Measuring Amp Power Output Under a Load
Has anyone tried measuring an amp head's (micro in my case) power output using something like a Kill A Watt?
I'm interested in how amp power output is measured under a load. I'm looking at getting an amp capable of pushing more power than I need at the moment. But I like the future flexibility and great price of the head. Ohm loads are not the issue with this amp since it can handle funky 2.67 ohm loads down to 2ohm. So, I'm not really looking for amp/cab suggestions.
A high powered amp can be used with a variety of cabs as long as the gain, master volume, and low freq. boosts are not cranked up. So the question becomes: where exactly (which given the gain/volume/boost variables becomes approximately) can you safely dial the power output knobs and where should you stop?
I've read and heard various suggestions, all good ones, about 'listen for cab distortion' and 'a 900watt amp pushing a 450watt rated cab shouldn't go past 12:00 volume'. Good enough except when dealing with high quality speaker cabs and gain/low freq boosts all over the place. Yeah, there are peak lights but they don't seem to tell me to turn down and all I hear is 'cranking dude!' when I see them.
It seemed to me that there should be a way to measure the power output and protect my precious cabs. And I have this Kill A Watt anyway. Anyone else tried this?
That device measures power consumption at the wall socket, not power output at the speaker.
The "proper" way to measure an amp's output power is to connect a dummy load to the output (instead of a speaker). Typically a sine-wave is then input into the amp and the output of the amp monitored--using either an oscilloscope or a distortion analyzer-- until the amp is observed to clip or reach a certain distortion level. At that point current and/or voltage can be measured and the power output calculated.
There's some simplifications going on with this method, but it's a pretty fair way to see what an amp can actually do.
If you are wanting to know how much power your amp is putting out into the speaker at any moment, you need a wattmeter. This is a kit that probably just measures voltage across the speaker, and you set a switch to select 4 or 8 ohms, and the LEDs light up to show you power level.
There's only a few frequencies when the load curve intersects is the fixed 4,8,etc. Ohms line the calculations are based on.
On other frequencies the result is incorrect.
The method nashvillebill explained is the correct one.
The meter You're linking to can be used to guesstimate a lot of things in MI/PA world though, including the average power draw of Your amp for inverter or generator use for example.
If talking about SS:
Class A operates at ~25% efficiency.
Class A/B at ~50% efficiency.
Class G & H at ~65% efficiency.
Class D at ~90% efficiency.
Tube amps ~10% less, respectively.
And since class D tube circuit is very different from SS class D, nowhere near that.
It's worth noting that if the amp uses a SMPS for its power supply, wall meters won't be trustworthy. This was noted a few years back among those who review computer power supplies, with reviewers actually getting higher than 100% efficiency numbers because the wall meters weren't getting a proper reading.
Also, in the realm of bass guitar, we almost always run out of speaker excursion (i.e. the cab starts to "fart out") before the speaker gets enough power to worry about overheating. Ears are king for determining bass limits.
The "watt" ratings of speakers are meaningless because they apply to the thermal power handling of the voice coil, and not the excursion limit of the system. As Mehve points out, most bass speakers will hit the excursion limit before they hit the thermal power limit.
Also, a simple voltage or power measurement won't tell you what the voice coil is doing, because excursion is frequency dependent. The measuring device would have to model the motion of the cone. I think that Eminence may be doing that with their new speaker protection system.
Yeah, thanks Mehve and fdeck. I hear you. The speaker excursion is going to fry my cabs before the voice coil stars going nova. Guess I just need to learn to listen more for that 'farting' speaker excursion sound.
You can't always here that kind of distortion. With a well designed cab, you may take it up to it's limit and it may sound fine. Especially if you have any addition "distortion" tone in your signal chain.
Or running for a long period while hitting thermal limits may melt something and ruin the driver.
The driver is in an enclosure, the driver manufacturer rated it free air. Heat has less chance to escape from a cabinet unless the cabinet designer thought of this.
The best bet is use a limiter set for half the power rating of the drivers. This is still not foolproof. Sometimes drivers blow but they can be repaired. Lessons learned, don't be so hard on it next time.
I don't do those kind of gigs anymore.
I'm not quite sure, however, I understand how a THD Analyzer might apply here since I was hoping to measure power at different points and AFAIK the Analyzer is good for determining the amps ability to do accurately what it says it will do (duplicate a sine wave) given a certain amount of power. But again, I've never used one.
