Oh, please, not !
Vu meters (who use them, nowadays ?) had an inertia. To compensate this inertia, and give some coherency between what you see on them and what you hear, with musical signals (transients), it was recommended to give *them* ~8db of supplementary gain. We don't have to measure continuous *levels* with them. But not with the *peak meters (5ms).
According to DYN, the standard consumer(dBu) input level is 0.775mv eff, asymmetrical, without load. And the professional level was +4dBu: 1.228 volts eff under 600 Ohms of load, symmetrical.
Since the digital, we tend to use dBV: 1Veff, whatever the load.
In order to do not make the things too simple for the pro audio industry, the odb of the digital crest factor meters (LED = no inertia), is set between -14 to -18dB from the clipping level. This is where you put your reference tone, depending of the brand of the equipment manufacturer, or the requisite of the client (TV as an example).
And, because some technical managers are not very intelligent, or have enough comprehension of the technical issues, some TV channels measure the crest of your musical content at this -14/18 dB and do not allow any overtaking: we have lost those 14/18dB of dynamic for...nothing.
Of course, CD and other digital media producers want their products to sound as loud as possible: thus they normalise the whole levels of the musical content of the final "master tape" in order to set the max higher peak at 100% digital level, and all the previous has no more sense anymore...
Last edited:
Ultra high end should really be focusing on bringing a bespoke solution from start to finish, so the gain structure should be known throughout.
As it stands, most consumer gear has way, way too much gain with all kinds of negative consequences. Realistically speaking, anyone with a good DAC should be wanting their system to be pretty much at it's sweet spot some 12 dB down from its max output (where SINAD usually hits its peak). I want no other volume control in my signal chain.
Obviously the knuckle draggers who still use analog sources will have to figure out their own solutions, (I'm kidding, of course
) but there's the universal adage of getting one's gain as early in the chain as possible.
As it stands, most consumer gear has way, way too much gain with all kinds of negative consequences. Realistically speaking, anyone with a good DAC should be wanting their system to be pretty much at it's sweet spot some 12 dB down from its max output (where SINAD usually hits its peak). I want no other volume control in my signal chain.
Obviously the knuckle draggers who still use analog sources will have to figure out their own solutions, (I'm kidding, of course
) but there's the universal adage of getting one's gain as early in the chain as possible.
During the good old days of analog, the tape had no absolute level limitation, but a progressive increase of distortion with levels. So, it was always possible to push things a little harder ...
With digital, the content is (let's say) undistorted up to the max level, and the signal/noise level increase with levels ...then it clips at this max level.
This imply a different way to think.
With digital, the content is (let's say) undistorted up to the max level, and the signal/noise level increase with levels ...then it clips at this max level.
This imply a different way to think.
Last edited:
Ultra high end should really be focusing on bringing a bespoke solution from start to finish, so the gain structure should be known throughout.
As it stands, most consumer gear has way, way too much gain with all kinds of negative consequences. Realistically speaking, anyone with a good DAC should be wanting their system to be pretty much at it's sweet spot some 12 dB down from its max output (where SINAD usually hits its peak). I want no other volume control in my signal chain.
Obviously the knuckle draggers who still use analog sources will have to figure out their own solutions, (I'm kidding, of course
Yeah, a lot of the latest current output DAC ICs have pretty hot output if you use a reasonable I/V resistor value. The Benchmark DAC3's default balanced output is +24 dBu! They do offer attenuation, though.
The SE out is high-ish too at 3vrms (11.7dbu), but has a "theater" mode at 2vrms and 8.2dbu. This is more inline with the unspoken spec others seem to be going with. I wouldn't expect Benchmark to hit clipping really ever, and probably have these figured well, but that's only at 3.7vrms and 13.5dbu.
Last edited:
I believe the Audioquest Niagara 7000 has a tank circuit in it that filters like this. The designer Garth Powell has a video on YouTube where he explains what is in the 7000. The MSRP of the Niagara 7000 is $8,000. Not cheap.
I don't imagine laser diodes and detectors are not that expensive, nor an interferometry circuit with enough bandwidth for a woofer. Surely you don't need a laboratory grade system for this.
if you just Google on Cheap laser interferometers.... you get a lot of DIY systems for cheap. Too much for AD but 25$ or less for one home made is OK IMO. From there it can be less as refined for the app. You could use the guts from a cheap laser pointer.
Not exactly usable for our app but has ideas in it ---->
Self Built Interferometer Measures Nanometer Displacement | Hackaday
THx-RNMarsh
Not exactly usable for our app but has ideas in it ---->
Self Built Interferometer Measures Nanometer Displacement | Hackaday
THx-RNMarsh
Last edited:
Hi Tournesol,
Consumer 0 levels are -10 dBu then, it's still 0.316 V with a continuous tone. Now to give you something else to dispute, the Dolby reference level is 580 mV, continuous tone.
The ballistic characteristics of a meter has zero to do with this, just to be clear on this. We are talking a continuous tone for calibration, so the meter reading will be steady state.
I wouldn't have gone very far in audio by using your higher levels for zero. Just think of all the calibrations I've done (thousands) that have met with approval from distributors and manufacturers over the years! I calibrated a lot of both professional and consumer products. Your +4 dBm reference for pro gear is correct. Maybe your level for consumer is normal and accepted for your country, but in North America it's 316 mV. Even Hewlett Packard thinks so.
-Chris
Consumer 0 levels are -10 dBu then, it's still 0.316 V with a continuous tone. Now to give you something else to dispute, the Dolby reference level is 580 mV, continuous tone.
The ballistic characteristics of a meter has zero to do with this, just to be clear on this. We are talking a continuous tone for calibration, so the meter reading will be steady state.
I wouldn't have gone very far in audio by using your higher levels for zero. Just think of all the calibrations I've done (thousands) that have met with approval from distributors and manufacturers over the years! I calibrated a lot of both professional and consumer products. Your +4 dBm reference for pro gear is correct. Maybe your level for consumer is normal and accepted for your country, but in North America it's 316 mV. Even Hewlett Packard thinks so.
-Chris
Yes, but this was to go in computer speakers cringing starts at $1. Counting and keeping track of fringes with interferometers can be a problem if you want a no fail system. Remember any vibration of the apparatus is a signal, most lab applications have the instrument on a VERY solid base not in the box with a woofer putting out high SPL.
DIY fiber optic sensors might be a possibility:
Fiber Optic Sensors for Displacement & Ultrasonic Vibration Measurements
Regards,
Braca
Fiber Optic Sensors for Displacement & Ultrasonic Vibration Measurements
Regards,
Braca
Seems like some kind of optical system might work. Reflect some light, block some light, something like that. Maybe use a modulated beam and synchronous detection to reject ambient light. Maybe consider measuring from the back side, through holes in the frame? Or, could be at the front side.
Last edited:
Now you are confusing me, I thought this was for $1000 type subwoofers. The entire BOM for all the mirrors, lasers, prisms, beam-splinters, actuators etc. in a Blu-Ray player is <$2. The accelerometer solution works and the problems have been solved and there are cheap IC versions now.
Last edited:
Apologies on (perhaps) misrepresenting your position. But, yes, if the goal is to lower distortion materially, then one's going to need decent fixturing to get a reliable measurement in what is definitely a high-vibration environment.
My understanding is the present discussion is more interested in voice coil position, but as a proxy, looking at the coil/cone interface would probably be the best place to measure.
- Status
- Not open for further replies.