Amp testing/comparison
Printed From: the12volt.com
Forum Name: Car Audio
Forum Discription: Car Stereos, Amplifiers, Crossovers, Processors, Speakers, Subwoofers, etc.
URL: https://www.the12volt.com/installbay/forum_posts.asp?tid=35718
Printed Date: May 16, 2024 at 10:42 PM
Topic: Amp testing/comparison
Posted By: Leif
Subject: Amp testing/comparison
Date Posted: July 16, 2004 at 8:29 PM
Hi there, I'm Leif, and I'm new here. :)
I'm currently building a system, and I'm trying to figure out which amp to use for my front speakers (Alphasonik components in kickpods).
I have three amps laying around.
Sony Xplod XM-7547 - a 4 channel amp, 4x75w or 2x300w bridged at 14.4 volts/4 ohms, according to the manual. Fused at 100 amps.
Alpine MRV-1507 - a 2 channel amp, 2x225w at 14.4 volts/4 ohms according to the manual. Fused at 80 amps.
Alpine MRV-F407 - a 4 channel amp, 4x60w or 2x160w at 14.4 volts/4 ohms according to the manual.
I've learned not to put that much weight on specs, so I decided to test them to see which one would put out the most power.
I hooked them up to my (running) car, and two Alpine Type-R 12 inch woofers (yielding two 4 ohm loads).
I also connected an Oscilloscope in parallel with one of the woofers - so that I could actually look at the waveform and see when they clip.
Then I played both deep, bassy music (Ying & Yang twins - Salt Shaker) and very peaky bass (Alicia Keys - If I ain't got you).
I tested each amp in turn separately, and turned up the gain on the amp until I saw clear clipping on the scope.
I was quite unpleasantly surprised when I noticed that ALL THREE AMPS CLIP AT VIRTUALLY THE SAME LEVEL.
I.e., all three amps are outputting the same power - regardless of the specs!
So, my question is.. Is it really supposed to be that way? I mean, the MRV-F407 is half the size of the other two. I suppose maybe the MRV-1507 would produce more power into a lower impedance load, but..
What's going on? Did I do something wrong, or is this normal? And, if I want more power than this into my front speakers, which amp should I get instead?
///Leif
Replies:
Posted By: stevdart
Date Posted: July 16, 2004 at 9:07 PM
What was the power output at the highest level?
-------------
Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.
Posted By: customsuburb
Date Posted: July 16, 2004 at 9:11 PM
Alpine is pretty honest with their specs, sony is not. There is no need to find the maximum power output. This rating really has nothing to do with sound quality. Pick the amp that has the continuous power rating closest to your speakers continuous power rating (besides the sony).
Posted By: Leif
Date Posted: July 16, 2004 at 9:44 PM
To clarify, power output is a term so commonly misused that I don't even know how I would measure it. Clipping voltage however (at a given load) is more of an irrefutable measurement, and really tells how loud an amplifier can play without distortion.
So, I didn't actually measure absolute power - I instead compared the three amps (with very differently specced output powers) and noticed that they clipped at the same voltage, thus all producing the same power.
Customsuburb, the two alpine amps with different specs had equal maximum output power at 4 ohms. Let's assume that my speakers can handle infinite power (just to take that out of the equation), and that the amps sound identical (which they do), then it comes down to which amp has the highest power output.
///Leif
Posted By: stevdart
Date Posted: July 16, 2004 at 9:50 PM
So, let me ask again. What was the power output?
-------------
Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.
Posted By: Leif
Date Posted: July 16, 2004 at 10:02 PM
Stevdart, I just said I don't know how exactly to measure it.
Or to be precise, I don't know how to translate the voltage reading I got into watts.
Could you tell me how to do it?
///Leif
Posted By: stevdart
Date Posted: July 16, 2004 at 10:09 PM
Oh sure, I thought you would just give the voltage reading. Use P = Esquared / R, where you square the voltage reading and divide by the ohm load to find power in watts. You showed the amp ratings but I couldn't tell which of the Alpine amps is rated closer to what you found.
-------------
Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.
Posted By: Leif
Date Posted: July 16, 2004 at 10:20 PM
Stevdart, I believe I was getting around 12 volts peak to peak, which would equal close to 150 watts. (24*24=576, 576/4 = 144).
I suppose this makes a lot of sense, because unless they put a switching power supply in to get the input voltage above 12, a bridged amp would only give you -12 to +12 volts.
So, with these measurements, the smallest and cheapest amp (the Alpine MRV-F407) had the truest specs.
