Formula for wattage when impedance is changed?

[brokenstrings]

I was wondering if anyone knows a formula to figure out what the new wattage requirements would be if two 8 ohm speaker cabs were ran together in parallel at 4 ohms?

I intend on running 2 pairs of 8 ohm speakers (250W program @8ohm) with one power amp at a 4 ohm load. I’m having difficulty choosing the right power amp due to not knowing the amount of wattage I will need per cab @ 4 ohms.

Hopefully Rockpage can come through.

[g1wgs]

Let me start out by getting the terminology straight.

the RMS or continuous rating of a speaker is what it will handle heat wise. Give it more power than this on "average" and it will probably fail. This is usually measured using a sine wave or "test tone."

The Program rating is usually double the RMS figure and is basically what the box will handle with music or speech.

The Peak rating means next to nothing, and is usually double the program rating. Pay no attention to it.

That said, if you have two 8 ohm cabinets in parallel, you are correct, it would be a 4 ohm load. Each speaker wants 250W optimally. So, you would need an amplifier that can produce 500W into a 4 ohm load.

The general rule of thumb when selecting amplifiers is that you want to give your speaker 1.5 to 2 times the RMS rating of the SPEAKER.

For instance, that 250W program speaker you have is probably 125W RMS. It should get 190-250W from an amp. For the amp, that's watts RMS.

I would not however, run an amplifier bridged into a 4 ohm load. If not limited and done correctly, the amp will probably overheat. So, you will need to find an amp that will do 500W rms into 4 ohms stereo.

I'd suggest the following
QSC RMX1450 or RMX1850HD
QSC PLX1602 or PLX1104
Behringer EP2500
Crown XLS602
Crown Xs500
Crest CA6

[lonewolf]

The Peak rating means next to nothing, and is usually double the program rating. Pay no attention to it.

The peak rating is very important when using professional high-power PA systems. It tells you how much instantaneous power the speaker can handle. This is very important when selecting the power amp and setting your headroom. To get the maximum headroom of 6db out of a typical professional speaker, you want to match the RMS power of the power amplifier to the speaker's peak rating.

In the case listed above, this formula would probably not be appropriate. You would probably do well with a minimum 500W per channel @ 4 ohms stereo power amp.

Behringer EP2500 2x650W@4ohms for about $330
Crown XLS 602 2x600W@4ohms for about $350

The Behringer is probably a little better deal since this is Crown's bargain basement model.

Musicians Friend is having a big sale on the Crown CE2000 @ $480 -- excellent price for this quality of amp.

[bassist_25]

I always have bridged power amps into 4 ohm loads and never had any problems with overheating, even when the amp was to the point of clipping. While a 4 ohm load will work an amp harder than an 8 ohm load, anything with the QSC, Crown, or Crest name on it shouldn't go into thermal shut down just because of the impedance load attached to it, so as long as it's designed to be able to bridge down 4 ohms. Generally rule, if an amp can run down to 2 ohms in stereo mode, it can usually bridge down to 4 ohms. That's not always case, for example, Crown's Micro/Macrotechs can bridge at very low impedances. Though always check with the designer's specs before making an amp purchase to make sure that it meets your needs.

You can't go wrong with any of the recommendations that g1 made. Just don't buy generic when it comes to power amps. Stay away from those cheap ass power amps on Ebay advertised as "Dj amps". Their power ratings are a bit misleading. They may have some amazing rating of 1600 watts or something, but it's not CLEAN power. If you're still new to the world of power amps, I would suggest going with something from QSC's RMX line, since they rather user friendly. Crown's are great, but they are designed for professional sound engineers, and can be rather imtinidating if you don't know what you're doing.

[g1wgs]

The Peak rating means next to nothing, and is usually double the program rating. Pay no attention to it.

The peak rating is very important when using professional high-power PA systems. It tells you how much instantaneous power the speaker can handle. This is very important when selecting the power amp and setting your headroom.

No it's not. The peak ratings on speaker ratings are myths. They are usually to give indication of the excursion limits of the cone. YOu should never use the number to select a power amplifier. There is no accepted standard for measuring peak power handling, so from speaker brand to speaker brand, it will mean different things.

To get the maximum headroom of 6db out of a typical professional speaker, you want to match the RMS power of the power amplifier to the speaker's peak rating.

That is completely wrong. To get the maximum head room out of a loudspeaker, you want to have an amplifier that can supply double the RMS rating into the given load.

Amps are measured for RMS with a 1kHz sine. A sine wave has a 3dB crest factor. If you then play a typical music program through the amp, you will never get the rated RMS power out of the amp because music has a crest factor of 6-10dB. HENCE, the need for an amplifier that is double the RMS rating of the speaker. A music signal with a 6dB crest factor can be reproduced at 125W without clipping by a 250W RMS amplifier.

