The Impedence Problem

[Mikey]

A recent discussion on the forum was about PROBLEM versus ISSUE, and though it was humorous, it had some truth to it. A problem is something that is realized and for which a solution is being sought, but an issue is something which is realized but no real solution is being sought. One of the most difficult long-standing ISSUES on Bill's forum is ... How many speaker cabinets should I hook-up to my amp? A simple question. Unfortunately, every person has their own perspective and reasons for providing an answer. Even though there are a lot of intelligent people here who are well educated in audio, electronics, physics, math, etc, nobody can seem to agree on this ISSUE. I think that there IS something which we can all agree-on, though ... it's high time that we collectively treat this as a PROBLEM and seek to solve it.

So, how do we go-about solving the problem, "How many speaker cabinets should I hook-up to my amp?"

1) We keep this topic confined to this thread, so that it will no longer plague Bill's entire forum, and so that all information about this subject will be readily available here, without being culled.

2) We all approach this problem with the goal of finding a SINGLE solution which is as SIMPLE as possible. This is, no doubt, a phenomenal goal, but it's not unattainable. I ask that everyone who wishes to participate in this discussion will do-so in accordance with reaching this goal.

3) As I've told many of my students, "You can't learn anything if you already know everything". Even though you may already have your own personal solution to this problem, that doesn't mean that it is necessarily the best solution for everyone, and you may discover that it's not necessarily YOUR best solution. Keep an open mind, have respect for others, and focus on the goal (NOT just proving yourself to be right).

4) "Mikey, who in the hell do you think you are, starting this thread?" ... It's pretty much common knowledge around here that I'm not one of the more highly knowledgeable members when it comes to the sciences envolved in speaker design and related topics. I'm just middle-of-the-road. Hell, that may even be a stretch, but at least I admit it, and I'm always trying to learn. That puts me among The Unwashed Masses who are the target market for Bill's quest. I'm the Joe Schmoe who is here seeking answers, not the opinionated bullshit that I've experienced on every other forum. I'm the typical guy who would like to see this problem solved.

5) In perspective, this forum is a think tank comprised of a huge number of very intelligent, highly-educated people, all here with a common interest. If there ever was a group with the potential to solve this problem, this is it. I think that this thread could have the potential to be a definitive resource to the audio community, if we all pull together for the common goal.

And last but not least ... in order to keep this thread moving forward and on-track, at the end of EVERY POST, I ask that you copy and paste this:

PLEASE DO NOT POST TO THIS THREAD UNTIL YOU HAVE THOROUGHLY READ ALL OF THE PRECEDING POSTS IN THIS THREAD!

I now open the floor for discussion...

[Mark Coward]

A recent discussion on the forum was about PROBLEM versus ISSUE, and though it was humorous, it had some truth to it. A problem is something that is realized and for which a solution is being sought, but an issue is something which is realized but no real solution is being sought. One of the most difficult long-standing ISSUES on Bill's forum is ... How many speaker cabinets should I hook-up to my amp? A simple question. Unfortunately, every person has their own perspective and reasons for providing an answer. Even though there are a lot of intelligent people here who are well educated in audio, electronics, physics, math, etc, nobody can seem to agree on this ISSUE. I think that there IS something which we can all agree-on, though ... it's high time that we collectively treat this as a PROBLEM and seek to solve it.

Bruce is the only one who has consistently posted his belief that a speaker's measured minimum impedance can be reached in actual use and therefore you can blow up an amp if you depend on nominal impedance ratings. It appears that you and Davy are in this camp as well. Personally I think that Bill has always posted reasonable and conservative guidelines. I've used tubas and titans for some four years now, and have never blown up an amp. I know of no one else who has blown up an amp. I've read for years on other audio and pro sound forums, and have never heard of this being an issue. The only argument that I've seen concerning minimum impedance was the discussion about Wayne Parham's Prosound shootout. Basically he insisted on measuring minimum impedance, then calculating the input voltage relative to this figure. Tom Danley and his camp, and pretty much everyone else told him to use nominal impedance as it is the industry standard:

http://srforums.prosoundweb.com/index.p ... /0/0/5451/

IMO, passing our Txxx boxes off as 10 ohm nominal is doing a huge disservice to our users.

If our users want to pretend their subs are 10 ohms instead of 5.8 ohms at 40 and 80 Hz, they are welcome to it. They can run four in parallel and pretend the 1.45 ohm load is higher than it is. They are welcome to red-line their amps with ultra low impedance loads. But I don't want to hear any crying about failures. I have no sympathy for users who either ignorantly or stupidly abuse their amps and speakers, and then wonder why they blow up.