///Leif
Posted By: stevdart
Date Posted: July 16, 2004 at 10:37 PM
In order to judge the power output comparison of the three amps, you would have to read AC volts at the speaker terminals that the load is connected to. Use a digital multimeter. Here is a link to a good tutorial for measuring power output, although it's not done at the car. ------------- Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.
Posted By: Leif
Date Posted: July 17, 2004 at 2:35 AM
Indeed, except I wasn't trying to measure power outage in watts, I was comparing three amps, and they all came out very close to one another.
So I guess it comes down to living with it as is, or getting another Alpine MRV-1507 amp (since it can be bridged, but will then only drive one speaker).. Of course, that'd be 900w RMS per speaker or so - might be a little much for a set of 6.5" components, even if they are alphasonik's :).
I guess I'll just go for the smallest of the amps (space is always a consideration).
Unless someone else has a different take on it, of course :).
///Leif
Posted By: stevdart
Date Posted: July 17, 2004 at 7:30 AM
Sorry, dude, but if you want to believe that you made some kind of realistic comparison of those three amps, go ahead and believe it. But it's flawed. If you don't make an accurate measurement, then you are obviously relying on the volume that you perceived to hear. That's no basis for an anouncement of amplifier power vs. published ratings. I suggest you study up on the fundamentals of audio electronics. There's a lot to be learned in this forum from some knowledgeable people, who insist they are learning something new every day. (I am so far behind many of them that I learn 100 things a day.) And welcome to the forum! ------------- Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.
Posted By: Leif
Date Posted: July 17, 2004 at 9:56 AM
I think there's been a misunderstanding of my test methodology, let me clarify again. If I'm mistaken, please enlighten me, but be specific.
I played three different amplifiers with the same program material into identical loads (the same set of woofers) and looked at the test signal with a scope. (Woofers aren't acceptable for absolute power measurements, but fine for direct comparisons)
When testing the first amp, I adjusted the scope so that the amp's clip level (clearly visible on the scope) ended up exactly on the -50% and +50% lines on the scope. I subsequently didn't touch the scope at all, but just hooked up the different amps and adjusted the gain until I saw clipping. (I wasn't listening to the woofer output, so my ears are out of the equation for this one.) This is where there accuracy comes in - if the scope shows that the amps all clip at the same level, they're producing the same power.
Which part are you disagreeing with exactly?
///Leif
Posted By: stevdart
Date Posted: July 17, 2004 at 3:11 PM
When you proclaim that three amps, from two companies, all with different power ratings, are identical in power output because you devised your own way of testing, which is not an accurate testing procedure, someone is going to call you on it. You have to use a proper testing procedure (please read the link I provided to you) to make such a proclamation on a forum that is read by as many people as this one is. Frankly, I don't care that you think the three amps produce the same power. But somebody reading this may get the impression that the test was proper. When you said that the voltage reading was "about 12 volts peak to peak", that gave you away. You know just enough about the terminology, and have a fine grasp of the English language, to sound almost convincing. But anybody who has tested voltage output knows better. You have an oscilloscope, now you should obtain the rest of the equipment described in the link and bench test those amps properly. You'll be glad to learn it the right way. Then you will be able to speak with authority. Sorry to call you out like this, but when you brush off a knowledgable expert like customsuburb, and announce conclusions that are false, it has to be done. He gave you the best advice in this thread. You should re-read it. And good luck with the testing. If I am wrong, somebody on the forum call me on it. Then that'll be another new thing I've learned. I'm done with this topic, unless I get called out. Then I'll apologize to you and then be done. ------------- Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.
Posted By: DYohn
Date Posted: July 17, 2004 at 3:26 PM
To second Stevedart, if you actually read 12 volts AC on the output of the amplifier into a 4-ohm load, that corresponds to 36 watts of power. Ohm's Law. This fact alone should clue you that there is some flaw in your testing methodology, especially if three different amps gave you the exact same results. Besides the fact that this is way below the ratied power on any of the amps, three amps with the same make/model numbers will not often produce the exact same results. ------------- Support the12volt.com
Posted By: kgerry
Date Posted: July 17, 2004 at 5:41 PM
just to chirp in as it hasnt come up yet....when doing any type of measurement like this try to do it with a quantifable known source signal, like a solid 1K test tone, etc...measuring a music floor is rarely accurate because the source signal is bouncing all over the place.....some may say it's a more realistic measurement re: dynamics but i always use either a 400 Hz or 1 Khz test tone.......
