Any Rules For Mixing Cabinets?

[BadExample]

So a fifteen could just as easily have half the total impedance of a 410 at 80hz and die four times faster?

Not likely to murder the 15, because the minimum Z of the coil is = R (unless a parallel path, then that path takes power about as much as it drops the R). Of course there are many other factors, wattage rating of the drivers, wattage applied, and for how long. Electrical death of spool/wrapped coil like a VC comes from enough heat generated for long enough to raise the temperature high enough to melt the coils varnish insulation. That shorts wire(s) together, dropping R and Z, worsening the problem causing more melting and shorting, until their is enough heat to burn the copper or more likely, what's driving it gives up. I would be surprised if anything would be damaged playing music through mixed cabs as long as the math is done and the nominal Z is = or > min Z of the amp. That is assuming all the knobs are not turned full CW

 

[ThisBass]

So a fifteen could just as easily have half the total impedance of a 410 at 80hz and die four times faster?

No, not this way. The measurements I did was ported versus sealed which is a combination that should be avoided anyway. Never the less the resonant peak which is bold with ported cab has a more remarkable impact.

I'd tell it this way, any assumption of a "nominal" Impedance is nearly true (+/- some deviation) only for rather typical program material, or Pink Nois, and "flat" response reinforced by an amplifier.

I woukdn't expect "identical" nominal impedance with different cabs anyway. A strong bass boost (which is actually selective gain) shifts power density to the lows where this remarkable impedance peak generates an increase to the nominal "characteristc". If both mixed cabs are shifted to higher impedance the amplifier can't push it's full power anymore. So there is no concern to consider a weakest link.

But indeed any assumption of an exactly even 50/50 power distribution with different cabs is dubious, at least total system impedance numbers like 5.33 Ohm or 2.67 are way to optimistc.

It's like a paradoxxon, althogh power to the cabs is decreased due to increasing impedance the potential of system max SPL is somewhat increased cause rail voltage sag of the power amp stage is somewhat decreased with highish going load.

Nominal impedance isn't always just the same, everybody who knows a couple of things about the legendary "Killer-Kappa" HighEnd cabs knows how diabolic a nominal impedance can be.

 

[Downunderwonder]

Not likely to murder the 15, because the minimum Z of the coil is = R (unless a parallel path, then that path takes power about as much as it drops the R). Of course there are many other factors, wattage rating of the drivers, wattage applied, and for how long. Electrical death of spool/wrapped coil like a VC comes from enough heat generated for long enough to raise the temperature high enough to melt the coils varnish insulation. That shorts wire(s) together, dropping R and Z, worsening the problem causing more melting and shorting, until their is enough heat to burn the copper or more likely, what's driving it gives up. I would be surprised if anything would be damaged playing music through mixed cabs as long as the math is done and the nominal Z is = or > min Z of the amp. That is assuming all the knobs are not turned full CW

I am referring to a situation of abundant amp power and equal nominal cab impedance with mechanical over excursion failure mode.

R being around 2/3 of nominated Z until things heat up.

 

[DaveAceofBass]

The math I am familiar with says that only the impedance curves, and whether the cabs are connected in series or parallel (and they're virtually always in parallel), matter, as far as determining power distribution. So there can be minor differences in power distribution between two "8 ohm" cabs, but they are explained by the impedance curves.

I do not see how either the speaker configuration or the sensitivity of the cabs would make any difference in how the power is distributed.

Can you explain it to me?

Suppose you have a head that puts out 250w at 8 ohms and 500w at 4 ohms. You hook up two 8 ohm cabs...Brand A has a 210 w/ tweeter with a sensitivity of 95 and Brand B is a 112 w/ 6 inch mids and tweeters with a sensitivity of 105. Hypothetically the 210 should be louder, but the 112 is clearly louder.

Now suppose that head can go as low as 2 ohms. The same cabs are used, except this time you use a 4 ohm version of the 210 and the 8 ohm version of the 112. You would expect the 4 ohm cab to be louder due to more speaker surface and lower impedance. However, in this case the 112 is still significantly louder. This is one of the exact scenarios I've experienced, but not the only test.

Usually the greater the speaker size of the cab, the louder it is. So a 410 would be louder than a 115, as there is more speaker surface. But I don't think that's what is in question...it's how much power each can receives, right?

Two well known amp companies have explained this to me directly. By hooking up two 8 ohm cabs, the amp "sees" 4 ohms, but that doesn't mean a 500 watt amp sends 250w to each cab. Imagine you have a faucet and you connect a Y-shaped valve to it, where one side of the Y is much larger in diameter and the other much smaller. When you turn on the water, the majority of the water flows to the larger diameter opening, but some water is still diverted to the smaller opening. This is how the amps work, in my experience, and from what I have learned from experts.

