Hf Ratio

Mosts specs of most drivers will give SPL @ 1W/1m figures, and by doubling power (1W, 2W, 4W, 8W, etc) you will get 3dB more for each doubling of power. They will either give max SPL (ignoring power compression), or you can calculate something similar.

A lot of HF drivers will be measured with some form of horn attached, usually stated in the spec. If your choice of horn differs from the one it is measured with, its the total dispersion that will affect SPL figures. If the measured horn is a 60x40, and yours is 60x40, very little difference in SPL. If measured is 40x40, and you want to use a 90x60, perhaps allow 3dB less SPL from your horn/HF combo, as the same power from diaphragm is being spread over a bigger surface of a sphere, a greater area of mouth, than measured. SPL, is, simply, power divided by area, same power over double an area, will be -3dB.

Picture is 1000 words - this site explains a lot of the fundamentals of sound physics. Should be OK to follow, just keep looking at it, until you get it!!

Working from that, you should be able to work out how loud each driver will contribute to the total system SPL. You will want every frequency band, to give a similar SPL. If it doesn't, you can pad down too loud bands on your LMS, or, if you have enough headroom before amp clip or driver physical damage, add gain at the LMS.

One extra thing, is for every doubling in number of drivers, you get a 3dB gain in output. So a single 12" may give 95dB 1W/1m for a particular box/passband. 2 x 12" would give you 98 dB 1W/1m, 4 x 12" would give you 101 dB 1W/1m, etc.

Equally, one of those 12" may give 95db 1W/1m, so 98dB 2W/1m, 101 dB 4W/1m, 104 dB 8W/1m, etc.

Equally, inverse square law is 6dB loss per doubling of distance, so the same 12" may give 95db at 1W/1m, 89dB at 1W/2m, 83dB at 1W/4m, 77dB at 1W/8m, etc. Roughly 20dB down for 10x distance.

All this is strictly reflex, a horn cab may add 3-6db to the efficiency of the driver, depending on design, but measured, or Horn Response, may give a good estimate of 1W/1m numbers for system design help. However, apart from a change to the initial 1W/1m figures, all other rules will generally apply.

Horn coupling effects tend to increase horn mouth related shizzle, so no great gain above the normal 3dB for doubling the number of drivers/horns, in terms of SPL, but gains in passband flatness, and low cut off frequency.

If you want it sound "heavy on sub", simply allow a required 3 or 6dB louder pass band, for required SPL.

So something like 133dB HF, 130dB Mid, 136dB Kick, 142dB sub design goals for EDM? +3dB is electrically double (1W, 2W, 4W, 8W, etc) but +6dB is perceived by the human brain as being audibly double. Most trained ears can perceive a change of around 1 to 1.5dB as a change in volume. Very good ears can hear a 1-1.5dB change in a graphic fader, but jury is out on whether that is straightforward perception of the change of SPL at a frequency, or whether you are hearing the group delay change as the fader is moved at whatever frequency on the GEQ.

The only other thing to consider, is you may not need the same SPL from 20Hz to 20KHz. Look-up Fletcher -Munson curves for how the ear perceives 2-4K far more easily than 20Hz or 20KHz. You then use a thing called the phon, which is dB SPL, corrected for brains perception of a frequency.

Further spanner, is that the F-M curves and phon are different depending on absolute SPL, the louder the noise, the less 2.5K you need!

I will let you have a cup of tea after that lot......

The art lies in the nuances of how to get drivers in different passbands to work with the units in the above and below passbands.

I'm not very artistic, so have to try and keep things on a theoretical physics domain. 

That is a fairly basic primer on the basics of decibels, but its a highly useful set of concepts to get you to a basic design, which then gets you a good starting point to evolve from. Train, then trust, your ears.

Real elegant design will match details like horizontal dispersion at X-over points, so the -6dB points are at roughly similar angles, so off axis response is just a quieter version of on axis. Equally phase/group delay at X-0ver, through SMAART or equivalent measurement, aids inter passband integration for pleaseing and constructive progression from one driver to the next.

The basics are well within the grasp of most, the real hardcore details sets the few above the rest. Just a bit of physics.

At full tilt the low mid would produce 123.5db @ 225w-1m and the HF 130db @ 25w-1m.

So using those what would the calculation be to combine the two and get one number for the cab in it's entirety? I appreciate that it might not be 100% accurate...