1x12" Cab Design Diary

[Phil Starr]

Hi, this is a 'diary' of how we are designing our 1x12. It is a bit behind the actual discussions we are having but I'll catch up as quickly as I can.

I’ve posted this in response to our earlier thread. Three of us at Basschat (Stevie, LawrenceH and myself) have decided to publish a series of open source cab designs and people asked us to describe how we created the design so they could follow our reasoning. This is going to be the design diary for the 1x12.

The first decision was which cab would be first. I initially suggested 1x15, a 1x12 and a 2x10 as initial projects. Stevie and Lawrence were both planning an 8” cab so that will be added in as an early design and there was some discussion about 2way designs.

In the end the decision to go for a 1x12 was an easy one. We’ve got a range of 12” drivers around and a single driver in a straightforward box makes an accessible start to home building. We should be able to come up with a design that will keep up with a drum kit on its own and be enough for small-medium gigs. Adding a second cab or building it into a 2x12 would cover most people’s needs.

Initially we didn’t have a design spec, but what has arisen through our discussions are these:

Capable of keeping up with an unamplified drum kit (120dB across most of the frequency range)
Compact (60l or less)
Neutral /clean sounding
Easy to handle
Good low frequency power handling
Readily available components, and easy to build
Value for money

We didn’t specify power handling. The critical thing is not how big an amp you have to use but the actual volume of sound you can expect the speaker to produce before it farts out. If it is loud enough with just 50W then this would be a positive, if it handled 1000W then it would probably be loud enough but fail the value for money test. As it happens nearly all 12's handle about 300W and have similar sensitivity of about 95dB/W.

Easy to handle ties in with weight, but also size and proportion. You need to be able to lift your cab, but it also needs to fit through doors, including car doors, and not drag along the ground. We decided that 60l was the biggest our 1x12 should be and this would intrinsically make the weight and size manageable.

Good low frequency power handling is a difficult thing to define. Almost no commercial 12” design will go down to low B at -3db or better and be capable of that at 120dB and ours won’t either. ( I don’t know of any commercial bass speaker that will but someone else will) What we are looking for is something that won’t shake itself to bits at high power and reproduces low E cleanly and at a decent level. We also want to see something that can handle a decent proportion of its rated power without distorting . More about this later.

UPDATE 

A lot of information has been lost in the last couple of years with changes to both this site and Photobucket which I put my old pics on. If you want to build these cabs I've retrieved the dimensions below. I need to go back and see if I can come up with something better when I get time but it ought to be possible to build one of these cabs from just the dimensions. If you want to do a build PM me and I'll fill in any details

•  
• Members
• 60
• 2,401 posts
• Location: Chard,Somerset

• Report post
•  

Posted July 8, 2015

If anyone wants to start a build before I've had time to write this up and get drawings done it shouldn't be too difficult. The cab is constructed out of 12mm ply and fitted with a Beyma SM212 driver available from Blue Aran and others.The external dimensions are 360x424x530mm the baffle front is set back 30mm from the front of the cab so internally the depth is 306mm. The four ports are made of standard guttering downpipe 190mm for a 50Hz tuning, and 260mm for a 40hz tuning. All this information is elsewhere in the thread but it's probably hard to find. I'll cut and paste the important bits later but everything is there below including some frequency plots measured from the prototypes.

 

Edited January 9, 2019 by Phil Starr

[Phil Starr]

To be honest I won't pretend the search for a driver was quite an objective trawl through every available 12” speaker. We’ve all built cabs before and between us had a number of 12” drive units sitting around to try. We had a fair idea of what we were looking for, though if we had turned up a speaker dramatically better we would have used it. For clarity I’m making our search look more systematic than the process probably was

It makes designing more interesting though when you know you have to justify your decisions to other people who probably know more than you, and who won’t mind saying what they think!

The first thing to do is to trawl through the spec sheets, keeping a sceptical eye open for manufacturers over optimistic claims. There are two things to keep an eye open for; the overall ‘sound’ of a speaker is determined roughly by its frequency response, and the bass performance by the design of the motor cone and suspension.

Most speaker manufacturers publish the frequency response as a graph http://www.eminence.com/pdf/Kappalite_3012HO.pdf The thing to look for in the frequency response is the shape of the graph above 1000kHz, which is where our ears are more discriminating in what we hear. Below this the shape also depends as much on the box as the speaker itself and they aren’t all measured in the same box. If you look carefully on the spec sheets there’s usually some small print to tell you how the tests were done.

