How to tell if amp is sufficient?

[mt2]

Hello all, I recently scored a great deal on some Revel 105s for a bedroom (20x10x10ish) and being completely new to HiFi decided to pair with the most convenient thing I could find on Amazon, the WiiM Amp. Revel recommends 50-120 Watts (per speaker or per pair?) and the WiiM Amp on paper meets this, if at the lower end. I have an initial setup, but I'm not convinced I've developed the right ear to tell if it's working to its potential. So, this is a question more of technique than individual product recommendations.

How do you go about testing an amp/speaker combo, what should you listen for to know if an amp is driving the speakers correctly? Do you all have any go-to recommendations for samples to listen to to achieve this? What differences would a more expensive amp bring to the table for sound quality? Looking to educate myself out of potential buyers remorse I understand at one level, just enjoying the sound is all that matters, but I don't have enough experience to say what I like! How did you develop a taste for it?

 

[NTK]

Welcome to ASR!

The usual recommendation is to use this peak SPL (sound pressure level) calculator to estimate the volume at the listening position. Below shows my example results. Amir measured the Revel M105 (link) and it has a sensitivity of 85 dBSPL @ 2.83 Vrms @ 1 m (= 2 W @ 4 Ω, which is about the M105's nominal impedance). The Wiim Amp's output is 122 W @ 4 Ω (review). Since the calculator uses sensitivity at 1 W, the max output of the amplifier is reduced by 50% to 61 W to match measured sensitivity using 2 W.

My assumed listening distance is 12 ft, with typical conditions (2 speakers playing, placed not too far from the wall). The estimated peak SPL at the listening position is ~98 dB. That usually means that the amplifier is sufficient for moderately loud listening, but not enough for very loud (usually means the THX reference level of 105 dB).

Peak SPL Calculator

[Edit] The general rule of thumb is that the peak level is about 20 dB above the average listening volume (i.e. the amplifier/speakers should have 20 dB headroom). In the above case the amp should be sufficient for an average listening volume of 78 dB SPL.

 

[pogo]

The speaker sensitivity is not determined over the entire frequency range. The calculator also does not take into account the min. impedance values in the lower frequency range. This calculator is better suited for this purpose: Link

If tone controls and/or a DSP are used, the calculation looks completely different again!

 

[Blumlein 88]

I thought you just purchase an amp at least double the max rating for the speaker and don't worry about it.

 

[Sal1950]

The speaker sensitivity is not determined over the entire frequency range. The calculator also does not take into account the min. impedance values in the lower frequency range. This calculator is better suited for this purpose: Link

If tone controls and/or a DSP are used, the calculation looks completely different again!

Of course your right, but you have to start someplace and that calculator is just as good as anyplace to start.
The if, and's & butts can drive you to madness. LOL

 

[Holmz]

Hello all, I recently scored a great deal on some Revel 105s for a bedroom (20x10x10ish) and being completely new to HiFi decided to pair with the most convenient thing I could find on Amazon, the WiiM Amp. Revel recommends 50-120 Watts (per speaker or per pair?) and the WiiM Amp on paper meets this, if at the lower end. I have an initial setup, but I'm not convinced I've developed the right ear to tell if it's working to its potential. So, this is a question more of technique than individual product recommendations.

How do you go about testing an amp/speaker combo, what should you listen for to know if an amp is driving the speakers correctly? Do you all have any go-to recommendations for samples to listen to to achieve this? What differences would a more expensive amp bring to the table for sound quality? Looking to educate myself out of potential buyers remorse I understand at one level, just enjoying the sound is all that matters, but I don't have enough experience to say what I like! How did you develop a taste for it?

Features are also worthwhile.
The problem many people have, is that they have older gear that is still good, adequate and quite usable,
And that makes a basis for cobbling together a decent system.
For instance, I have an older preamp that my daughter is getting - and it is stunningly good, but has been replaced.

IMO I’d consider at least looking at something like a Lyngdorf TDAI 1200 or 2400 (0r something like that).
They do all the streaming stuff, they also have TV inputs, and built in DSP to correct for the room and comes with a microphone to push the auto correction gear to set up the thing.
And then you don’t have multiple boxes, and better Feng Shui if it is stylish Batchelor pad... or trying to be…

 

[J.M. Noble]

Get an inexpensive true RMS multimeter with a peak hold function and play music while measuring. You'll learn a heck of a lot and you'll also have your answer.

