T18's

I've put this in general for discussion, not sure if sub or bass would be right.
Covered this on the other side ages ago, got head up arse comments.
T18's as I understand are a sort of BP. A long time ago we loaded some with the cone facing out. In doing this they had the following traits:

Two side by side, useless. One on top of the other, useless. Bottom one upside down, useless. Top one upside down, massivly loud and loads more, "throw". Four square like this, top two upside down, enough 90's bass for a damn big venue and loads of, "throw". We did use them up to 250hz (I know, we were inexperienced) but I don't think that would have made any difference as EV actually reccomended it.
When the drivers were loaded magnet out they didn't seem to have these traits. It WAS a long time ago, should we have run them out of phase or something? Even one bin loaded cone out used to, "throw" loads more than magnet out.
I checked this with BP Sound who still has four loaded cone out, he confirms, top ones upside down in a group of four chucks it very far.
What on earth is the reason????

Regarding the first part, whether a sub or bass cab is usually signal dependent. In a system, I always assume that if there's a sub cab, there will always be a bass cab in the system. Hence the term sub bass. a lot of people don't see it like that. The EV cab was designed many years ago when you only had a bass cab on the bottom of your system.
Regarding stacking, With any kind of horn (ish) cab you never know how they will turn out until you put them together. Regarding throw, four will always, due to acoustic coupling, be better than two.
Because of the size of low frequency waves, I find it hard to believe that changing the way cabs of the same design are stacked can make a difference, but it does. My own experience, years ago, involved my BS600 cabs. I also had a block of four to consider which way up they would stack. Many people including me, in the old days would want to create a large horn by stacking two one sided horns apposed to each other. I had my old RCF analyzer with me so I checked all configurations. The end result had me stacking them up all the same way so that the end of the curve was facing the floor. I guess that we can conclude that cancelation, phasing etc. is horn length and shape dependent. This situation wouldn't be expected from set of reflex cabs.

It's nice on here, thanks for the explanation Tony.