The first thing is to see how high the response goes, if you don’t have a decent output (within 6dB) up to 3000Hz then your bass is going to sound as if the tone control is rolled right off before you start. A lot of 12” drivers are designed to be the bass unit in a PA speaker so they don’t need any top and you can reject all of these for a single speaker design like ours. Ironically if the speaker says bass driver or LF it probably isn’t going to work as a bass guitar speaker. Ideally I’d like an output within 6dB up to at least 4K.
You’ll see two types of response. All speakers have a fairly smooth responset up to 500Hz ish which we think of as mids then the response starts to wobble. This is where the cone won’t work any longer as a piston, usually starts to flex and and resonate, and more of the sound comes from the central part of the cone. Depending upon the cone design the response will either bulge upwards by a few dB’s falling off at somewhere over 2K or be fairly even, with a bit of a wobble.
The bulge is really useful for adding colour to the sound and most commercial speakers have the upper mid bulge, You have to decide early on whether you want a coloured ‘old school’ sound or something more neutral.


To predict bass response I like to look at three parameters in the spec sheets, [b]Fs[/b] is the resonant frequency and tells you how deep the speaker will go. It isn’t an absolute limit but once you put a speaker in a cab it will start to droop in bass output around and probably above this point.
The next thing to look at is [b]Xmax[/b] (and Xlim if the manufacturer gives it). Xmax is about how far the cone can move before the voice coil moves out of the magnetic field. This is important because the sound will distort and compress at this point. Be careful because different manufacturers have different ways of specifying Xmax. Some simply measure the magnet and the coil which gives a worse case Xmax, others like Eminence give Xmax under a certain level of distortion which makes them look a bit better but, to be fair, a bit more ‘real world’. Where we could we chose to use the voice coil overhang measurements but they weren’t always available. If Xmax is low (3mm or less) the speaker isn’t likely to handle a lot of deep bass without distorting at high sound levels, so we wouldn’t look at these for bass unless we were designing a multiple driver cab.
The next thing we looked at is a weird parameter called [b]Qts[/b]. I won’t try to explain it here but a high Q (above 0.5) means the speaker is poorly damped. It will only give a flat response in a big box, in anything of a practical size it gives a big bass hump arund 100Hz and will sound warm, woolly, and very old-school. A low Q speaker (below 0.3) is going to be well damped. Just like a sports car tracking the road it is going to track the bass well, but the bass output will be tight and fall in output from quite a high point above its resonance. Just right for this bear is around the 0.35-0.4 point. By and large a powerful magnet creates low Q and better damping of cone movement so low Q speakers are more expensive and heavier.


The last thing to look at when choosing a speaker is the efficiency, how much sound you get for a watt. This is usually given as dB/W. Look at the graph ideally at the flat bit below 500Hz which gives a better idea of bass efficiency. Most manufacturers give an efficiency which includes the midrange humps and bumps, this makes their speaker look better than others that don’t use this trick. Nearly all the speakers we looked at were around the 95-98dB/W mark and there wasn’t much variation in sound levels once we modelled them.


So taking all of this into account we came up with a shortlist of candidates, which had to go up to 3kHz, have a Qts between 0.3 and 0.5, an Fs of less than 55Hz and an Xmax of at least 4mm. All the speakers we looked at had an efficiency above 94dB/W. I’ll run through the choices we modeled next time.

[Musky]

I'm looking forward to this massively, so thanks to you Stevie and Lawrence for putting the effort in.

[funkle]

Awesome. Thank you for sharing your thought processes.

[ebenezer]

i too am looking forward to this!....just for info...I swapped the 12 inch driver in my aguilar gs 112 for a eminence 3012 ho kappalite expecting good results!!....
all i can say is,i was unhappy with the sound(yes i know the stock driver is 4.8 xmax and 95db......3012 6.2 xmax and 100.5 db)...i found it only a small
amount louder than stock driver.....sound wise,it did not mesh well with tweeter and crossover...too aggressive in the mids and the bottom end not that
good.....3012 now back in the box!....returned cab back to standard and sounds great!.....i think correct cab design plays a large part .it seems the lesser
deltalite 2 might have been a better option(looking on talkbass)

regards....mark

[Muzz]

Fascinated by this. Thanks for the effort.