What really bothers me about this subject is users getting the idea that BFD cabs are "different" than commercial cabs. We are no more stupid and ignorant than JBL, QSC, just about anyone you care to name. Folks buy a JBL box, it says 4 ohm nominal impedance. They bridge an amp into it. JBL doesn't publish impedence sweeps, nor do users typically do this.

Bruce referred to the QSC guy to "prove" his point of view, while you can see from QSC's published specs and recommendations for their amps and cabinets it appears that they are just as ignorant and stupid as the rest: 3,600w into one cab, 4 ohm nominal impedance. Now, being a logical person, it appears that QSC doesn't make an amp that will produce 3,600w/4ohm in one channel. Therefore, the only way to supply that much power is to bridge an amp. 4 ohm is the minimum bridged load for their capable amps. Unless they've done something magical with the drivers and cab, it is going to have a minimum impedance(s) somewhere below it's nominal 4 ohm rating. So it appears to me that they are recommending doing exactly what Bruce is telling us not to do, with their own amps and speakers. I guess they want people to blow up their stuff?

The issue is not that cabs and speakers have minimum impedance below nominal - which btw nominal is an average, it would be impossible for Danley's box to have a minimum above nominal - I'm sure Bill is correct and he just rounded down to 4 ohm to be conservative. I'd have no problem with this approach, especially since all of Bill's new designs are only adding about 2 ohm from horn loading, let's just use the nominal ratings of the drivers and consider the horn loading as extra "insurance" against overpowering. That's what I would do if selling BFD cabs commercially, forget about explaining horn impedance and why the "odd" number. One 8 ohm driver, cab is 8 ohm, and for all practical purposes it will work just fine and folks would be a little less likely to blow them up.

The real issue is supposing that minimum impedance can be reached in real world use. Theoretically it could happen, sure I could compose a piece of "music" that features a pure 40hz sine wave and nothing else but I imagine the audience would leave long before something blows up. I have seen no proof from Bruce or anyone else that this is a problem in the real world. Just look at the single fact of 2x18 cabs being the most popular sub, and "needing" 3k or more @4ohm means that a majority of users are bridging an amp per cab. If minimum impedance were adversely affecting amps, we'd see them blowing up regularly. To my knowledge, no cabinet design can reduce impedance - the lowest impedance cannot be lower than that of the driver alone.

BTW, Danley publishes both minimum and nominal impedance values for his gear:
The TH-115 nominal impedance is 4 ohms.
Minimum impedance is 5 ohms at 78 Hz.

Link to TH-115 Page

Danley's specs leave no absolutely doubt about the minimum impedance of this box and where it occurs. The published impedance chart shows a whole lot of signal at the 5-ohm level. I find this interesting, as the nominal value is less than the measured impedance over the entire passband. This differs completely from Tubas and Titans, both of which have substantial impedance dips below nominal, almost down to the DC resistance value. IMO, passing our Txxx boxes off as 10 ohm nominal is doing a huge disservice to our users.

Assuming that minimum impedance could actually be higher than nominal tends to indicate a bit of ignorance. Apparently it's easier to believe that Tom Danley can work miracles. Nominal simply means average. Could you average making $4/hour if your minimum rate is $5/hour? I think that Bill's 10 ohm nominal figure for the Titans is conservative enough, and not in any way a "disservice" to users. The 3015LF according to Eminence has a zmin of 5.31 ohm. At Wayne's shootout, Leland's 3015LF was tested to have a zmin of 6.25 ohm. Either Wayne's gear is suspect, or the horn loading adds impedance as Bill has said. Bill said that the original tuba design added 4 ohm impedance. I've run four tubas loaded with 4 ohm drivers, in parallel on a single amp channel, for 5 hours or more at a time with no problem. I've run two tubas with 4 ohm drivers in parallel, on a bridged amp, for 5 hours or more. Bill said the cabs would have a nominal 8 ohm impedance, and in my experience in actual use it is true. He has never condoned running amps with 2 ohm loads, that's my decision but it has nothing to do with the minimum impedance issue. There are plenty of folks who adhere to the philosophy of only running 4 ohm stereo loads, that is a conservative viewpoint and I understand the merit of it. But the only people I've ever seen concerned about minimum impedance are Bruce, Davy, Mikey and Wayne Parham.

[Tim A]

I'll chime in before the thread gets whacked.

If you look at this strictly from a pragmatic standpoint:

FACT: There are hundreds of thousands of amp/speaker configurations in use all over the world on a daily basis, who's users are ignorant to anything beyond the impedances listed on the speaker and amp. They work just fine. Simply becoming aware of the difference between an actual impedance load at a given frequency and nominal impedance does not change that fact.