-------------
Kevin Gerry
Certified Electronics Technician
MECP First Class Installer
Owner/Installer
Classic Car Audio
since 1979
Posted By: Leif
Date Posted: July 17, 2004 at 11:53 PM
Fair enough, time for me to put up or shut up. (I must admit I am a little curious to see if the standard test methodology will yield the same results.)
I have a 500ft spool of 18GA wire (stranded single conductor, insulated), its resistance is 3.6 ohms according to a digital multimeter. Stevdart, do you have any problem with me using this wire as the dummy load? Also, should I perhaps unwind it into a disorganized pile to make it less of a coil?
Furthermore I'll pick up an Optima Yellow Top battery (there's no way I'm carrying all the gear out to the car again, way too much work) to use as the power supply. I'm planning to keep a 10A 13.8V supply (radio shack) hooked up to it to at least mitigate any battery discharge between testing the different amps.
I will use a modern discman (digital volume control) with continuous test tones recorded on CDR, a digital multimeter in AC mode to monitor the output voltage, the same oscilloscope to keep the gain just below amp clipping, and another multimeter in DC mode to monitor the supply voltage.
If there's anything wrong here, please let me know quickly so that I can get it right this time.
///Leif
Posted By: Leif
Date Posted: July 18, 2004 at 8:43 PM
Okay, the measurements are done. I scrapped the idea of a long-wire resistor (research showed that it would be a highly inductive load) and instead combined 25 watt high power resistors to a total resistance of 5.0 ohms, 150 watts. (Resistance increased to 5.3 after they heated up from the testing).
Supply voltage was a brand new Optima Yellow Top battery - voltage measured at 11.9 volts during testing (full signal load).
I tested with a clean 50hz sine wave, and used the oscilloscope to make sure it was just a hair from clipping. Just to get it out of the way, I also looked at the amp input signal on the oscilloscope - no clipping.
The results are:
Sony XM-7547 - spec says 2x300w into 4 ohms - actual voltage was 29.2 volts 50hz AC at 5.3 ohms, yielding 160.8 watts.
Alpine MRV-F407 - spec says 2x160w into 4 ohms, actual voltage was 27.7 volts 50hz AC at 5.3 ohms, yielding 144.8 watts.
Alpine MRV-F1507 - spec says 2x225w into 4 ohms, actual voltage was 29.5 volts 50hz AC at 5.3 ohms, yielding 164.2.
Stevdart, this is to me very close to my original figure of 150 watts per channel. Sure, it wasn't absolutely accurate - now it is.
So, 4 hours and $200 later, now that the test methodology is out of the way, would someone please answer the original question, which is, is it normal for amplifiers of such different specified wattages (and different purchase costs) to produce for all intents and purposes identical output power?
(The difference between 164 and 144 watts is much less than one single decibel, and inaudible.)
Thank you.
///Leif
Posted By: haemphyst
Date Posted: July 18, 2004 at 9:39 PM
I'm pretty sure I am on Leif's side... c'mon guys! What is wrong with his methodology?!? Sounds to me as though he standardized pretty well... I am curious to know what you think he has done wrong NOW...
Additionally, I am pretty sure he didn't have to go through all of this TWICE. I got what he was saying from the beginning. I am curious to know what you feel the real answer is as well.
The Sony, in my opinion is the poser in this scenario. From what I have seen the Sony stuff is NOT rated at an RMS output, so the specified 300 watts is a bit (OK, a LOT) misleading, and the 160 is probably the real deal as far as RMS output. As far as the Alpine stuff is concerned, I think those numbers are probably pretty close to what they can really do. They are close to the specified ratings of the amplifiers, so I would say "Take those as the real numbers" By the time you were to actually have a 4 ohm load on the outputs, you would see the numbers (as far as actual power output) go up a little bit, bringing you closer to the specified numbers. The peak numbers at clipping may sag a little bit from your 5.3 ohm levels, but it shouldn't be much more than a few tenths of a volt.
If you were to figure it backwards, Leif, the 407 would be 25.3v at 4 ohms, and the 1507 would be 30v at 4 ohms. I think the numbers you came up with are pretty close to the mark.
Good job on the testing, BTW...
-------------
It all reminds me of something that Molière once said to Guy de Maupassant at a café in Vienna: "That's nice. You should write it down."
Posted By: stevdart
Date Posted: July 18, 2004 at 10:13 PM
haemphyst wrote:
If you were to figure it backwards, Leif, the 407 would be 25.3v at 4 ohms, and the 1507 would be 30v at 4 ohms.
Ohm's law with these numbers gives the Alpines exactly what the published ratings are. 25.3*25.3 /4=160 30*30 /4=225 Considering they published the ratings using a 14.4V source, I would say their ratings are as accurate as can be, even somewhat underrated. Good follow-through, Leif. ------------- Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.
