Exceeding driver Xmax, and multi-cab setup query

[LawrenceH]

Hi there,
Looking to upgrade my amp to something lighter and louder, I managed to pick up a Tecamp combo thanks to guitarguitar's offer (discussed elsewhere). Thinking about the limitations of my previous kit and looking for cost-effective solutions, before this deal came up I was thinking about self-build. Finding the amazing wealth of information online about DIY loudspeaker cabinets and the superior performance of these to many 'big name' cabs has been fascinating and I am still wondering about adding a home-made extension cab. Reading discussions from Alex, Greenboy, Bill Fitzmaurice, Bgavin and others has been very illuminating, but I am struggling to find answers to a couple of questions, perhaps someone here can help.

1) Particularly with the modern high-end Neo drivers, roughly how far can one exceed Xmax a. most importantly safely, but also b. useably in musical terms with bass guitar?

I ask because playing with driver parameters in WinISD I notice that with a typical reflex cab with a tuning frequency somewhere around the driver Fs, excursion rapidly rises below the box tuning frequency, effectively placing a lower limit on the safe operating frequency the box can handle. However there is a hump roughly an octave or less above this which for lower box tunings acts more as a limit on the wattage the unit can handle. This is of particular concern because with typical cab tunings of 40-50Hz the excursion 'hump' is placed right in the range of the lower string first harmonics, where guitar output is very strong. Lots of my cab designs look good freq response-wise but the hump will skirt over Xmax at this point, at the wattage I would likely feed it when running full-tilt. I'd love a rule-of-thumb estimate of what's acceptable. I've read somewhere Xmax plus 10% but this didn't seem to be referencing PA drivers particularly, and I've read elsewhere that Xmech can be 2 or even 4 times as much as Xmax with some of these! If Xmax is currently typically measured as THD=10%, then this seems to an extent slightly arbitrary without understanding how the THD versus displacement curve up to Xmech plots out or how THD relates to musicality.

2) My second question relates to combining multiple full-range cabs: with cabinet matching, is the main concern for bassists one of matching group delays?

I realise that bi-amping with crossovers is the 'gold standard' here but with single channel micro-amps that's not an option. Lots of what I've read says that combining multiple cabs is only acceptable if they contain the same drivers and presumably are identical in size and tuning. Different frequency responses of different drivers will give the cabs different colorations but that in itself is not necessarily a problem, in fact when attempting to extend the bottom of a smal combo with an extension cab it could be beneficial. But I can see that group delay at the lower frequencies could result in nasty cancellation or booming. I'm guessing that what matters is aligning the phase plots of two cabs versus frequency, but I'd like to know if I'm barking up the wrong tree or if there are any other serious problems.

Many thanks to anyone who has bothered to read all this!

[Mr. Foxen]

Isn't the answer to the first one xlim?

[quote]Xlim is expressed by Eminence as the lowest of four potential failure condition measurements: spider crashing on top plate;vVoice coil bottoming on back plate;vVoice coil coming out of gap above core; or the physical limitation of cone. A transducer exceeding the Xlim is certain to fail from one of these conditions. High pass filters, limiters, and enclosure modeling software programs are valuable tools in protecting your woofers from mechanical failure.[/quote]

The second bit, if you have the cabs, try it and see, get a long cable and wander about, to se if there are bad cancellations. The people madly criticising mixing cabs and drivers and such are either people in pro audio that are very fussy on the minutiae of good sound, or people who have read their statements and taken it to mean the world will implode of you go against them, because they are experts. In most cases, you won't cause many real problems mixing cabs. Some stuff is going to upset totally modelled stuff, but is fine in practice, for example, if you do the highly reccomended pair of 2x10s stacked on their sides, the bottom one is radiating in half space cause of the floor, but the top one isn't, so they are mismatched.

Edited November 9, 2009 by Mr. Foxen

[LawrenceH]

[quote name='Mr. Foxen' post='649325' date='Nov 9 2009, 01:13 AM']Isn't the answer to the first one xlim?



The second bit, if you have the cabs, try it and see, get a long cable and wander about, to se if there are bad cancellations. The people madly criticising mixing cabs and drivers and such are either people in pro audio that are very fussy on the minutiae of good sound, or people who have read their statements and taken it to mean the world will implode of you go against them, because they are experts. In most cases, you won't cause many real problems mixing cabs. Some stuff is going to upset totally modelled stuff, but is fine in practice, for example, if you do the highly reccomended pair of 2x10s stacked on their sides, the bottom one is radiating in half space cause of the floor, but the top one isn't, so they are mismatched.[/quote]

Ah, yes, sorry I should have been clearer - many manufacturers don't produce Xlim figures so is there a reasonable 'guesstimate'? But I also don't understand if the modelling of input power versus excursion beyond Xmax can be considered accurate, if not Xlim could be reached much earlier than expected. I'm thinking really about what's acceptable for the hump above the tuning frequency.

Re cabs, the thing is I'd rather not build it to try out without having avoided obvious pitfalls! I definitely hear what you're saying re 'real world' and it's good to hear that you find mixing can work in practice. OTOH my experience of sound engineering with a bassist's ears has certainly shown me how bad some set-ups can be at the low end! What I've learnt is that it's worth getting each component of the signal chain as close to optimum as is reasonable/affordable, as errors are additive, but beyond a certain point for each individual component the price/performance ratio places a limit on what's worth it and time/money's better spent elsewhere (e.g. a semi-decent entry level condenser mic on vox always makes an improvement compared to the SM58 with even the crappiest PA, but a Neumann+pre through a guitar amp probably isn't worth it!).

Thanks a lot for the reply!