No point ever made using a spectral analyzer as its backbone has impressed me, period. I've taken two very different sounding amp tones (Vox combo vs. Marshall stack) playing the same part (reamped from the same DI) and compared them using a spectral analyzer. These tones are about as different as you can expect, but the spectral curves were very very similar to the eye. It's not that the curves were not different at all, but the graphs were unremarkable compared to the obvious audible differences, and didn't always correlate in ways one would expect. This case and situation has repeated itself over many times in my experience, and as such I no longer care for the use of spectrum analyzers. The human ear can detect many things that analyzers seem to miss, and this is again a perceived difference, just like you said in your original post, that's the best point you've made so far. It's also something I'm very familiar with, and I use it frequently when performing Reamping/IR/Mix work.
Spectrum analyzers are not intended to look at frequency response, they are used to look at total power and how that power is distributed across the bandwidth that you are analyzing. To properly analyze frequency response you are better off plotting the raw PCM data of the system when ran through a sine wave sweep, this is how such frequency response charts are done with pro audio gear like speakers and microphones. Even then you can't rely on VST spectrum analyzers either, the real, $2K+ devices will be way more accurate and will give much better readings than any cheap VST will do. Still a VST plugin will still do a good job at letting you see how the power in the system is distributed among the bandwidth you are using.
A small difference to the eye, but in actuality 80Hz to 50Hz is a difference of almost a full octave. This actually deflates your argument in my eyes quite a bit as you are already making potentially large differences seem smaller to the uninformed readers.
Octaves are an interesting topic, the sense of an octave is the result in how our brains process sound. Out minds, our brains are logarithmic in nature, although sound, energy etc are linear devices in most cases. While the differences may be large to our ears, at a linear scale, the difference is negligible. What does happen though at lower frequencies is that the RMS or overal power drifts relative to peak power, meaning that a wave at such low frequency becomes so low that the RMS or rather average heating of a device will increase and decrease enough to where it won't be accurate to the theoretical average power (Wrms) dictated by the peak magnitude of the signal. This means that at lower frequencies the device is more likely to slightly heat up more than it would with higher frequencies. The bigger problem though is the extra mechanical stress that is being put on the device, however with guitar speakers and bass speakers if there is a signal that is too low for the inductance of the coil to react to and it begins to level off, the low frequency signal will be attenuated. I talked about this before in a thread that talked about reactive loads. If the signal is too low in frequency, the coil's inductance will see the signal as DC and after the time constant of the speaker has passed, the charge in the magnet will be lost and the speaker will begin to move back towards center, even though there is still a signal present. The effective output signal or the audio will be a mix between the signal from the amp, and the loss of charge of the magnet, resulting in an attenuation. The threat though of overheating is a concern hear mostly because the signal is lost however is still dissipating and if the user thinks that they don't have enough low end (the same can happen with tweeters at high frequencies too), they will try to boost it. That can cause over dissipation of both the speaker and the amp. This isn't just a problem with bass amps, a guitar into a guitar amp into a guitar cab is equally prone to this. For that reason as mentioned earlier all loudspeakers are designed to be able to handle much more than typical conditions. For that reason, a T-75s rating of 75W means that the manufacture determined that the speaker in a worst case scenario can average that power transfer with no risk whatsoever from mechanical failure or electrical failure. That average and rating is usually much lower, sometimes up to 20% lower than the point which the avalanche effect would take hold and cause destruction. This is why in actuality 5150s with four 6L6s have been known to dissipate up to 130+ Watts, it even says this in the 5150 manual, so that would mean the 6L6 is handling 32.5W RMS or 46W peak, even though they are rated for 30W. This is because the 30W maximum is a conservative maximum to ensure that they are rugged enough handle the abuse in everyday use. This goes for all electrical components.
I've not tested this, maybe I will. Another point you didn't address is that guitar bridge humbuckers don't put out a whole lot of fundamental energy, neither do bass guitars' bridge pickups, and it has to do with what portion of the vibrating string the pickup is excited by. The string length closer to the bridge has much more second harmonic (the "honk" of a bridge pickup) and subsequent harmonics and almost no fundamental at all. If you take the same guitar and switch to the neck humbucker, you will see a marked boost in the fundamental energy present in the output signal, and a harmonic structure more similar to a bass guitar.
This makes sense because bass guitars aren't usually designed to be used with the bridge pickup alone. Most bassists I know will use neck pickup or a blend of neck and bridge pickup far more often than a metal guitarist who will keep the guitar parked in the bridge pickup position for an entire set. This means that although basses do only have a one octave difference in fundamental energy, their tonal focus in those ranges is many times greater than that of a guitar since what lows they do have are much more focused and will be amplified in a linear way which only serves to accentuate this. Tonal presence always trumps raw dB measurements.