While I certainly understand the issue, and to my mind it is an issue, not a problem that requires solving, it simply cannot be that dire or we'd all be blowing shit up on a regular basis. At that point, it would be a problem.

[Sydney]

a SINGLE solution which is as SIMPLE as possible.

(1) Measure the actual load ( including ALL cabling and connections ) in the intended frequency range of op.
(2) Find out as much as possible about the amplifier driving it ( even if it is only field reports).

That no Professional amp's blow in use, is a testament to the design measures that are implemented. This is done because of the rough duty that is to be expected. Lethal load conditions are usually protected by shutdown or other over current protection measures. Including such things as thermal switches on the heatsinks. Power supplies are more robust and fans are included for dissipating heat. Pro Amps have more extensive heatsinking.
The same abuse leveled at many home stereo amp would destroy the output section promptly. ( I have seen it )
Again we are talking about longterm reliability.
To use an analogy: A combustion engine can be set up and driven very conservatively and will last hundreds of thousands of miles. The same engine driven near red line constantly will not last as long. The same engine set to produce max power at the edge of performance envelope ( ie an engine in a dragster producing an enormous amount of power briefly ) has to be rebuilt after every race, because of parts fatigue.
It's not the impedance, it's the increased amperage that causes heat that shortens component life.

[Mikey]

Mark, thanks for getting the ball rolling!

Bruce is the only one who has consistently posted his belief that a speaker's measured minimum impedance can be reached in actual use and therefore you can blow up an amp if you depend on nominal impedance ratings. It appears that you and Davy are in this camp as well.

I think you're way off the mark, here. I doubt that there are just the few of us who feel it's foolish to run an amp at it's minimum-rated impedence based on calculations of total nominal driver impedence. Bear in mind that this isn't simply an issue of blowing an amp up suddenly. As was recently shown, the lower the impedence of the total driver load, the less life duration you can expect from your amps. Personally, I feel that it's irresponsible to tell people to run at the minimum rated impedence of an amp when it's known that this will greatly shorten amp life.

let's just use the nominal ratings of the drivers and consider the horn loading as extra "insurance" against overpowering. That's what I would do if selling BFD cabs commercially, forget about explaining horn impedance and why the "odd" number. One 8 ohm driver, cab is 8 ohm, and for all practical purposes it will work just fine and folks would be a little less likely to blow them up.

That doesn't work, either. The more cabinets you add, the less difference there is between the drivers' nominal impedence and the added impedence of the horns. For instance, four 8 ohm drivers in parallel equal 2 ohms, and four Titans with 8 ohm drivers equal 2.5 ohms. Your "insurance" is shot.

The real issue is supposing that minimum impedance can be reached in real world use.

It does. The impedence dips are at various frequencies. If a song you're playing happens to be in the key at or about that frequency, you're smacking that minimum impedence. Even a 5th or 7th harmonic could push that minimum impedence frequency. BUT, there are other factors which affect whether these dips would have enough adverse effect to shut an amp down.

This whole solution-finding effort isn't about pointing fingers or naming names, it's about arriving at an optimal way of configuring speakers to amps so that there is minimal chance of blowing an amp and ensuring reasonable amp life.

[mullett]

I'm just middle-of-the-road. Hell, that may even be a stretch, but at least I admit it, and I'm always trying to learn. That puts me among The Unwashed Masses who are the target market for Bill's quest. I'm the Joe Schmoe who is here seeking answers, not the opinionated STUFF that I've experienced on every other forum. I'm the typical guy who would like to see this problem solved....

Hey Guys one of those that have been around for awhile dont post much but have read alot. I think Tim hits it on this one.

FACT: There are hundreds of thousands of amp/speaker configurations in use all over the world on a daily basis, who's users are ignorant to anything beyond the impedances listed on the speaker and amp. They work just fine. Simply becoming aware of the difference between an actual impedance load at a given frequency and nominal impedance does not change that fact.

I have had to learn alot in the past year or so. Most of the stuff you guys taught me was far beyond what I understood before starting to read here. Even reading other places has taught me plenty as well.
Fact is, most, as I would refer to as consumers, dont know half as much as they think, or understand half of what they think they know.
There has to be a good general rule of thumb and I think that already exist, for the very knowlegable people here to try and change the rule of thumb and re-educate the masses is a very difficult thing to do. Just keep doing what you do, But dont get too hung up on trying to educate those that already know everything(not refering to any one in particular). Beside if they already know everything then why are they even reading this......


