How to determine amplifier output

[tejcurrent 13]

I have read many places different ways to determine the output of your amplifier as it is in your car, but none have given me the information I need to do this myself. I will be auditioning several amps in the next few weeks, and would like to have real numbers as they pertain to what my subs will really get.

Can anyone break it down to where I can mess it up, as far as how to go about this, what equipment I will need, and the approximate cost of the equipment? I'm sure I'm not the only one that would love to do this, and this will allow me to give accurate results and power output when I'm testing the amps.

Thanks in advance.

[CBFryman 2]

Osciliscope

True RMS Amp meter

True RMS volt meter

or

True RMS amp meter

True RMS volte meter

Stable load that doesnt induce current phase change.

Osciliscopes range in price. Osciliscope cards for computers run for about $150 and a little more for good software. Stand alone units can be had for $200-500 used and new they can be several thousand.

True RMS amp meters run around $100 and I reccomend Fluke Highly.

True RMS volt meters can be had in almost any decent DMM. Good DMM's usually cost over $50. Agian I reccomend Fluke.

For a stable load you can purchase them, or get a bunch non inductive 25w cremic resistors (like $1.50 each), wire them to the load you need, then submerge them in a bucket of distilled watter after cleaning them with acetone so the water doesnt get contaminated with electrolytes.

[///M5 2,833]

You left out one major piece Chris. You will need some way of measuring distortion as well and along with that you need to determine what level of distortion is acceptable and test up to that level. To make things even more difficult, you could question which harmonics are actually important to you in that measurement and what actually defines distortion. A lot of speaker designers that I have worked with in the past measure up to the 50th harmonic, which at low frequencies is fairly easy but even up at 20kHz is a completely different story. You'd be surprised how many manufacturers cannot even do this measurement and have to outsource it to a testing house. To do it right it is really expensive. The load that Chris proposes isn't stable enough in my book.

[CBFryman 2]

Good point sean.

[mrray13 30]

bah..lol....sean does this crap for a living and his equipment is way beyond the means of mortal men.

what chris suggested will work for most in car, daily applications. (my god, i'm agreeing with chris. someone call hell and check the temperature) or if your after musical rms, just use your pair of dmms, check input dc voltage and your output ac voltage. do the math and viola' ! a fairly accurate representation of what your amp does in real world.

truth is, a singular stable impedance doesn't happen in real world application, even in burp boxes. so an amp's power is never a stable item as a result. i say go real world, or go home, lol..

wheeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

[///M5 2,833]

Come on, it isn't fair to compare one amp at 10% distortion to another at .1%. Without measuring the spectral content you could easily do that by accident as well. How do you think these chump companies can claim they are selling an 8000w amp, when in reality it might do 50w a channel. It is pretty easy to meet a spec if you change how you measure that spec, without taking this into consideration in your measurement it won't be accurate enough to draw the conclusions that you are trying to.

[DevilDriver 28]

I couldn't agree more with Sean. If you're just wanting an approximation for a "daily application", then just use whatever the manufacturer publishes. If you're going to test a product with the goal of producing any measurements worth comparing to the manufacturer's published information, then it only makes sense to test in the same conditions.

[helotaxi 46]

My opinion, if you are using it for competion (SPL that is), test it on the actual load it will be driving. Tests into a stable load don't really matter other than giving a benchmark for general comparison. If the usage is specific, so should the test be.

For basic personal enjoyment, the real test is "is it loud enough for me before audible distortion?" If the answer is "yes" the RMS power output = enough. If the answer is "no" the RMS power output = not enough.

For daily use the power breakdown is really that simple: enough or not.