In the case listed above, this formula would probably not be appropriate. You would probably do well with a minimum 500W per channel @ 4 ohms stereo power amp.

It is completely appropriate. And you stated the EXACT recomendation I did.

Behringer EP2500 2x650W@4ohms for about $330
Crown XLS 602 2x600W@4ohms for about $350

oh, surprise surprise. Two of the models I recommended.

[BDR]

My dad can beat up your dad ...

r:>)

[g1wgs]

I always have bridged power amps into 4 ohm loads and never had any problems with overheating, even when the amp was to the point of clipping. While a 4 ohm load will work an amp harder than an 8 ohm load, anything with the QSC, Crown, or Crest name on it shouldn't go into thermal shut down just because of the impedance load attached to it, so as long as it's designed to be able to bridge down 4 ohms. Generally rule, if an amp can run down to 2 ohms in stereo mode, it can usually bridge down to 4 ohms. That's not always case, ...

It really depends. If the music content is very dynamic you should be OK. However, bridging amplifiers is not for the faint of heart. Some amps require special wye adaptors to flip the polarity of one channel, and then using both the red posts on the back of the amp. If one doesn't understand it, stuff and melt and catch on fire.

If you put a compressor on your mix and keep the signal within 6dB of the average like some bands I see, then run that signal into a amp that is bridged 4 ohms and run it into clipping, it will most definitely thermal. Especially outside on a hot day, or in a stuffy club where thre amp's heatsink gets coated with a flithy grime from cigarette smoke.

4 ohm bridged is only recommened if you limit the signal and know what you are doing.

[lonewolf]

The Peak rating means next to nothing, and is usually double the program rating. Pay no attention to it.

The peak rating is very important when using professional high-power PA systems. It tells you how much instantaneous power the speaker can handle. This is very important when selecting the power amp and setting your headroom.

No it's not. The peak ratings on speaker ratings are myths. They are usually to give indication of the excursion limits of the cone. YOu should never use the number to select a power amplifier. There is no accepted standard for measuring peak power handling, so from speaker brand to speaker brand, it will mean different things.

To get the maximum headroom of 6db out of a typical professional speaker, you want to match the RMS power of the power amplifier to the speaker's peak rating.

That is completely wrong. To get the maximum head room out of a loudspeaker, you want to have an amplifier that can supply double the RMS rating into the given load.

Amps are measured for RMS with a 1kHz sine. A sine wave has a 3dB crest factor. If you then play a typical music program through the amp, you will never get the rated RMS power out of the amp because music has a crest factor of 6-10dB. HENCE, the need for an amplifier that is double the RMS rating of the speaker. A music signal with a 6dB crest factor can be reproduced at 125W without clipping by a 250W RMS amplifier.

In the case listed above, this formula would probably not be appropriate. You would probably do well with a minimum 500W per channel @ 4 ohms stereo power amp.

It is completely appropriate. And you stated the EXACT recomendation I did.

Behringer EP2500 2x650W@4ohms for about $330
Crown XLS 602 2x600W@4ohms for about $350

oh, surprise surprise. Two of the models I recommended.

I made the same recommendation as you because this rig was obviously not going to be run by a trained engineer. Your rules of thumb are valid for systems that are run by musicians or roadies, but not trained engineers.

[g1wgs]

I made the same recommendation as you because this rig was obviously not going to be run by a trained engineer. Your rules of thumb are valid for systems that are run by musicians or roadies, but not trained engineers.

No, you said:

In the case listed above, this formula would probably not be appropriate.

Please explain how proper amplifier selection works in the "pro" world.

[lonewolf]

This is how to get the most out of a pro speaker system by using specifications. This method used to appear on engineering data sheets before the advent of PROGRAM POWER:

Pro PA speakers have an instantaneous PEAK power rating of 4x the RMS power rating, or +6db of instantaneous power headroom above RMS built into the speaker.

If you want to exploit this feature, you need clean, unclipped power all the way up to the peak value of the speaker. To guarantee the amp will not clip at this power level, it must have an RMS power rating equal to the speaker's peak rating. Of course, you only run the amp at about 1/4 to 1/2 power so that only the musical peaks reach full rms power.

Care must be taken if this is used.

Here is a disclaimer from an EV Engineering Data Sheet regarding this configuration:

The caution cannot be made strongly enough, however, that this arrangementis only for experts or for those who can discipline themselves against “pushing” the system for ever-higher sound levels and who can avoid “accidents” such ascatastrophic feedback or dropped micro-phones.

[g1wgs]

This is how to get the most out of a pro speaker system by using specifications. This method used to appear on engineering data sheets before the advent of PROGRAM POWER:

Pro PA speakers have an instantaneous PEAK power rating of 4x the RMS power rating, or +6db of instantaneous power headroom above RMS built into the speaker.