You hit is right on the head, the strength of the fundamental is what determines the power of the low end even for gutiars. One thing to not forget is that when it comes to what a speaker can handle, it all comes down to power dissipation and mechanical energy. Don't forget that it all comes down to power, you know P=I*V. Just because the bass guitar has more energy in the lower frequencies, when we look at the entire power being distributed to the load, you will notice that the bass guitar lacks harmonics and other information and thus compensated by putting that extra power in the lower frequencies. By that I mean, when you push the volume up higher, instead of your headroom being eaten up by higher information, you have more room to boost the lower spectrum. Power or energy being distributed to the speaker is still the same, 100W of 50Hz is still the same as 100W of 5KHz. The big difference is the mechanical stress on the speaker due to the lower frequency vibration which as mentioned before could cause the magnet glue to come undone and you would loose the magnet from the rest of the speaker assembly. But the amount of power required to even do that to a guitar speaker is huge.
Back to the perception of loudness being logarithmic, power is the same way. a 100W amp is only perceived as twice as loud as a 10W amp, and a 10W amp would only be perceived as twice as loud as a 1W amp. The 100W amp being 4 times as loud as a 1W amp. You want to hear the difference of 2 times as loud, open a track in your DAW, copy it, and set the second track to -6dB and toggle between the two channels, the louder track is eating up ten times more power than the more quiet track, pretty crazy right. Well it gets even crazier. You know the speaker sensitivity ratings manufactures give them, well speakers have such a poor efficiency that a better way to measure their effectiveness to do the job is to measure how much acoustic volume it makes for a certain power rating. If you go to Celestion they rate for example the V30 to have a sensitivity of 97dB. That means that it has a measured volume of 97dB (extremely loud) 1m away from the speaker with a 1W amp. That means a 10W amp it would be 100dB 1m away and 103dB with a 100W amp.
As Sloan mentioned it takes power to get lower frequencies audible, this is why you need a huge bass amp (500 Watts at least) so a bassist can compete with a mere 120 Watt amp. The Fletcher-Munson effect is a bitch in that regard, but that just means that you would have no problem if you are just recording and micing a bass guitar, there is no issue running an amp a very low power to get the volume you need to mic the cabinet. Remembering that you can get "relatively" blistering volumes with very little power considering the Fletcher-Munson effect, if you ran a bass guitar into a bass amp or even a guitar amp into a guitar cab, conversation or just slightly louder, you might be eating up 1W of clean power if you are lucky and it the signal strength would still be loud enough to get a good track.
The preamp of a guitar amp has a significant filtering effect on the low end present, so even though they use an F string, it's all getting attenuated before it reaches the power amp anyway. Also tubescreamers are a staple of that genre, and they significantly cut lows below 700Hz even before the signal reaches the amplifier in the first place. Not a fair comparison at all.
Filtering happens in the power amp too. Most of the low end coming from the power amp is actually coming from a mix of the Global Feedback circuitry, and the speaker's resonance where it creates is own voltage and signal as the cone and voice coil travel through the magnet as well as the cabinet's resonant frequency which can actually be much lower.
And as for tubescreamers, that 700-800Hz low cut is actually not a low cut, its a low pass filter, it actually removes the highs above that point. The reason they do sound brighter is again, harmonics and the illusion of low end being removed because harmonics are being created by the clipping circuit, which is then filtered out by the low pass filter which is effectively removed when the tone knob is all the way up.
You completely missed my statement about the peaks being unclipped when they enter the power amp. You present yourself in a way that makes me think you might know even more about engineering than I do, and I respect the knowledge you have but you completely fail to mention that power output of a power amplifier is directly proportional to the voltage at input. When Djenty Dave plugs his guitar into his trendy ENGL amplifier, it is being clipped and squared all to hell. This means that the output of the power amp will not have as many dramatic swings and so the wattage will be more constant to the speaker and it will be easy to judge when the speaker is being stressed or not.
In my instance the input of the power amp is DI signal, which as we all know, is VERY peaky. When the DI swings the power amplifier will swing along with it until it reaches its voltage rails. The speaker in this case will receive a "punch" from the amp, whereas with the example of a squared signal being amplified, the amp is only "leaning" on the speaker. We are dealing with a tube amplifier here, and tube amplifiers are notorious for being able to swing much higher and cleaner than their spec sheet rating given certain conditions or when the equipment is being used outside of its intended use, which it is here. You wont even have time to sit and judge whether the signal is too loud because by then the transient has already come through and punched a hole in your speaker cone. This is a HUGE gray area, and it's one that is very difficult to calculate or account for because of: the unpredictability of DI signal, the "professional inaccuracies" or aka "white lies" of amplifier speccing, and the sheer danger of the fact that a speaker which is overloaded with frequency content below its effective response range may not show signs of wear before it ultimately craps out without warning. The speaker will fail from over-excursion long before it fails from a wattage mismatch.
