Using an oscilliscope to measure RMS output?

[OzzyGreg]

.
Guys,

I have recently been given an oscilliscope. I have no idea how to use it but I seem to recall someone telling me that an oscilliscope can be used to measure/calculate the output of an amplifier.

Can I use my oscilliscope to measure/calculate the power output (RMS) of my bass amps?

If so, how do I do it?

Any advice greatly appreciated.

Cheers,
Greg

[Bassman822]

1: Connect output of amplifier to a dummy load of the correct impeadance and with sufficent power handling capabilities
2: connent a sine wave generator to instrument input OR preferably power amp input
3: Connect o-scope input to dummy load and measure peak to peak voltage before clipping
4: Multiply p2p voltage by 0.3535
5: Square results
6: Divide by impeadance- that will be the rms output

[seamonkey]

Sure get a power resistor rated at what ever resistance you want to test, like 8 ohms then make sure it can handle the wattage of your amp.

8 Ohm 200W Non-Inductive Dummy Load Resistor

Input a sine wave at 1khz and crank. Measure V across the resistor. Watch for flattening of the peaks because that is clipping. Distortion is warping of the sine wave. It's harder to see. In both cases turn it down a little.

Measure the Peak to Peak voltage, and get the RMS
Vrms= Vpk/sqrt(2)

Then use Ohms law to figure out the power:
Ohm's Law Calculators
P = E^2 / R

Vrms is the "E" here. Square it and divide by R

And as some will point out "RMS" doesn't equate to power. There is RMS voltage and RMS current, so people take this to mean RMS power.

You need to measure this quick, because over time your amp will overheat.

Repeat at several frequencies, like at 100hz
You'll see a lot of power loss at lower frequencies where basses play.

Speakers are Reactive devices and vary in their impedance across the frequency range. You have use a resistor otherwise current and voltage are out of phase and power is hard to calculate.

[lbwdog]

Now.... you got that?...

[Rick Auricchio]

And after you do all that, throw all your results away. When you drive a speaker, its impedance varies widely across the frequency spectrum. Cabinet design and tuning also affects the impedance curve.

So your measurement with a dummy load is only data point, and it has little to do with what power is actually delivered to a speaker. And power doesn't tell you a thing about loudness (decibels), which is what our ears respond to...

[lbwdog]

Originally Posted by OzzyGreg

.
I have recently been given an oscilliscope. I have no idea how to use it
Any advice greatly appreciated.

Cheers,
Greg

ADVICE? Sell it!! You'll then get the maximum benefit.

[billfitzmaurice]

Originally Posted by Rick Auricchio

And after you do all that, throw all your results away. When you drive a speaker, its impedance varies widely across the frequency spectrum. Cabinet design and tuning also affects the impedance curve.

So your measurement with a dummy load is only data point, and it has little to do with what power is actually delivered to a speaker. And power doesn't tell you a thing about loudness (decibels), which is what our ears respond to...

Unless it's a tube amp a dummy load isn't necessary. Measure the amp at 100 Hz and you'll get a pretty good real world result. But since you can't alter the output of your amps what's the point, beyond idle curiosity?

[fdeck]

We wouldn't be human without idle curiosity.

I find it interesting to look at what happens to the waveform when the amp reaches its clipping threshold. For instance, does the amp exhibit "hard" or "soft" clipping?

[OzzyGreg]

.
Thanks for all the responses guys!!!

But since you can't alter the output of your amps what's the point, beyond idle curiosity?

Bill, that is precisely the reason I'm doing this!
I have an old Peavey MkIII head and a Yamaha B115 combo that I'm curious about.

Ok, all this is new to me so i'm going to have to do a lot of thinking before I move on it!

Greg

[aborgman]

Originally Posted by seamonkey

Repeat at several frequencies, like at 100hz
You'll see a lot of power loss at lower frequencies where basses play.

Not generally.

Most amps will put out rated power pretty much across the entire frequency range.

My 1965 B-18 has no problem putting out 50W at 30Hz, 1KHz, or 10 KHz.

[Mr. Foxen]

I definitely think the measured output opposed to manufacturer's claims in an interesting and worthwhile observation to be making.

[seamonkey]

Originally Posted by fdeck

We wouldn't be human without idle curiosity.

I find it interesting to look at what happens to the waveform when the amp reaches its clipping threshold. For instance, does the amp exhibit "hard" or "soft" clipping?

If it's a dual trace scope you can superimpose the original sine wave onto the screen to see what non-linearities the amp adds. And what the heck, try different wave forms. Saw tooth, Square Wave, ...