PLEASE DO NOT POST TO THIS THREAD UNTIL YOU HAVE THOROUGHLY READ ALL OF THE PRECEDING POSTS IN THIS THREAD!


Just my two cents....I have also seen Bill hack the threads that get way outthere and agree with most of what he says
Jeff Flowers

[Mikey]

FACT: There are hundreds of thousands of amp/speaker configurations in use all over the world on a daily basis, who's users are ignorant to anything beyond the impedances listed on the speaker and amp. They work just fine. Simply becoming aware of the difference between an actual impedance load at a given frequency and nominal impedance does not change that fact.

While I certainly understand the issue, and to my mind it is an issue, not a problem that requires solving, it simply cannot be that dire or we'd all be blowing shit up on a regular basis. At that point, it would be a problem.

Part of the reason we're all here is because we want more for ourselves than is available to the average musician, in terms of both gear and knowledge. Didn't your folks ever use the classic phrase, "If your friends go and jump off a bridge, does that mean you have to follow"? Sure, granted, there are countless amps out there running 2 ohm loads, but that doesn't mean that's the BEST way to do it. Right?

[Mikey]

for the very knowlegable people here to try and change the rule of thumb and re-educate the masses is a very difficult thing to do

The fact IS that amps DO blow, for many reasons. How many blow due to low impedence abuse? Who knows. How many wear-out prematurely due to constant thermal abuse? Who knows. The point is that amp life and reliability can be greatly improved by ensuring that you operate your amps with a reasonable margin of safety. Knowing that, why would a responsible person recommend that someone else throw caution to the wind and go ahead and operate their amps at minumum impedence? It's not a matter of educating the world, it's a matter of ensuring that good advice is given here on the forum. Then whatever people choose to do with that advice is up to them.

Can anyone supply any facts supporting the stance that running an amp at it's lowest recommended impedence has no adverse affects?

[davygrvy]

I have lots of thoughts on this topic, but I'll limit myself to just this for the moment (cause I'm at work).

The IEC document 60268-5 defines "rated impedance" for loudspeakers in section 16.1 as

The value of a pure resistance which is to be substituted for the loudspeaker when defining
the available electric power of the source shall be specified by the manufacturer.

That's very clear. We want a substitution number for ease in calculating loads.

The lowest value of the modulus of the impedance in the rated frequency range shall be not
less than 80 % of the rated impedance. If the impedance at any frequency outside this range
(including d.c.) is less than this value, this shall be stated in the specifications.

That part means an 8 Ohm loudspeaker shall never dip below 6.4 Ohms at any frequency including DC else it be stated as such.

If our users want to pretend their subs are 10 ohms instead of 5.8 ohms at 40 and 80 Hz, they are welcome to it.

7.25 Ohms would be the rated value for that 5.8 minimum if one followed the IEC recommendations.

[Mark Coward]

Knowing that, why would a responsible person recommend that someone else throw caution to the wind and go ahead and operate their amps at minumum impedence? It's not a matter of educating the world, it's a matter of ensuring that good advice is given here on the forum. Then whatever people choose to do with that advice is up to them.

Can anyone supply any facts supporting the stance that running an amp at it's lowest recommended impedence has no adverse affects?

You are confusing the issue Mikey. I have not, and have never recommended running an amp at it's lowest recommended impedance. The issue is that regardless of what impedance load you want to run on amp, the industry standard is calculating these loads based on a speaker's nominal impedance NOT the minimum impedance. Bill doesn't do it, Tom Danley doesn't do it, QSC, JBL, ad infinitum. Virtually every 8 ohm "nominal" driver/cab is going to have a minimum impedance of around 5 - 6 ohm. But in practical use the minimum impedance is not a factor, there will be a range of frequencies in it's bandwidth and the actual impedance will be an average.

The real issue is supposing that minimum impedance can be reached in real world use.

It does. The impedence dips are at various frequencies. If a song you're playing happens to be in the key at or about that frequency, you're smacking that minimum impedence. Even a 5th or 7th harmonic could push that minimum impedence frequency. BUT, there are other factors which affect whether these dips would have enough adverse effect to shut an amp down.

And where is your proof for these statements? What testing have you done to confirm it? Do I just take your word for it? I don't see this supposed "problem" occurring in real life.

