

[davewantsmoore]

53 minutes ago, almikel said:

Add a 2nd driver with opposite polarity you have

 

.... a dipole.    Just like an "open baffle" speaker.

[davewantsmoore]

20 hours ago, Primare Knob said:

Power and SPL seem more than adequate for my preferences. My preferred listen volume is about -12.5 dBFS, which would mean a peak SPL of 102.5dB.

Yes, if you've really calibrated your system to full scale being 115dB.

 

~102dB@20 is do able with a reasonable excursion 15" driver.     The vast majority of how it will sound is to do with how it's positioned in the room (and whether or not there are big issues), and what EQ is applied to it.

[almikel]

20 hours ago, Primare Knob said:

My preferred listen volume is about -12.5 dBFS, which would mean a peak SPL of 102.5dB.

subs usually need to cope with higher SPLs - especially with HT

http://www.acousticfrontiers.com/2013314thx-reference-level/

 

Mike

[almikel]

1 hour ago, almikel said:

Add a 2nd driver with opposite polarity you have

 

10 minutes ago, davewantsmoore said:

 

.... a dipole.    Just like an "open baffle" speaker.

and as a lazy DIYer, dipole bass is too hard to get right - so I'll stay with  monopoles for subs - much easier

 

Mike

[Nada]

On 26/10/2018 at 10:25 PM, Primare Knob said:

 I am looking into building two subwoofers for HT use, and I am thinking .....

You might save a lot of time, some money and pain by paying for expert input:

 

http://www.redspade.com.au/audio/rythmik.php

 

http://www.rythmikaudio.com/DS1500ci.html

[Primare Knob]

a single driver in a box is a monopole alignment.
Add a 2nd driver with opposite polarity you have a type of 'isobaric" alignment where the pressure inside the box remains the same as 1 driver moves out and the other driver moves in...highly likely to produce bass suckouts in the room - try it if you want - my lazy DIY gene says don't bother.
You want all sub drivers moving in the same direction at the same time - increasing or decreasing pressure in the room.
"Funky" designs of push/pull may work - but will need experimentation/prototyping.
 
I had a large woofer box with 2x15" drivers in the same cabinet but separated inside. When I replaced the drivers and fired it up I couldn't believe how poor the bass was, until I realised one driver was out of phase creating massive suckout. 
 
Monopole subs are easy to design and will produce predictable results - IMHO stick with the well traveled road.
Sealed monopole enclosures have lots of flexibility in size - you can go with very small enclosures if you have EQ/amp power/driver excursion/multiple subs.
 
mike

I believe that Isobaric designs have only one woofer in contact with the outside air, and one woofer in a sealed enclosure without contact with the outside air, to keep a constant pressure in the enclosure used for the other driver.

Why would 2 drivers moving in the same acoustical direction create suckouts in a room?

System is calibrated at reference level, and I am aware about the extra 8dB required due to the use of bass management.

I understand that simple things are easier to build, but I like to build a quality sub, wether that is easy or not. I don't like the idea of 4 subs in my room, and don't really need them.

Still have lots to learn, and likely gonna adjust my ideas once I understand things better.

[Primare Knob]

What Qtc tells you, is "this is the shape of the frequency response" around the resonance frequency of the system.
 
....  If you are going to apply some type of EQ to change the shape of the frequency response, then Qtc is irrelevant.
 
You will see people focus on Qtc sometimes because they don't plan on EQ'ing that part of the frequency range......   For example a 2-way 6" woofer and 1" tweeter in a box with a passive crossover.    There is no EQ applied to the speaker below 100Hz.... and so the frequency response will be whatever the box tells it to be (ie. Qtc).
 

At the moment my subs are placed at the half way length of the room. Including my listen position it needs 3 EQ filters, for a rather flat response.

My biggest problem with my current subs, is that they hardly go louder below 50Hz over 95dB as it seems to hit a limit. The other problem I have is that in the low frequencies, there is a lot of undefined rumbling that doesn't sound very pleasant, which I assume is distortion.

I understand that box size isn't the best starting point for a subwoofer build, but large subs will make it rather unpractical in my small room.

What I want to achieve is undistorted fast transient response bass at a max peak level of 105dB over the whole LFE frequency range. I am still learning what parameters to look for that are related to this.

I am assuming that EQ-ing the sub can be done in software, including cross overs, phase settings and gain, and bypass the need for analogue parts. I am probably thinking to easy on this, as usual.

The whole Qts idea, is coming from reading a few posts and articles that low Qts values are desired for clean and fast responsive bass with little overshoot.

[Primare Knob]

You might save a lot of time, some money and pain by paying for expert input:
 
http://www.redspade.com.au/audio/rythmik.php
 
http://www.rythmikaudio.com/DS1500ci.html

I know, but that also takes away a lot of the learning path. Plus my stubbornness hasn't caved in yet [emoji15]

[Primare Knob]

 
.... a dipole.    Just like an "open baffle" speaker.

But will it still be a di-pole if both driver move in the same acoustical direction?

[davewantsmoore]

10 hours ago, Primare Knob said:

But will it still be a di-pole if both driver move in the same acoustical direction?

No.

 

That is just "two monopoles".    If two monopole are close together (measured in wavelengths of the frequency of interest - as they are for you), they can be considered "one monopole"