Posted By: stevdart
Date Posted: July 18, 2004 at 10:21 PM
OH, I see now how you came up with those numbers, hemp. You used the ratings to determine the voltage, I thought you were figuring a difference in voltage based on the different ohm load.... But seriously, Leif, you know that original guess of 150 was a shot in the dark, or 144 or whatever you came up with. ------------- Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.
Posted By: Leif
Date Posted: July 18, 2004 at 10:55 PM
Thanks, Stevdart!
Though.. Shot in the dark? No. Guesstimate? Hell yes. Since a doubling of power is only 3dB, even if i had been 33% inaccurate, that would have been just one dB.
Anyway, I realized that with the speakers I have (alphasonik 6.5" two way components, specced at 200w RMS).. I previously used them on the sony amp, the one that produces 2x160w RMS. That was plenty loud. With the woofers high passed around 70hz, I doubt they could really handle much more power than I was feeding it. (Not that 1000w peaks would kill them, but the cone won't move any further than the rubber surround allows no matter what you do)..
Perhaps I should just be happy with my front stage. It sounds damn fine, anyway. And now I know, I can choose between the three amps simply based on size and looks! :)
If I go for the smaller amp, I could use the extra real-estate for another amp for the tweeters (biamping) and that electronic crossover I never got around to using.
Anyone wanna buy an Alpine MRV-1507 or a Sony XM-7547?
:-D
///Leif
Posted By: flatulatta
Date Posted: July 19, 2004 at 12:07 AM
im curious to see if anyone has a problem with that???.... i wonder if someone has or could make a subscription site with all of these rating on it....maybe you could get sponsered by manufactuers in an attempt for them to get more money.. or it would make them create better amps to see that people are figured out the real specs
Posted By: greenlantern
Date Posted: July 19, 2004 at 8:19 AM
Why won't a sony, alpine or anyother amp rep reply? I think I willl email them this thread to look, and have a cnace to defend their claims.
Posted By: Leif
Date Posted: July 19, 2004 at 1:56 PM
I forgot to thank you for your reply, haemphyst, so thanks :-). Voltage is voltage whether you're looking at clipped kickdrum transients or a near-clipping sinewave - I'm glad there's other people here that understands this too :).
Greenlantern, I have a feeling they would just repeat some of Stevdarts arguments (i.e. the test wasn't done properly, our amp works fine in OUR lab) :).
///Leif
Posted By: stevdart
Date Posted: July 19, 2004 at 7:43 PM
My arguments were quite valid, as you well know. When a newbie makes a claim without having done his homework, and just relies on BS to get by, someone is going to say "WTF?" You didn't know some of the basics of audio, didn't know ohm's law, didn't know that a measurement of output voltage was necessary to find power, tried to bullsh*t me with some dumb*ss "12 volt peak to peak" crap, then somehow changed 12 volts into 24 volts so it would fit better into your guesswork, and tried to set yourself up as some kind of a genius who has made some brilliant discovery about the industry. "clipped kickdrum transients"...give me a break with your terminology. I don't know a whole lot, and I haven't installed any more systems than you have, but I realized a long time ago that a little knowledge is dangerous. With a lot more knowledge comes humility. Run your expensive components at 2 ohms, who gives a crap? But refrain from giving bad advice. And what's with the "Thanks, Stevdart"? I just said that it was good that you followed through. That's all. ------------- Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.
Posted By: dpaton
Date Posted: July 19, 2004 at 9:57 PM
ALtho I'm something of a noob here, I'm versed quite deeply in both analog electronics and test methodologies, as well as a tidbit I want to share about the recording industry. One thing I'd like to point out here (aside from the test methodology, which I'll gladly comment on later if desired), is that using prerecorded music for testing is a crapshoot at best. So much of the CDs out there now are clipped in mastering that you don't have a prayer of getting a clean waveform out of them when played back in your system. You'll see a clipped waveform at any level large enough to show you the nature of a kick transient. If you're serious about testing things, make your own CD of known good tones or buy a quality function generator. There's lots of good used Tek gear over on evilbay.
-dave
-------------
This is not a sig. This is a duck. Quack.
Posted By: stevdart
Date Posted: July 19, 2004 at 10:36 PM
And I differentiate you as a "newbie", not because that's a bad thing, but because if you had been around this forum for awhile you would have found that the people who take time from their lives here are just helping others with the problems they face with car audio installations, not taking the opportunity of widely-read bandwidth to elevate themselves. So you can get down off your pompous perch and start helping others with tried-and-true solutions to their problems. Or read and learn, as there is much to be learned from the posters on this site. Even for you, leaf.
