Reason for me was: no experience and following amp specifications from positive reviewed amps on stereophile.
My Sonus Faber speakers can handle 22Vrms long term, but about 8Vpp on my scope is the max I can handle long term.
With the NIOSH app on my IPad, I measured today 109dB Max when pushing my speakers. For the amps this was a walk in the park.
https://www.diyaudio.com/forums/solid-state/96192-post-solid-pics-683.html#post6288826
Maybe a moderator should delete this thread.
Didn't think you would take is so literally.
I guess if one is not following the scientific method, one shouldn't post such things.
My only reason for posting the video was to illustrate that even with 87db speakers, things can get loud and that people shouldn't worry so much about power output.
I found the same at home.
You know, it's easy to build a couple dozen power amps of all different power outputs, but not so easy to build a proven, affordable, DIY 95db capable loudspeaker. I haven't seen many, not even a handful.
Didn't think you would take is so literally.
I guess if one is not following the scientific method, one shouldn't post such things.
My only reason for posting the video was to illustrate that even with 87db speakers, things can get loud and that people shouldn't worry so much about power output.
I found the same at home.
You know, it's easy to build a couple dozen power amps of all different power outputs, but not so easy to build a proven, affordable, DIY 95db capable loudspeaker. I haven't seen many, not even a handful.
I listen most times to a PC USB DAC that is bus powered.
Probably less than a watt.
Doesn't raise the roof but comfortable listening level.
Probably less than a watt.
Doesn't raise the roof but comfortable listening level.
That's quite unlikely. Peaks at +18dB happen once a month if that. It is a very safe margin for re-re-re-recording (and re-listening) operations.
In listening, 14dB headroom is rarely noticed unless the passage is repeated multiple times (as in post-production).
Totally agree that the DVM is NOT the tool to measure audio peaks. The caveman approach to the question is to just set-up a LM386 on 9V supply (pretty near 1 Watt max). I also love the 1980 Realistic SPL meter (and have one here by my chair) BUT I would not trust it in dynamic monitoring.
Attachments
Firstly yes it does demonstrate that 87db speakers can in fact play loud, but the overall premise of the video is that they are loud at 1W which is where the problem really lies. Anyone without understanding the science behind it may go away thinking you only need 1W of amplifier power to get sound that loud from 87db speakers. This is clearly not true.
Since when is it a bad thing that false information is called out. In a world full of fake news I think it is more important than ever.
A few points have been made, which should help anyone reading understand why this video doesn't demonstrate what is claimed.
1, The method of measuring the voltage being delivered to the speakers. Unless you feed a constant sine wave, that is within the frequency range that the multimeter is accurate, then you will get false readings.
2. Sound intensity drops as per the inverse square law. At twice the distance the sound will be 6db lower. Since the speakers are rated at 87db 1W 1M then at 2M the speakers should be able to produce 81db, the speakers are arguably even further from the Mic's than 2M so that should be even lower.
3. As pointed out the SPL meter was registering 96db peaks (and that still may have been inaccurate) so assuming the speakers were 2M from the mics (probably further in reality) they are producing around 11db more than they should be at 1W ) a 10db increase requires a 10X increase in power so at a minimum those peaks that registered would need to have been 10W).
If you did the same test with a 1W amplifier and avoided clipping, yeah it might be reasonably loud, but it will not be anywhere near as loud as what the video portrays.
Something a lot of people don't realize is that to be twice as loud (human perception) as 1W you need 10W, and to be twice as loud as 10W you need 100W of power. So yes 1W gets you a long way, but the headroom needed for recordings with large dynamic range is something that also needs to be taken into consideration, especially when dealing with lower efficiency speakers.
Tony.
I have 98dB speakers and do not listen to more than 1/2 Watt peaks.
My amp is unity gain.
The source is a CD player 2Vrms output
The speakers never see more than 0.5W on peaks.
My son says, "dad can you turn it down I can hear the music outside the house" on his way home from school.
So I guess you could say an 87dB speaker + cd player as source multiplied by 11dB of gain (3,5) is 2 x 3.5 = 7Vrms peaks is probably adequate for an 87dB speaker
49/8 is approximately 6 Watts at clipping, let's call it 10W to prevent clipping.
If you have a huge room or having a celebration and cranking it you'd probably want 8 times that - shortly followed by the police showing up, hahaha. Power requirements increase dramatically (Voltage squared), but by the same token they also reduce dramatically depending on the application.
I honestly believe people over estimate the power they think they need. I was also in this camp till I started playing around with unity gain amps.
My amp is unity gain.
The source is a CD player 2Vrms output
The speakers never see more than 0.5W on peaks.
My son says, "dad can you turn it down I can hear the music outside the house" on his way home from school.
So I guess you could say an 87dB speaker + cd player as source multiplied by 11dB of gain (3,5) is 2 x 3.5 = 7Vrms peaks is probably adequate for an 87dB speaker
49/8 is approximately 6 Watts at clipping, let's call it 10W to prevent clipping.
If you have a huge room or having a celebration and cranking it you'd probably want 8 times that - shortly followed by the police showing up, hahaha. Power requirements increase dramatically (Voltage squared), but by the same token they also reduce dramatically depending on the application.
I honestly believe people over estimate the power they think they need. I was also in this camp till I started playing around with unity gain amps.
Last edited:
Since I built a 2 x 4W amplifier and found out I never turn it up the (obvious) bonus I discovered is that I can build good sounding amplifiers at moderate cost and small size. As it uses a TDA7297 it is bridged and uses a single PSU with ideal rectifier and a single secondary transformer. The transformer that was planned was a 12V type but it turned out to be too high for the casing so I used a 6V 4A Rcore transformer that did fit just to avoid the "failed project" status. The transformer puts out 7.xV at all times and the losses of the ideal recifier are negligable. So I expected nothing (the amplifier driving 88 dB 4 Ohm speakers) but I really enjoy the sound. I haven't measured but I estimate that I don't go over 1W as volume is never above 11 o'clock with 2Veff sources. Contrary to some other amplifiers no listening fatigue at all.
Stupid thing is that almost all sources have transformers with higher voltages 🙂
Stupid thing is that almost all sources have transformers with higher voltages 🙂
Last edited:
^ It is implied. He says this is how loud 1W is (when it was clearly much more than 1W) He probably believed it was only one watt.
Tony.
Tony.
Actually an average level of 18dB below peak is quite standard for mastering. Millions of CDs and LPs were mastered and cut at that level. These days it's hotter, 16dB or 14dB below peak is now common, even more squashed because of the loudness wars.
But for decades 18 below peak as widespread. I also have many classical, big band and even some rock that was mastered at 20-22dB below peak. And peaks happen frequently on every track (except some classical). I can provide documentation, if you like.
If you look at the poll in the thread found in my sig line, you will see that the majority of people need 5 volts or lower for a very high SPL. 1 Watt average can be loud indeed.
Guideline from Mr. Pass article with 1Watt window show 🙂
source : http://www.firstwatt.com/pdf/art_cas_2015.pdf
source : http://www.firstwatt.com/pdf/art_cas_2015.pdf
Attachments
That graph shows 4V full scale, which is the peak voltage that an amp rated at 1 watt into 8 ohms should be able to do. Thus the 1 watt window.
Under anechoic conditions, of course. In real rooms, not true, depending on size and absorbtion.
That's right. In many rooms the sound bounces around and comes back again many times. Think of a small room where the walls, floor, ceiling are all polished concrete. Your speakers are going to sound a lot louder in there than they will sitting out in the middle of clover field. 🙂