Now let's suppose I am wrong and each cab does get 250w in my hypothetical scenario. Even if that is the case, the speaker with the higher sensitivity will be louder. I can tell you this from first hand experience. Now even if both speakers received the same power, the main conclusion that we all need to know is that the speaker with higher sensitivity will be louder.

 

[Passinwind]

Two well known amp companies have explained this to me directly. By hooking up two 8 ohm cabs, the amp "sees" 4 ohms, but that doesn't mean a 500 watt amp sends 250w to each cab.

Ohm's Law was repealed? Cool!

But in all seriousness, I'd guess that something got a little lost in the translation from what those speaker makers told you. And again, relative loudness was not what Duke was addressing.

 

[Passinwind]

Indeed there can be noticeable difference which depends on frequency spectrum.
it's not unusual that the average Impdance increases with boost of lows (EQ adjustment).

The measurement to proof the Impedance versus spectrum dependency is rather simple.
Take a pink noise and do a measurement for Urms and Irms and then calculate Z=Urms/Irms.

Then boost the low end by +12 dB and measure Urms and Irms and calculate Z.
There should be a very remarkable increase for Z

That's a new one on me, but I look forward to testing it once I get my impedance curve tracer back up and running. I assume you also decrease overall level by 12dB to compensate for the bass boost and keep all levels well below onset of power compression?

 

[ThisBass]

That's a new one on me, but I look forward to testing it once I get my impedance curve tracer back up and running. I assume you also decrease overall level by 12dB to compensate for the bass boost and keep all levels well below onset of power compression?

yep, power compression wasn't of any conern in my consideration cause (I think) it makes no sense to tread a sytem beyound it's "practical" performance charateristica.

The measuremts I did was with True RMS meter bandwith 20Khz for Voltage and 5 kHz for Current. Some nitpicker may criticize the current AC bandwith but. it's of no "practical" concern. The remaining crest error in measuremnts cancels out broadly by Urms/Irms.

BTW thermal power handling of drivers is commonly predicted with theoretical nominal Z. There is actually no current RMS measurement to calculate the "actual" power consumption.
May be that some manufacturers do a P = Urms * Irms determination but, a very lot of manufacturers refere to AES or different approved societies.

To the contrary, the older AES reference recommended to use Zmin instead of Znom to determine power handling. Differencies can't be more obvious like that.
The benefit of a P=Urms^2/Zmin prediction wasn't intended to promote marketing department goals, it was meant to "easily" predict current ability of amplifiers at max SPL.

 

[Downunderwonder]

Suppose you have a head that puts out 250w at 8 ohms and 500w at 4 ohms. You hook up two 8 ohm cabs...Brand A has a 210 w/ tweeter with a sensitivity of 95 and Brand B is a 112 w/ 6 inch mids and tweeters with a sensitivity of 105. Hypothetically the 210 should be louder, but the 112 is clearly louder.

Now suppose that head can go as low as 2 ohms. The same cabs are used, except this time you use a 4 ohm version of the 210 and the 8 ohm version of the 112. You would expect the 4 ohm cab to be louder due to more speaker surface and lower impedance. However, in this case the 112 is still significantly louder. This is one of the exact scenarios I've experienced, but not the only test.

Usually the greater the speaker size of the cab, the louder it is. So a 410 would be louder than a 115, as there is more speaker surface. But I don't think that's what is in question...it's how much power each can receives, right?

Two well known amp companies have explained this to me directly. By hooking up two 8 ohm cabs, the amp "sees" 4 ohms, but that doesn't mean a 500 watt amp sends 250w to each cab. Imagine you have a faucet and you connect a Y-shaped valve to it, where one side of the Y is much larger in diameter and the other much smaller. When you turn on the water, the majority of the water flows to the larger diameter opening, but some water is still diverted to the smaller opening. This is how the amps work, in my experience, and from what I have learned from experts.

Now let's suppose I am wrong and each cab does get 250w in my hypothetical scenario. Even if that is the case, the speaker with the higher sensitivity will be louder. I can tell you this from first hand experience. Now even if both speakers received the same power, the main conclusion that we all need to know is that the speaker with higher sensitivity will be louder.

typo much? A 95dB cab is 10dB quieter than 105dB cab. You can't make that up with even double the power.

 

[BadExample]

I am referring to a situation of abundant amp power and equal nominal cab impedance with mechanical over excursion failure mode.