[Phil Starr]

[quote name='ebenezer' timestamp='1390557034' post='2346686']
i too am looking forward to this!....just for info...I swapped the 12 inch driver in my aguilar gs 112 for a eminence 3012 ho kappalite expecting good results!!....
all i can say is,i was unhappy with the sound(yes i know the stock driver is 4.8 xmax and 95db......3012 6.2 xmax and 100.5 db)...i found it only a small
amount louder than stock driver.....sound wise,it did not mesh well with tweeter and crossover...too aggressive in the mids and the bottom end not that
good.....3012 now back in the box!....returned cab back to standard and sounds great!.....i think correct cab design plays a large part .it seems the lesser
deltalite 2 might have been a better option(looking on talkbass)

regards....mark
[/quote]

Hi Mark, the 3012HO is one we are modelling. It's a good example of what I was saying, Coincidentally the link http://www.eminence.com/pdf/Kappalite_3012HO.pdf is for that driver. If you look at the graph the sensitivity below 500Hz is only 98dB/W and when we modeled it it was slightly below this. The 100.5 they quote includes the mid peak, it's fair enough because they tell you how they test but also misleading. I wouldn't have expected much increase in sensitivity. To get the best out of it you would have had to have modified the crossover which was designed for the stock driver.

Ideally we'd like to run tests on the 3012HO driver but we don't have one to hand. If yours is still back in its box then pm me.

[6v6]

This is great, thanks for sharing!

/me can feel another cab build coming on...

[stevie]

[quote name='ebenezer' timestamp='1390557034' post='2346686']
I swapped the 12 inch driver in my aguilar gs 112 for a eminence 3012 ho kappalite expecting good results!!....
all i can say is, i was unhappy with the sound(yes i know the stock driver is 4.8 xmax and 95db......3012 6.2 xmax and 100.5 db)...i found it only a small amount louder than stock driver.....sound wise,it did not mesh well with tweeter and crossover...too aggressive in the mids and the bottom end not that good.....3012 now back in the box!
[/quote]

I’ve run both of these drivers through my modelling program and can show you what happened when you changed your speakers. You'll notice that the difference between the two speakers is largely in the midrange, with the Delta 12LF looking fairly balanced throughout its frequency range and the Kappalite HO looking decidedly mid-prominent (the Kappalite is the green curve).



The most interesting portion of the graph is below 100Hz. Here the drivers are very close – as you’d expect because this is where the cabinet dominates system performance. What this means in practice is that if you want to maintain sensitivity but get more bass extension you must increase the size of the cabinet. If you want to keep your existing cab you can only increase bass extension if you are prepared to accept a reduction in overall sensitivity (as Acme have done, for example). You can't simply fit a more sensitive driver and expect it to produce more bass.

It is worth pointing out that the cabinet Phil is building will allow the use of a range of drivers, allowing you to choose the type of sound you want within reason. We will be modelling different drivers, verifying their published specifications and showing you how they measure in a finished cab.

Edited January 24, 2014 by stevie

[Thunderpaws]

This is all well written, interesting stuff (for dumb asses like me). Thanks v much for doing this...looking forward to seeing the results.

Once you are finished designing the 12 can you design a 108 or 110 for home practice?

[LawrenceH]

A 108 (or perhaps two) will hopefully appear at some point. I have already built a pair of 110s, loosely documented on basschat somewhere, that might represent a suitable basis for a 110 cab design - however we would need to identify alternative drivers since the Celestions I used, although very good, have unfortunately become prohibitively expensive.

Those who read or commented in the 'Ideal Cab' thread, those comments were taken on board - but note that the initial prototype will be focused on getting the basic design elements right. A lot of the requirements people had were to do with ergonomics which, although important, can often be incorporated with relatively minor adaptations at a later point at the cost of increasing complexity or budget.

My feeling so far is that Phil, Stevie and I each come at this from slightly different angles, though they may well disagree? But my assessment is that Phil has pushed for a cab that prioritises as straightforward as possible construction and very solid performance on a budget, making the design as accessible as possible. Stevie (at least in his own designs) probably prioritises absolute sound quality above all else, using very nice drivers indeed. As for me I like to maximise portability, which, if it's not to significantly compromise sound quality, increases build complexity and often cost. We also each have slightly different tool sets. I understand Phil has a fairly good workshop for prototype construction and has a lot of experience as a joiner. Stevie has a decent testing rig and a range of commercial software that will allow us to test drivers, model, and (importantly) subsequently assess the cabinet. Meanwhile I have much cheaper measurement hardware and have used free software for my own designs, which may be of interest to those basschatters who would like to do their own thing without such a significant initial outlay. In particular, when it comes to a 2-way design, I have not seen any documentation of the crossover design process on basschat, which might well be of interest to other DIYers.