I don't like to argue for the sake of argument. This bothers me because the implication is that Bill's recommendations are flawed, that he is being irresponsible in posting the nominal impedance ratings. I have seen no theory to prove this, and no anecdotal evidence. The industry standard is based on nominal impedance, it's worked for a long time. While "We've always done it that way" and "if it ain't broke why fix it" are cliches, I think they are appropriate to this discussion. In the four years that I've been here, I have not heard of one amp shutting down. On the other hand, there have been quite a number of blown drivers. Personally I have changed my opinions on max power since learning from Bill, previously I was in the 2xRMS camp. I may choose to run amps at 2 ohm loads, or bridged 4 ohm, I don't advise anyone else to do so. But it is an extremely prevalent practice, for the reason I stated above - the inefficient direct radiator cabs "needing" gobs of power to produce maximum output. When you "need" 3k or more wattage for a single 4 ohm cab, you pretty much have to use a bridged amp to get there if you're running on 110V power. Bridged into 4 ohm is the minimum load rating for most pro amps. All nominal 4 ohm cabs are going to have a minimum impedance of less than 4 ohm, probably about 2.5 ohm - 3 ohm. Where are all these amps that are going up in smoke?

[davygrvy]

But in practical use the minimum impedance is not a factor, there will be a range of frequencies in it's bandwidth and the actual impedance will be an average.

Is it? What's the math on that? Is that banded pink noise in the intended frequency range monitoring RMS current and coming to a Znom=Vrms/Irms result? Hmm.. I should try that method to see where it lands me

EDIT: I could probably do that in a pure analogue domain with a Analog Devices AD633 (see figure #7) and an RMS detector

[Sydney]

Thanks for that IEC ref. David.

When analogue synthesizers first became available ( Mini Moog early 70's ), a lot of people started paying attention to the work of Joseph Fourier, the mathematician who discovered the basis of what we now call Fourier analysis.
All waveforms can be broken down into sine waves of various frequencies or conversely recreated given enough sine waves.
See Additive Synthesis:
http://www.soundonsound.com/sos/jun00/a ... nthsec.htm
This enabled electronic musician to create "replicas" of acoustic instruments provided they had enough oscillators for the fundamental and harmonics.
As many discovered a flute bears a very similar sonic footprint to a sine wave, and one of the easiest to synthesize.
With the above in mind, it is not whether a single sine wave is involved but the how strong is power distribution of the fundamental.
Granted the simple pure oscillator instrumentals of the early synthesizer players were predominate in the 70's and few musicians go for that simple retro sound anymore, or play very loud flute solos ( sorry Ian Anderson ).
However a maker of a ( hideously expensive ) amp warns "that In the case that measurement are done with the non significant method of continuous sine wave input signals (applied for more than 5 seconds), the system is self-protected from that unusefull (sp) signals, and lower output values will be found".
So avoid LF feedback, and If you absolutely need to sustain a note longer than 5 secs , you may want to use another amp.

[Tim A]

Sure, granted, there are countless amps out there running 2 ohm loads, but that doesn't mean that's the BEST way to do it. Right?

No, but it certainly isn't the pending doom it's being made out to be by virtue of this thread. You say you want to give people the best advice, then cause a riot.

Almost everyone here has been doing something with sound for a long, long time. Some straight forward 'turn it up and play' and some very knowledgable tech types. By show of hands, how many people here have ever fried components or cabs and traced it back to this? On the other hand, how many people blissfully went along with the manufacturer's recommendations for years and had no problems? I'm going to guess the outcome here will be heavily slanted.

Do I regularly run my amps at 2ohms? No. Will I if need be? Yes, in a heartbeat. And if I knock a month off the amp's life, so be it.

And quite frankly, to suggest that you can start a thread and have everyone come to consensus on a specific topic is ludicrous. That would be true for virtually any forum and any topic, from the perils of smoking to the need for training wheels on a tricycle.

[Mikey]

Ya know what ... I thought I was doing something good by starting a thread that would keep this topic from rearing it's head time and time again throughout the forum, and I hoped that it would lead to constructive, cooperative discussion. Instead, I get attacked for it. Screw it.

[Sydney]

RE: My 1st post:
#2 Find out about the amp

how many people blissfully went along with the manufacturer's recommendations for years and had no problems?

The problem is Ad copy implies 1 thing ( heavily disclaimer-ed and qualified ) and the fine print in the manual says another

For instance the popular QSC RMX series ( 850 - 2450 )
http://www.partsexpress.com/pdf/rmxman.pdf
The ads for this amp prominently display high power ratings for 2 Ohm load *
( Guaranteed to get guys excited and increase sales )
* Continuous Sine Wave @ 1% THD - Thermal or Overcurrent cutback limits duration.
( So If your after hi-powered distorted short bursts of Sine Waves...)

* However on page 13 in 4 languages:
DO NOT USE 2 OHM LOADS.

Chin up Mikey ( so your head is above the hype )