R being around 2/3 of nominated Z until things heat up.

Perhaps the hypothetical person who would crank the big amp enough to mechanically destroy the 15 with a 4 x 4 added would have already destroyed the 15 before he bought the 4 x 4 to add

 

[ThisBass]

Two well known amp companies have explained this to me directly. By hooking up two 8 ohm cabs, the amp "sees" 4 ohms, but that doesn't mean a 500 watt amp sends 250w to each cab. Imagine you have a faucet and you connect a Y-shaped valve to it, where one side of the Y is much larger in diameter and the other much smaller. When you turn on the water, the majority of the water flows to the larger diameter opening, but some water is still diverted to the smaller opening. This is how the amps work, in my experience, and from what I have learned from experts.

A different but rather valid consideration,
the amp sends a distinct voltage to the cabs and the cabs try to suck appropriate current from the amp. If frequency spectrum emphasize the range around Zmin then current draw is rather big, if the spectrum is EQ'd to emphasize bandwith around the impdance peak then the current draw is deminished, thus the rather dynamic charicteristic of "average" Impedance an amplifier can see within different conditions.

That's why I dont't like phrases like "the cab is powered with "X" watts ".
Of course drivers don't know how much of current is noxious, neither they don't know how much of voltage to force excursion beyond Xmax, or reach Xlim in worst case scenario.
The power needed to exceed Xmax is widely unknown cause, again, impedance curve isn't a steady number. as well as frequency spectrum isn't a steady thing like The ten commandments.
Even power distribution with multiple cabs, ... oh my.

Note, 15" loaded cabs tend to be tuned lower than their 10" loaded counterparts thus they can take some benefit regarding machanical "power" handling, which is actually robustness versus highish voltage at sub lows.

 

[DukeLeJeune]

Thank you DaveAceofBass for taking the time to reply.

BadExample pointed out that you and I might simply be having a semantic disconnect - when I say "power distribution", I'm talking about how the amplifier's output (wattage) is distributed between the two cabs. If you are talking about something else, like how loud each cab will be, then we're talking about different things.

That being said, here are my replies to your post:

Suppose you have a head that puts out 250w at 8 ohms and 500w at 4 ohms. You hook up two 8 ohm cabs...Brand A has a 210 w/ tweeter with a sensitivity of 95 and Brand B is a 112 w/ 6 inch mids and tweeters with a sensitivity of 105. Hypothetically the 210 should be louder, but the 112 is clearly louder.

Agreed.

Now suppose that head can go as low as 2 ohms. The same cabs are used, except this time you use a 4 ohm version of the 210 and the 8 ohm version of the 112. You would expect the 4 ohm cab to be louder due to more speaker surface and lower impedance. However, in this case the 112 is still significantly louder. This is one of the exact scenarios I've experienced, but not the only test.

Ignoring the idiosyncracies of the individual impedance curves, in this example the theoretical 500 watts coming from the amplifier would be distributed as follows: 333 watts into the 210, and 167 watts into the 112.

In this scenario, the 95 dB 210 cab would theoretically be hitting about 120 dB, and the 105 dB 112 cab would be hitting about 127 dB, ignoring thermal and mechanical compression. Calculating the combined SPL of the two is more math than I'm willing to do right now.

Usually the greater the speaker size of the cab, the louder it is. So a 410 would be louder than a 115, as there is more speaker surface. But I don't think that's what is in question...it's how much power each can receives, right?

That's what I've been talking about, yes, how much power each cab receives.

Two well known amp companies have explained this to me directly. By hooking up two 8 ohm cabs, the amp "sees" 4 ohms, but that doesn't mean a 500 watt amp sends 250w to each cab. Imagine you have a faucet and you connect a Y-shaped valve to it, where one side of the Y is much larger in diameter and the other much smaller. When you turn on the water, the majority of the water flows to the larger diameter opening, but some water is still diverted to the smaller opening. This is how the amps work, in my experience, and from what I have learned from experts.

In this analogy, if each is an 8-ohm cab, then (ignoring the idiosyncracies of their individual impedance curves) both sides of the Y are the same diameter. However if you are combining a 4-ohm cab with an 8-ohm cab, the 4-ohm cab's side of the Y is 1.414 times greater in diameter (which works out to double the cross-sectional area).

So your hose analogy is very close, but if the impedances are the same, then the hose diameters would be the same (again, ignoring the idiosyncracies of the individual impedance curves).

Now let's suppose I am wrong and each cab does get 250w in my hypothetical scenario. Even if that is the case, the speaker with the higher sensitivity will be louder. I can tell you this from first hand experience. Now even if both speakers received the same power, the main conclusion that we all need to know is that the speaker with higher sensitivity will be louder.