[LukeFRC]

[quote name='LawrenceH' timestamp='1390592296' post='2347440']
A 108 (or perhaps two) will hopefully appear at some point. I have already built a pair of 110s, loosely documented on basschat somewhere, that might represent a suitable basis for a 110 cab design - however we would need to identify alternative drivers since the Celestions I used, although very good, have unfortunately become prohibitively expensive.

Those who read or commented in the 'Ideal Cab' thread, those comments were taken on board - but note that the initial prototype will be focused on getting the basic design elements right. A lot of the requirements people had were to do with ergonomics which, although important, can often be incorporated with relatively minor adaptations at a later point at the cost of increasing complexity or budget.

My feeling so far is that Phil, Stevie and I each come at this from slightly different angles, though they may well disagree? But my assessment is that Phil has pushed for a cab that prioritises as straightforward as possible construction and very solid performance on a budget, making the design as accessible as possible. Stevie (at least in his own designs) probably prioritises absolute sound quality above all else, using very nice drivers indeed. As for me I like to maximise portability, which, if it's not to significantly compromise sound quality, increases build complexity and often cost. We also each have slightly different tool sets. I understand Phil has a fairly good workshop for prototype construction and has a lot of experience as a joiner. Stevie has a decent testing rig and a range of commercial software that will allow us to test drivers, model, and (importantly) subsequently assess the cabinet. Meanwhile I have much cheaper measurement hardware and have used free software for my own designs, which may be of interest to those basschatters who would like to do their own thing without such a significant initial outlay. In particular, when it comes to a 2-way design, I have not seen any documentation of the crossover design process on basschat, which might well be of interest to other DIYers.
[/quote] YES, yes indeed

[ebenezer]

looking at the graph,it seems to me the delta 12lf is, for the price, a cracking driver! (a recommended replacement for aggie cab). All this begs the question are we expecting too much from the humble 12 inch driver?

[LawrenceH]

[quote name='ebenezer' timestamp='1390642471' post='2347774']
looking at the graph,it seems to me the delta 12lf is, for the price, a cracking driver! (a recommended replacement for aggie cab). All this begs the question are we expecting too much from the humble 12 inch driver?
[/quote]

That's the danger with modelling graphs, they tell you nothing about soumd quality. The delta LF variants are straightforward pressed-steel chassis mid-woofers with moderate extension. Not bad, but nothing special either. But there is truth in what you say about what to expect from a single 12" driver. Two mediocre 12" units will very easily match or exceed what a very expensive one will do alone in terms of output capability, and it becomes a question of what your threshold for required peformance levels is- hence the 120dB approx goal for a single cab.

Edited January 25, 2014 by LawrenceH

[LukeFRC]

I'm going to say... can this thread be pinned please mods?

[paul_5]

Better off pinning [i]and[/i] glueing...


I'll get me coat.

[Phil Starr]

[quote name='LukeFRC' timestamp='1391449422' post='2357292']
I'm going to say... can this thread be pinned please mods?
[/quote]

thanks.

It might be better to create a new thread once the design is finalised with the design at the top, then we can drop further designs into that thread. We can link to this diary from there. Our hope is that enough people will build these and then tell us of anything we might do to improve future designs.

[sifi2112]

I also want to see the list of suitable driver ... please

Enjoyed the read so far ... gonna start work on my next cab soon.

[stevie]

I can give you a list of the 12" drivers that are in our possession and that we may be trying out:

Beyma SM212
Ciare 12.00NdW
Celestion TF1225
Celestion TN1230
DAS Audio P12
Eminence Beta 12
Eminence DeltaLF
Eminence Kappalite 3012LF.

It's not a list of recommended drivers though. When we look more closely at these it will become apparent which ones are suitable and which are not - and why. This should help you choose your own drivers in the future.