If you want to exploit this feature, you need clean, unclipped power all the way up to the peak value of the speaker. To guarantee the amp will not clip at this power level, it must have an RMS power rating equal to the speaker's peak rating. Of course, you only run the amp at about 1/4 to 1/2 power so that only the musical peaks reach full rms power.

Care must be taken if this is used.

Here is a disclaimer from an EV Engineering Data Sheet regarding this configuration:

The caution cannot be made strongly enough, however, that this arrangementis only for experts or for those who can discipline themselves against “pushing” the system for ever-higher sound levels and who can avoid “accidents” such ascatastrophic feedback or dropped micro-phones.

I disagree. That is ONE manufacturers take on it. EAW recommends only 1.5 to 2x the RMS rating on ALL of their cabinets.

There is no industry standard for measuring peak power. 6dB of power headroom is NOT always built in. Some speakers will go beyond the excursion limits of the cone and damage can result. The two main failure modes of any loudspeaker are heat, and over excursion. Given a signal of high enough RMS voltage, you will melt or severely damage the voice coil. Given enough peak power you will damage the motor structure of the cone and/or spider.

In actuallity, you can put any size amp on any given speaker. the CORRECT and PROFESSIONAL way of doing so is by using a quality limiter on the crossover outputs. By correctly calculating the voltage gain and maximum SPL required, the full benefits of a loudspeaker can be realized. It has next to nothing to do with using peak power specifications.

Most people don't have a true limiter at their disposal nor the cash to buy one. Following the 1.5 to 2x rms rule is pretty much standard practice in the sound reinforcement industry, along with bringing enough PA to the gig.

[lonewolf]

Mackie's (er, I'm sorry, EAW's) "rule of thumb" is also just one manufacturer's recommendation. But, simply do the math:

EAW's 2X "rule of thumb" is based on the assumption that the power amp will never clip at instantaneous peak power and that the amp's peak power = 2X RMS power. That is the only way to theoretically get 6db headroom out of this arrangement. In practice, it rarely attains this level.

To do this, it requires running the amplifier so that it is at the threshold of clipping every time the music peaks. The 4x method delivers guaranteed clean power at these levels with absolutely no chance of amp clipping. That is the chief difference between the two methods.

Also, class D power amps do not provide 2x peak ratings...the peak level is only at 1.4 x the RMS value. In this case, you will only get about 4db headroom.

I'm not saying the 4x method is for everybody, it just happens to be a legitimate configuration that uses a speaker's peak power rating.

If you don't agree with it, then you probably shouldn't use it.

[g1wgs]

Mackie's (er, I'm sorry, EAW's) "rule of thumb" is also just one manufacturer's recommendation. But, simply do the math:

Cute.

EAW's 2X "rule of thumb" is based on the assumption that the power amp will never clip at instantaneous peak power and that the amp's peak power = 2X RMS power. That is the only way to theoretically get 6db headroom out of this arrangement. In practice, it rarely attains this level.

Who the heck needs to get their headroom that way? The simple fact is that modern mixing consoles have +18-24dBu of headroom built into the mixer. Most if not all power amps clip between 0 and 6dBu of input. Additionally, running the speaker system to their limits when they will EASILY hit 130 dB, is ridiculous. The distortion at those levels will start to show. And so will the blood running from you ears. Attenuating the amps 10 dB or so, and running the console around 0 to +6 will provide the headroom and volume you need. If it's not loud enough in all the seats, then you didn't bring enough PA to the gig. By that I mean number of boxes with a mixture of long throw and nearfield cabinets.

In terms of handling peak transients, take a QSC MX1500a for example. The rail voltages are +/-96 and +/-48V. It's a two tier class H design. This amp is capable of delivering 500W per channel into a 4 ohm load.

In order to do so, the RMS voltage needs to be about 45V. In order to produce that, the peak to peak voltage only needs to be 127V. In the case of the mx1500a, the peak-to-peak voltage is 192V. That can supply 68Vrms. That would be about 1100W. Can the mx1500a do this for long.... no. But we are only talking about peak transients.

Now remember, having selected a 500W RMS per channel amp, we would have been powering a 250Wrms speaker. The peak headroom in the amp is sufficient to do this.

Like it or not, 1.5 to 2x is a defacto standard in selecting amplifiers. Class D amplifiers are pretty rare, Italian made Powersofts and Crown iTechs are the only ones I can think of off the top of my head. Though iTechs are a modified class D.

If you don't agree with it, then you probably shouldn't use it.

It's not that I don't think there are other methods. It's just that you touted this 4x rms as the way professionals do it. I think you are wrong, and it depends on the application.