Hard clipping/soft clipping, even/odd odd harmonics. People should try these - I think everyone would be surprised at the folklore versus real world results.

[seamonkey]

Originally Posted by aborgman

Not generally.

Most amps will put out rated power pretty much across the entire frequency range.

My 1965 B-18 has no problem putting out 50W at 30Hz, 1KHz, or 10 KHz.

Still have your measurements? I'd be curious to see.

[craig.p]

Originally Posted by seamonkey

If it's a dual trace scope you can superimpose the original sine wave onto the screen to see what non-linearities the amp adds.

And you can get an instant reading on whether the amp is inverting or non-inverting, a crucial bit of information should you ever decide to run two non-identical amps in parallel. An inverting amp will show the output waveform as a mirror image of the input waveform, if you position one above the other.

On other matters, I disagree with the idea that dummy-load measurements aren't worth much. If you know what the amp can do into a purely resistive impedance, you have a baseline, AND you're likely to uncover "grossness" in an amp's behavior that you wouldn't have uncovered had you not bothered with the test in the first place. I also disagree with the idea of testing into an open circuit; in such a case, the power section is applying pressure but doing zero work. You are not likely to measure (or uncover) anything meaningful relative to real-world operation under such a scenario.

[gaffster787]

Originally Posted by seamonkey

Still have your measurements? I'd be curious to see.

What seamonkey is getting at...is..

The music signal's voltage level across the speaker rolls of at the low end, or else you could putting out DC...Like NASA says..."we have a major malfunction". A circuit that filters out DC is considered to be "AC coupled" or "capacitively coupled". A series cap or transformer are common methods.

So how fast does it roll off? The "half power" or -3dB frequencies tell you that at these "corner" frequencies the amp delivers half the power it does at mid band. Usually audio amps are designed to deliver at least half power down to 20Hz, so it is usually not a concern, but worth knowing before you buy, especially if you play a fiver..low b is around 30Hz.. Of course the speaker itself super-imposes its own frequency response and is generally the overriding factor.

[aborgman]

Originally Posted by seamonkey

Still have your measurements? I'd be curious to see.

Driving the power amp only through the ext amp jack with a sine wave I was able to reach 58.3V (P-P) into a non-inductive 8 ohm load. The amp flat topped at approximately the same voltage (within .5Vp-p) across the entire audio spectrum.

That is about 53W.

[aborgman]

Originally Posted by gaffster787

What seamonkey is getting at...is..

The music signal's voltage level across the speaker rolls of at the low end, or else you could putting out DC...Like NASA says..."we have a major malfunction". A circuit that filters out DC is considered to be "AC coupled" or "capacitively coupled". A series cap or transformer are common methods.

AC coupling has zero to do with low frequency roll-off.

Originally Posted by gaffster787

So how fast does it roll off? The "half power" or -3dB frequencies tell you that at these "corner" frequencies the amp delivers half the power it does at mid band. Usually audio amps are designed to deliver at least half power down to 20Hz, so it is usually not a concern, but worth knowing before you buy, especially if you play a fiver..low b is around 30Hz.. Of course the speaker itself super-imposes its own frequency response and is generally the overriding factor.

Any competent amp built in the last 50 years is pretty flat across the audio band. Undersized output transformers can lead to higher distortion at high power/low frequency.

[MAMMOTHvolume]

Originally Posted by aborgman

Not generally.

Most amps will put out rated power pretty much across the entire frequency range.

My 1965 B-18 has no problem putting out 50W at 30Hz, 1KHz, or 10 KHz.

at 30Hz? i'd expect a drop off at around 50Hz

[seamonkey]

Originally Posted by aborgman

Driving the power amp only through the ext amp jack with a sine wave I was able to reach 58.3V (P-P) into a non-inductive 8 ohm load. The amp flat topped at approximately the same voltage (within .5Vp-p) across the entire audio spectrum.

That is about 53W.

One of those numbers seems to typo?

58.3Vp-p / sqrt(2) = 41.22 V RMS

41.22Vrms ^ 2 /8Ω = 212W ?

Maybe it was a 32Ω load?

[dincz]

Originally Posted by seamonkey

One of those numbers seems to typo?

58.3Vp-p / sqrt(2) = 41.22 V RMS

41.22Vrms ^ 2 /8Ω = 212W ?

Maybe it was a 32Ω load?

58.3V p-p is 20.6V rms. That gives 53W.