-------------
Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.
Posted By: haemphyst
Date Posted: July 20, 2004 at 12:19 AM
Take it easy, stevdart. I was hearing nothing in Leif's tone that warranted that kind of response. He may be a "newb", but you were once a newb too. (Relatively speaking, I am a newb to this board, but I am NOT a newb to electronics or electronics theory.) I am fairly certain that nobody jumped on you like that if you made any mistakes. The man is here, like everybody, trying to learn and provide some input. If he is a newbie here, that does not mean he is a newbie to electronics in general. I was an electronics technician (6 years) AND a high voltage electrician (5 years) in the Navy, a communications technician for the County of Kern (California) (2 years) and now I design and build home theaters for people (3 1/2 years) I can vouch for his testing. The results he received were sound. The "kick-drum transients" that you were so quick to dismiss IS a real method of testing the dynamic (as opposed to continuous) power of an amplifier. Granted, it is usually done in a lab with a pulse generator, but how many of us have one of those? It is a small function called headroom. An amplifier (depending on the quality and "stiffness" of the power supply) can supply MUCH more power (rated in dB over the continuous output) for a transient attack than it can for a continuous tone. Simply because somebody does not know all of the phraseology, it is not necessary to dismiss him as completely incompetent. I have read MANY of your posts, and you are a smart guy... I have generally respected your point of view. You have posted some valuable input, and you are part of what makes this board as good as it is. I have read things from everybody here that I was not necessarily in complete agreement with, but I was not in a huge hurry to flame them in front of the whole community. I am not saying you are completely wrong, please believe that, but I am simply asking for a little bit of tolerance from you. You might read your own post regarding knowledge, and humility. The way I was reading your last post, it seems you are not necessarily willing to heed your own advice. I think you should realize that you do not know everything, either, and listen a little bit more. I also do not know everything, I am 37 years old, and I have been interested in electronics or have worked in electronics related industry since I was about 6 years old. I knew Ohm's Law by the time I went to second grade. I read EVERY book I could get my hands on regarding electronics. I was building passive series crossovers (are you familiar with those?) by the time I was a freshman in high school. I have BUILT, from piles of PARTS, flame speakers, ribbon drivers, Tesla coils, VanDeGraaf, Whimshurst, and Mark's Generators, Quarter shrinkers, and many other high voltage and audio components. I repair amplifiers (both digital and analog) PC power supplies, and component level repairs on computer motherboards. I am STILL aware I do not know everything. I learn something new everyday. Take a chill pill, man. Read and heed your own posts.
I am done with this thread. I will NOT reply to any response directed at me, but I will continue to post elsewhere on the board. I will continue to respect everyone here, so long as the respect is mutual. I will continue to offer advice, whenever I can, and I will continue enjoying doing so.
-------------
It all reminds me of something that Molière once said to Guy de Maupassant at a café in Vienna: "That's nice. You should write it down."
Posted By: Leif
Date Posted: July 20, 2004 at 12:44 AM
DPaton, ain't that the truth about CDs! It's been a pet peeve of mine for a long time now, and it really drives me crazy.
The "Alicia Keys" cut I used for testing IS indeed clipped on the CD. However, enabling bass boost on the discman is a very effective way to round out the waveform. Yes, the odd order harmonics will still be here, but there won't be any flat edges anymore, so for this type of testing, it works. I did take this into account when doing the testing.
Also, even if bass boost hadn't been on, CD players generally have some sort of high pass filter in the output amplifier.. The cutoff of this may be very low, but even if it's at 20hz, any squarewave (the clipped edge) on the CD is going to start tilting towards the center. (Only a completely DC straight signal path will be able to pass a squarewave without tilting the flat tops.)
Thus, looking at the scope and ramping the volume up, it's still easy to see where the amp clips and where the source material was clipped.
(I'd love to show you some screen shots of this, let me know if you're interested. Incidentally, there are Mix CDs that have evidence of this double clipping - they started with clipped source material, then ran it through a non-dc straight analog path, and then clipped it AGAIN to make the cd "loud enough to sell".. heh. Sounds lovely, I promise!)
Stevdart, this board labels me a Newbie based on my post count. Having been here long enough to earn the fancy title of "Silver Member", one would think that you would understand this, and not make other assumptions based on it.
Haemphyst, thank you. It wouldn't have been right for me to say what you did, but I'm very grateful that you said it.
///Leif
|