We can't test every potential driver out there - so our list of recommendations will necessarily be limited to what we can get our hands on. However, there are a couple of drivers that we would like to test but which we cannot really justify going out and buying. These are the Celestion BL12-200X and the Faital Pro PR300. If anyone has one of these and wouldn't mind lending it to us when we reach the measuring/testing phase, that would be great.

[redstriper]

[quote name='stevie' timestamp='1390571323' post='2347001']
you can only increase bass extension if you are prepared to accept a reduction in overall sensitivity (as Acme have done, for example).
[/quote]

This is what interests me, I want a small(ish) light(ish) cab with deep bass extension for dub reggae, no need for much upper mids or treble.
I am happy to use a more powerful amp in compensation for the lower sensitivity.
The Acme flatwound 12 sounds perfect, but expensive and not available in the UK.

I would be very interested in an Acme clone or similar cab and I presume the Eminence Kappalite 3012LF would be the most suitable driver?
On a practical note - how do you reduce the overall sensitivity of such cabs?

[stevie]

Comparing the measurements in issue 8 of Bass Gear Magazine, the bass extension of the Fearful 12/6 - which uses the Kappalite 3012LF - is quite close to the Acme 12 - which uses a version of the Kappalite 12. So I reckon you would get fairly close. The Acme is tuned very low though, about 33Hz I think, which is about 15Hz lower than the Fearful. I'm not sure how much that contributes to what it does.

We can give you some guidance on experimenting with cab tuning. I prefer to tune for power handling myself, whilst Lawrence seems to favour a shelf-type bottom end that reaches lower by tuning lower.

There are various ways of reducing the sensitivity of a drive unit. A less powerful magnet, a more flexible suspension and a heavier cone are the main ones.

[Owen]

The ACME is lush! I have one and I love the low tuning. If importing another from the US was not so painfully expensive, I would have 2. The low end is pretty addictive. I have experienced the less efficient nature of the cab and flogged amps which were not happy to drive it, but I have no DB amplification GAS now (apart from the desire for a 2nd one). A low tuning option on this build might appeal to some people who were happy about the efficiency trade off.

[LawrenceH]

Actually I'd probably favour a smooth roll-off above all else, because they are easiest to integrate in a mix in my experience. I like the sound of sealed cabs in this respect. So for a reflex cab, a lower tuning if it reduces the shoulder in the response curve or pushes it out of bandwidth. I don't like a shelf with a sudden drop below tuning. Tuning for power handling makes perfect sense though and is what I've done in the past.

btw acme's own measurement data is not in total agreement with the bass gear magazine measurement - I'm not sure how the tests differed but one shows what is presumably baffle-step-mediated increase in midrange sensitivity while the other does not. I find these cabs' design philosophy interesting, and am curious as to how this approach would differ in subjective terms from a similarly low-tuned but more sensitive driver, suitably eq-ed to reduce the midrange response to comparable levels.

[redstriper]

Thanks all, I'm learning as I go......... so another stupid question that I just can't get my head round:

Why does a cab have to be less sensitive to be more bassy?

Is it because the volume is reduced when you filter out the higher frequencies and those frequencies are louder than the lower ones?

[Balcro]

[quote name='redstriper' timestamp='1391554584' post='2358641']
Why does a cab have to be less sensitive to be more bassy?
[/quote]

In the simplest terms it's because of interactions between 2,3,or 4 of the Thiele / Small parameters. These determine everything that happens. The lower the fundamental resonance (Fs) the greater the ability of the drive unit to produce low bass, e.g 40-80Hz at the same sound level as it does at 100Hz. The slope down from say 100HZ can be gradual, steep or shelved, it depends upon the system tuning, but the result is not in doubt.

You have a given strength in the magnet / motor-assembly. A powerful magnet will exert great [u]control[/u] over the movement of the voice coil and hence the back & fore movement of the speaker cone. If you make the cone heavier, the motor assembly still remains the same, likewise the front and rear suspension of the cone, but stopping & starting the cone (back and fore) is harder because of the increased moving mass of the cone.

The result is a lower Fs. To move the increased mass of the cone as far and as quick as before, you have to increase the electrical signal coming in (more watts needed). The drive unit needs more power input to sound as loud as before. It's less efficient.

This is analogous to taking a modern European car. Without changing anything else, stick 200Kgs more than normal over each suspension point and it will behave like a 50's American sedan. It'l bounce up and down. How do you make it go as fast as the European car; you put more power in with a bigger engine!


Well it's something like that.

Balcro