There is so much more that goes on in a gutiar amp than just clipping and making "square waves". In fact the output of a highly distorted guitar amp actually looks nothing like a square wave, in fact the signals to the speaker can exceed the voltage to the power amp, or depending on how you analyze it, the voltage of the power amp can increase higher than what it is supplied in certain conditions and even reconstruct resonant signals that have been clipped. The thing is while comparing the signal of the guitar to the output of the amp will look like two very different signals are equally as dynamic, if not the signal after the amp being more dynamic, sounds weird considering teh RMS readings of a gutiar amp are constant, but again that is based on perceived volume, while the signal after the cab may sound compressed and constant, it is way more dynamic than people think.
For all intents and purposes of a short conversation, lets just say that over-excursion is the same as over-disspation, and this is what I mentioned before about using your ears, if a speaker is experiencing over-excursion, it is the result of a very small amount of time which the speaker is being over-dissipated, too short for you to hear as volume (for numerous reasons that is a conversation and a half on its own) and you can hear that by a very nasty brickwall distortion. Speakers have safety features to keep them from imploding by changing the mechanical resitivness that increases the more the cone is pushed towards the end of excursion, preventing the cone from hitting the frame until maximum dissipation has been exceeded. I am actually working on a patent on the larger works of a circuit that incorporates the distortion or compression characteristics of such distortion or rather saturation is a better term. To say the least I have studied the way speakers behave and their safety features is an understament.
To eleborate on the safety feature of speakers, right after their sensitivty rating, the cone begins to resist additional movement that increases the further the cone moves away from center. This creates a marvelous and musical sounding compression but is a safety feature to prevent the speaker from suffereing massive mechanical shock resulting in faulire. Thereforethe biggest concern is over-dissiption, and like I have said, over dissipation and consequently over-excursion will make a very nasty distortion even if the signal only clipped for a very short amount of time. The moral of the story is to not plug the cab into a power amp that exceeds its power. If you had a 500W bass amp going into a 240W cabinet and had a bass player do really percussive slap bass, not only will you hear a horrd distortion, you will blow the amp. Your 5150, a 120W amp even if it has some extreme peaks, will not exceed the power handling of your cabinet. Even under a near short circuit load, becuase pentodes are current limiting governers, in a 120W setup, it is not ever possible to ever get a 5150 to put out more than 150-160 Watts, meaning, you will never have a peak that will over-excurd your 300W cabinet. Now when it comes to solid state power amps, as long as it is under 300W, no solid state power amp can exceed its maximum rating to the load, perdiod, it just makes an equally nasty sound as speaker excursion distortion. If you hear that, regardless if it is from the speaker or the amp, turn the volume down. You have 300W of power in the cab, just keep the amps you plug into the cabinet under 300W, preferably half that with tubes to 150W and it won't be possible to ruin your speakers.
I know a lot of dudes have chimed in hear and said that there have been people they knew who blew speakers, and since we don't know the setup I can't say what went wrong, but I can imagine a typical braindead musician doing stupid shit like having a high powered bass amp in a guitar cab that exceeded the rating and wondered why it blew. For the most part, most solid state bass amps are beyond the power rating of any guitar cab in the market and that would be your root cause for failure.
I appreciate your chiming in, but I do feel you are a bit condescending when you approach it with the attitude of "it's so easy dude, totally nothing to worry about?" and then you fail to take into account some aspects of the equation I am already familiar with or you fail to read the initial post to see what exactly I'm talking about doing. My math has never been strong and I'll be the first to admit to that fault, but my engineering and circuit background is there and I understand a lot of unprinted things that most people overlook because they're so busy looking at numbers and spec sheets and graphs to realize that these things can only approximate truth.
Spoken as a conversation between professionals with the deepest of respect, this is a purely academic discussion. I'm just not convinced the points you've brought forth hold the water that we both wish them to. I'm chiming in for the readers, not to continue an otherwise dead argument or start a pisswar
So far 3+ years of College, AS Degree in Electrical Engineering, another 3-4 years of additional study, including hundreds of dollars of additional reading outside college textbooks, the past year and a half creating my own designs, including at least a possibility of one pattent, hundreds if not thousand hours of study, if I don't know what I am talking about when I have spent over half a decade putting my life to this research then I fail at life. I just like to inform. The scope of this conversation is way beyond what any thread could contain, to understand even an inkling of this thread and concepts would take years of study to even grasp, very ambiguous for an internet thread.