[mrray13 30]

unless he's doing some sort of write up, ie: publishing a book of amp performance manufacturer specs vs true benching, there simply isn't a need for such stable measurements. unless his box/sub combo holds that same level of stability, it doesn't matter what the amp does on the bench. it's going to change in car. worried about distortion? fine. set the gains using a scope and measure output. but still use the load the amp is going to be seeing in car. then you'll have real world power, not what happens on a bench.

example, my hifonics nemesis is set using a scope. rated 750 wrms, it'll produce 840 wrms clipped. (according to ac output x dc input) unclipped, i'm around 450 (unclipped output. i haven't bothered to check the LOC converted input signal) . musical. i didn't use tones. why? it'll never see tones. no need to set that way. no truer measurement of power needed, IMO.

wheeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

[///M5 2,833]

My opinion, if you are using it for competion (SPL that is), test it on the actual load it will be driving. Tests into a stable load don't really matter other than giving a benchmark for general comparison. If the usage is specific, so should the test be.

For basic personal enjoyment, the real test is "is it loud enough for me before audible distortion?" If the answer is "yes" the RMS power output = enough. If the answer is "no" the RMS power output = not enough.

For daily use the power breakdown is really that simple: enough or not.

Since the OP's first post specifically mentioned "real numbers" outside of the car environment, he would need to test versus a stable load at a stable output and a consistent distortion level. However, I 100% agree with the statement above that I bolded. If the application here is to figure out which amp will do better for you in an SPL world, design and build your box first and then test the amps on the driver in the box you have chosen.

[CBFryman 2]

An LS1 is an LS1, no point in putting the car on a chassis dyno. Car weight, rotating inertia, and driver skill will all affect it's straight line perfomrance.

[mrray13 30]

I will be auditioning several amps in the next few weeks, and would like to have real numbers as they pertain to what my subs will really get.

hey sean, the OP wants to know "real numbers" his subs will be getting. as quoted above. that tells me in car, in enclosure using his subs as the final load. thus, using the methods described by chris and myself should be adequate, IMO

hehe..

wheeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

[helotaxi 46]

As would my "enough or not" measurment...

[Tirefryr 1,742]

If it's a quality product, I see no point in not trusting the manufacturers spec. Of course this can cause problems when you get something like RF who likes to rate some amps at more than 100% less of their actual output.

[tejcurrent 13]

hey sean, the OP wants to know "real numbers" his subs will be getting. as quoted above. that tells me in car, in enclosure using his subs as the final load. thus, using the methods described by chris and myself should be adequate, IMO

YES. I am not trying to write a book of definitive results. I am not trying to prove or disprove what any companys say is true. I am trying to give myself (and others) an idea of what they can expect from a given amp in a realistic environment, compared to the benches of those with excessively expensive testing equipment, and resources unavailable to most consumers.

I want to have results that are specific to my situation, in that it will be more realistic than what some will probably expect. I would like to start a topic stating something like, amp X put out 3913 watts @ Y load with Z volts. Many amps don't do what they claim, either over or under, but I would like to say for those looking for an amp that in a vehicle with X electrical hooked up to a single BTL dual 2ohm

amp 1 puts out xxx1 watts

amp 2 puts out xxx3 watts

And so on.

NO this will not be scientific at all in the classic sense, but it will be very realistic so that people don't get this 10gigawatt amp only to find out it does 3 gigawatts.

Loud enough or not is a very good measurement too, but if a person was trying to hit the highest #s possible in a comp, there wouldn't really be 'enough', just the constant quest to advance.

For the common consumer, what are the steps involved in testing their specific amp in their specific vehicle in their specific configuration?

Thanks for all the speedy replies. (T-shirt?)

Edited March 9, 2007 by tejcurrent

[///M5 2,833]

I want to have results that are specific to my situation, in that it will be more realistic than what some will probably expect. I would like to start a topic stating something like, amp X put out 3913 watts @ Y load with Z volts. Many amps don't do what they claim, either over or under, but I would like to say for those looking for an amp that in a vehicle with X electrical hooked up to a single BTL dual 2ohm

To what accuracy do you want to make that statement? The reason I ask is that with the methods described above you would still be at best with a 10% number, which in your example is nearly 400watts. To me a real number is somewhere in the better than 1% realm.