 

[pogo]

Which inexpensive true RMS multimeter with peak hold function offers reasonably accurate measurement results?
For music signals the multimeter needs to sample at >2*f!
This could also be sufficient for an approximation of the peak power: f is the highest frequency with lowest impedance

 

[DanielT]

Just keep in mind that if you're going to perform EQ and let's say it leads to lowering the sensitivity of your system, the speakers by 3dB in order to achieve a more pleasing sound character. This means that you need twice as much amplifier power.
In addition, you then need speakers that provide good power handling. They need, as a result of your EQ, to work harder and you don't want them to distort annoyingly because of it.

Subbass, the lowest frequencies need much more power than the highest frequencies. You can see that by how much amp power active subwoofers can have, 500 watts or more is not unusual. Especially with small sealed subwoofer boxes. With more cavity volume, however, less amp power is needed for subwoofers.
A tweeter, on the other hand, can easily only need 20-30 watts. Incidentally, you can see that in its power handling. An example this which is popular:

ND25FW-4

Power Handling (RMS) 20 watts
Power Handling (max) 40 watts

Dayton Audio - ND25FW-4 1" Soft Dome Neodymium Tweeter with Waveguide 4 Ohm

ND25FW-4 1" Soft Dome Neodymium Tweeter with Waveguide 4 Ohm

www.daytonaudio.com

Listening distance and amp power needs are usually based on calculations in free space. A normal listening room functions as a reverb chamber, so listening distance is in most cases nothing that you need to consider. As long as we are not talking about extreme cases like listen a few inches from the speakers vs ... or a listening room that is so treated that it is like an anechoic chamber. Very extreme cases.

 

[J.M. Noble]

Which inexpensive true RMS multimeter with peak hold function offers reasonably accurate measurement results?
For music signals the multimeter needs to sample at >2*f!
This could also be sufficient for an approximation of the peak power: f is the highest frequency with lowest impedance

You're overthinking this. (Disclaimer: I am an electronics engineer and I have a professional background in loudspeaker design.)

Unless you're blasting cymbal concertos or piccolo sonatas at max volume, the main frequencies of interest are in the range of a couple of hundred Hertz and down. I own a four channel 100MHz oscilloscope, but I just use the meter (a Fluke 87, but still just a meter) for this application and it's good enough for the purpose.

Do a little Ohm's Law to get the amp's rail voltages from the rated continuous output power into a specified impedance. See how often you're bouncing off those rails, if at all. Ideally you'd also measure SPL at the same time. The phone apps can be surprisingly accurate. I own a SPL meter and I also have a couple of calibrated mics, and both my iPhone and my Apple Watch are pretty close--certainly good enough for this application.

With this data in hand, you'll be able to make accurate predictions about a lot of things in your system and your room.

 

[pogo]

I am an electronics engineer and I have a professional background in loudspeaker design.

Then please can you tell me from your practical experience how high the error is due to the back-EMF (addition, subtraction, no effect, ...)?

 

[Chrispy]

Do you need the full potential in your bedroom? What levels will you be listening at? How far away are you? Will you add a sub?

As-is do you feel that you're getting audible distortion at your listening levels? What's missing/wrong as-is? As was said, don't overthink it. Spending more may get you more power or ability to handle lower impedances better, or better box aesthetic or features or connectivity....but that Wiim Amp seems pretty nice for a bedroom setup I'd think.

 

[antcollinet]

For music signals the multimeter needs to sample at >2*f!

Not really with the spectral power density of music. The vast majority of the power is (around) 3KHz and below. This is conveniently also where you typically get the lowest impedance dips.

 

[pogo]

Hence my simplified assumption:

This could also be sufficient for an approximation of the peak power: f is the highest frequency with lowest impedance

 

[antcollinet]

Hence my simplified assumption:

Yep - valid point (I initially read your post assuming f up to 20kHz) , and it will probably be valid in most circumstances. Perhaps not when the minimum impedance is higher than 3kHz (probably even 1kHz) - in which case it probably doesn't matter, since the music power will be 30dB or more down.

 

[J.M. Noble]

Then please can you tell me from your practical experience how high the error is due to the back-EMF (addition, subtraction, no effect, ...)?

You're still overthinking this. The goal is to ensure a reasonable reserve, not explore the very limits of performance. If you're off by 20% (i.e. 1dB in power terms) it's more than good enough. 20% would be a gross measuring error but inconsequential to the listener.

You're trying to do science, and that's fine, but I'd suggest employing practical engineering practices instead. Your room's acoustic treatment surely requires more attention than this issue.