Yes, how loud each cab will get is a function of its efficiency and how much wattage it sees.

But in my opinion, there is another conclusion that might be more important than which cab is louder: Will one of the cabs be in danger of being damaged??

Suppose the cabs in your second example above, the 4-ohm 210 and 8-ohm 112, are both rated at 300 watts. If you're driving both cabs in parallel with your 500 watt amp, knowing that the 210 would theoretically be seeing 333 watts at full power might be valuable information that would keep you from pushing your rig too hard.

(If we wanted to be thorough we would look at the mechanical power handling as well, but we'd probably have to rely on making educated guesses.)

Keeping cabs from being inadvertently overpowered is why I have focused on power distribution to the cabs, rather than on how loud each cab gets.

 

[ThisBass]

A different consideration on bulky impedance peak with ported cabs.
If TB was more consistent then the form would dislike any combination of tube heads with ported cabs. Unless the gear is evaluated the result is a crapshoot cause the impedance peak interacts strongly with system performance respectively tube amplifier design. Response is unpredictable, system damping is unpredictable, tonal feel is unpredictable, it's a very good bet to run into some troubles on stage, and nasty/muddy sound in the house.

A (mind thinking) solution was to explicitely suggest sealed cabs with tube heads.

 

[MVE]

A different consideration on bulky impedance peak with ported cabs.
If TB was more consistent then the form would dislike any combination of tube heads with ported cabs. Unless the gear is evaluated the result is a crapshoot cause the impedance peak interacts strongly with system performance respectively tube amplifier design. Response is unpredictable, system damping is unpredictable, tonal feel is unpredictable, it's a very good bet to run into some troubles on stage, and nasty/muddy sound in the house.

A (mind thinking) solution was to explicitely suggest sealed cabs with tube heads.

Or, only get a combination of tube amp and ported cab that is known to work well together... and the obvious place to query and fish for experiences and opinions would be right here on TB... Oh, it's a viscious circle. Lol.

 

[DukeLeJeune]

I agree that if there is any difference it will be slight but, as you know, the impedance of a speaker cabinet varies with the frequency of the signal. Is it possible that factors such as reactance in a particular cab and its drivers might result in a slightly different impedance from that of a different cab during operation?

Yes, the impedance does vary with frequency, which is why I said that the power distribution is explained (maybe should have said "predicted"?) by the "impedance curve", rather than by the "impedance", as the latter can imply a single number on a spec sheet.

There can be quite a bit of variation in impedance curves between cabs that have the same nominal impedance rating, but the minimums are what we need be the most concerned with, and those generally don't vary as much.

 

[ThisBass]

In this scenario, the 95 dB 210 cab would theoretically be hitting about 120 dB, and the 105 dB 112 cab would be hitting about 127 dB, ignoring thermal and mechanical compression. Calculating the combined SPL of the two is more math than I'm willing to do right now.

Total level adding of coherent signals
Invalid Link Removed

120dB + 127dB = 130.2 dB

 

[DukeLeJeune]

Total level adding of coherent signals
Invalid Link Removed

120dB + 127dB = 130.2 dB

I didn't know there was an online calculator for that - THANK YOU VERY MUCH! I bookmarked the link.

 

[Downunderwonder]

Thank you DaveAceofBass for taking the time to reply.


That being said, here are my replies to your post:



Agreed.

• In this scenario, the 95 dB 210 cab would theoretically be hitting about 120 dB, and the 105 dB 112 cab would be hitting about 127 dB, ignoring thermal and mechanical compression. Calculating the combined SPL of the two is more math than I'm willing to do right now.

not seeing where there was agreement with the first point.

 

[TTPbass1]

Just being realistic... My Markbass CMD 102P Combo (2x10) is rated 101 dB SPL (Front ported) and the Standard 151HR extension cabinet is 100 dB SPL (Rear ported). The Combo Head II (Little Mark III) is rated 500 watts at 4 ohms.

This combination sounds amazing, in the player's spot and on the auditorium. The combo alone sounds great, but when you add the 151 cabinet the sound grows in definition and roundness. They're a great mach together.

 

[Passinwind]

I didn't know there was an online calculator for that - THANK YOU VERY MUCH! I bookmarked the link.

That site has a lot of other useful calculators as well, plus they show you their homework.

 

[BadExample]

I've been using Electronics Calculators Shavano Music Online - Speaker Wiring; Impedance Calculator and JAVASCRIPT CALCULATORS some, not sure if they are in the FAQ/stickies.