[tejcurrent 13]

Well, I have posted a thread in the Fi forum, I will be comparing close to a 1/2 dozen amps if things work out right, all supposed to be putting out 4000 watts or higher. I will be testing using the same few test tones off the same cd, with all the conditions the same. I would like to be as accurate as possible, but there is simply a line I must draw as to how much I can spend on testing equipment.

You are right, I do not want to say 4000watts + or - 400. Whatever the margin of error, I would like to keep it equal for all of the amps I test. I am most interested in what these amps do in my particular setup.

[CBFryman 2]

In your case and osciliscope, and the two RMS meters is the route to go. If you just want to know aparent power (what 90% of these RMS tests you see on the web are) then skip the osciliscope except for gain setting.

[tejcurrent 13]

Well, I don't want to shell out the cash for the osciliscope, but the rms meters are a go. Maybe in a few months I'll get it, but right now it's just not practical. So using the 2 meters, can anyone systematically describe the best way to get the most accurate measurements? Like I said, I know the margin of error is much greater than would be acceptable for most standards, but this is really just for my personal use, and I think the info will help many others who are interested in finding out what's happening with their amps when put in a real world scenario instead of a laboratory. Hopefully it'll help those looking for a powerful amp get a better idea of what will really go to their subs in their vehicle. Thanks.

[///M5 2,833]

Unless you really have some money for gear, you are better off using a mic and just testing for peak output. Or of course following HT's " "enough or not" measurment..."

[tejcurrent 13]

Well, I don't want to shell out the cash for the osciliscope, but the rms meters are a go. Maybe in a few months I'll get it, but right now it's just not practical. So using the 2 meters, can anyone systematically describe the best way to get the most accurate measurements? Like I said, I know the margin of error is much greater than would be acceptable for most standards, but this is really just for my personal use, and I think the info will help many others who are interested in finding out what's happening with their amps when put in a real world scenario instead of a laboratory. Hopefully it'll help those looking for a powerful amp get a better idea of what will really go to their subs in their vehicle. Thanks.

OK, then I'll just forget it until I can get the proper equipment (come on used stuff). Anyway, I'll say 2k watts = not loud enough.

[mrray13 30]

an oscope can be had for cheap, i got mine for $20. to be realistic and fair, snag a scope, set the gains to where they aren't clipping and check your voltages. (the process is easy enough, measure and record dc input to the amp. measure and record ac output to the speakers and multiply the two. ie, 14 dcv input times 130 acv output = 1820 watts)

i'll ask my guy here, i think he has another scope he wants to get rid of.

wheeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

[helotaxi 46]

(the process is easy enough, measure and record dc input to the amp. measure and record ac output to the speakers and multiply the two. ie, 14 dcv input times 130 acv output = 1820 watts)

Curious as to the origin of that math sense it really makes no sense because it totally fails to account for load and thus current which are vital to any electrical power calculation. According to this, 30V on the output at 14V on the input is going to be the same output wattage regardless of load on the output.

[tejcurrent 13]

You know a place to get one on the cheap I'll pick it up. If someone wants to sell theirs I'll get it, just let me know.

[CBFryman 2]

(the process is easy enough, measure and record dc input to the amp. measure and record ac output to the speakers and multiply the two. ie, 14 dcv input times 130 acv output = 1820 watts)

Curious as to the origin of that math sense it really makes no sense because it totally fails to account for load and thus current which are vital to any electrical power calculation. According to this, 30V on the output at 14V on the input is going to be the same output wattage regardless of load on the output.

Good catch. I didnt see that before. RMS Vac output * RMS Aac Output = Apparent Power.

Vdc input * Adc input = Powerdraw on the system many Clamp meters can not read DC current though

Apparent Power / Power Draw* 100 = approxamite amplifier efficnecy at that level.