The dummy loads are easy http://www.parts-express.com/cat/dum...resistors/1535 combine these for your expected needs plus 50% for a safety factor. Non-inductive will help to keep the amplifier from being influenced by the load attached.
You are now opening a BIG can-o-worms. A "standard" was developed to try and level the playing field for amplifier claims. It involves running the amp at a minimum rated distortion level, measuring the peaks of the "clean" signal and taking a mathematical Root Mean Square (0.707) for the average of what a customer might perceive (hear). Amplifiers can and will far exceed these ratings with increased distortion depending on the ability of the power supply designed in the amp as well as the chosen output power handling devices abilities (tubes, SS devices, operational final amplifier class A, B, A/B, C, D, G, H).
Confused yet? :)
In other words it's possible, but a complete waste of time, money, and effort.
When we set limiters for subwoofers, you run a sinewave in the front of the system somewhere in the subs passband, 50-60hz or so, and measure voltage across the output terminals. ( music is A/C ).
Voltage gets calculated to various amounts of watts depending on the impedance of the speakers but the voltage stays the same. An amplifier is pretty much a "voltage multiplier" in very simple terms.
Edit to add:......when your strings vibrate across the poles of a magnetic pickup, it creates a very small voltage swing from positive to negative and back again. This is very small millivolts of a signal. The preamp takes that and multiplies it, boosting it up to some thing strong enough to "drive" the poweramp. The poweramp then takes that and multiplies it some more, in order to "drive" the speakers to achieve a volume level to fill the venue with sound. A "bass amp" is a preamp and a poweramp all built into one box.
Again, very simplified explanation here.
When you figure in the very dynamic thing that is a plucked/picked/whatever string vibrating over a magnetic pickup on one end, and the very dynamic thing that is a coil of wire "floating" in a magnetic field attached to some cone with a given mass, "stiffness", attached to a suspension with varying amounts of "give" or damping depending on power applied on the other end commonly called a "speaker".....it can get to be a bit much. :)
That is why "watts" are not not such a good ruler as we are not talking light bulbs or space heaters :D But it's all we have up to this point. ;)
I can utilize 2400 watts by touching the hot and neutral together on a 120vac service, but it doesn't make all that much sound for the power used.
you are interested in is below clipping level. Also, simply feeding in a sine wave and setting the
master volume for some particular output power won't help either; the sine wave test signal has
an arbitrary level. You need something directly related to a real bass signal.
The best and simpliest way is as already mentioned - listen to the sound from the speaker.
Simple rules like, 'a 900watt amp pushing a 450watt rated cab shouldn't go past 12:00 volume',
make a pretty big (and false) assumption that full output occurs at a setting of '10'. Further, they
also assume that a setting of '5' is one half of whatever power '10' is. Also false.
There are a couple ways you could do it though, to answer your question.
As already mentioned, typical power meters, which actually measure voltage and convert to power,
do not measure actual power over various frequencies; they are only accurate when the actual load
is equal to the expected load.
However, they can give you reasonably accurate relative power levels. So if you had an amp and
cabinet setup with full power capabilities, you could get a reference reading for full power operation
with real bass signals. Then with a lower rated cabinet, you would lower the master volume for
correspondingly lower readings on the power meter. But all your signal conditions have to be similar
for this to work. Any change in tone can significantly alter the power output.
The best way to do what you want would be with an adjustable limiter in the amp itself.
A limiter, if it's late enough in the ampifying stages, can act as a maximum power output control.
You could set the output power using a dummy load and an input signal large enough to drive
the limiter on. Set the limiter to give you an AC voltage based on the power and the load:
voltage = square root of (power x load resistance).
So if you wanted 200 watts, 200 w times 4 ohm = 800, square root of 800 is 28.3. You would adjust
the limiter setting to give you 28.3 VAC on a 4 ohm load.
You have to make sure the limiting action is in effect, i.e., increasing the input signal does not increase
the output, and that the output is not limited elsewhere in the amplifier stages.
|All times are GMT -6. The time now is 08:17 AM.|
Powered by vBulletin® Version 3.6.12
Copyright ©2000 - 2013, Jelsoft Enterprises Ltd.