how to calibrate a VU meter at preamp output ?

[andre tchmil]

Hey Guys , simple question.
I have a preamp and a buffered VU meter circuit at the output.
How do I calibrate the VU meter for todays digital levels ?
opinions please .

 

[PRR]

> VU meter at preamp output

It does not matter where it is.

Like a weight scale. It is calibrated the same whether in a store, in a house, or out on the street at night weighing drugs.

The VU meter was a semi-portable measuring tool to be used ANYWHERE along an audio line. When radio networks expanded over telephone lines, the levels might be fine in Chicago, weak in Denver, hot in Sacramento. Or at least that is what the local engineers thought. With a VU meter they could monitor the same live program (or tone) the SAME way at every point and report PRECISE levels (+/- several dB) to telco engineers.

> calibrate the VU meter for todays digital levels ?

Turn it up until it dances good at your usual levels. The VU meter is BAD for even guessing at peak levels, which is all that matters with >12-bit digital systems. VU meters are now just decorations.

 

[pvision]

> VU meter at preamp outputThe VU meter is BAD for even guessing at peak levels, which is all that matters with >12-bit digital systems. VU meters are now just decorations.

The old saying is that VU stands for Virtually Useless

I remember an engineer calling the maintenance guys in at a studio I worked at because the tambourine was distorting on tape. It was hitting about -3 on the Studer A80 VU

No PPMs in the studio so it took a while to talk him down

Nick Froome

 

[emrr]

If you want to be fancy, get a vintage Daven VU attenuator and dial it in to the sweet spot.  They were still cheap the last I looked, no one has really wanted them so far.  Every proper VU panel had one.  Or just build the right VU pad for your expected levels. 

 

[JohnRoberts]

Hey Guys , simple question.
I have a preamp and a buffered VU meter circuit at the output.
How do I calibrate the VU meter for todays digital levels ?
opinions please .

The problem with asking a simple question on the internet is that you don't always get simple answers.

The simple answer for an analog VU meter is to calibrate it to 0VU=+4dBu=1.23V. Level is level.

The complicated answer is "it depends". Meters in digital environments generally measure levels below digital full scale, or below clipping. Digital full scale does not map out to a fixed or standard analog level. Good practice is to calibrate analog 0VU to digital  -18dBFS or -20dBFS.   

Good Luck.

JR

 

[envelope]

Inject a 1kHz test tone into the preamp, then adjust your preamp gain so that you can measure 1.23V across pin 2 and 3 of your output XLR with a DMM set to AC volts , then adjust your buffered circuits' trimpot to position the meters 'pointer 'at your discretion.

with RMS meters, I sometimes adjust the pointer to about -6 on the Vu scale while measuring 1.23V on the output XLR when calibrating with a 1kHz sine-wave, so that with nominal +4dBu RMS music program material, the pointer ends up swinging in the 0Vu +2Vu range.

Or you can run some pre-mixed, commercially produced mega-normalized modern mixes out of your soundcard  directly into the Vu buffer circuit, bypassing your PreAmp altogether. Then adjust the buffer circuit so the pointer is deflecting around 0Vu or a little above, or wherever you want it to deflect.

Then when you reconnect the Vu buffer to your Preamp, it will be ('semi-calibrated already) dig-it?

This is just one 'self-help' method, and will open to criticism of course.

 

[envelope]

sounds like the engineer is not an engineer, at least not a capital "E" P.Eng. which a lot different that a small ' e' engineer with a diploma or certificate in recording arts. Proper Vu  readings are still critical if that is the only reference...  If used properly, they are just as meaningful as any PPM.  One can learn to translate where an RMS Vu will deflect at any given frequency, transient content or amplitude.
It takes proper training to be able to translate RMS Vu readings.  It is a lost art and has thus become trivialized...sorta like calibrating analogue tape machines....it is critical but often ignored in modern times.

> VU meter at preamp outputThe VU meter is BAD for even guessing at peak levels, which is all that matters with >12-bit digital systems. VU meters are now just decorations.

The old saying is that VU stands for Virtually Useless

I remember an engineer calling the maintenance guys in at a studio I worked at because the tambourine was distorting on tape. It was hitting about -3 on the Studer A80 VU

No PPMs in the studio so it took a while to talk him down

Nick Froome

 

[pucho812]

Hey Guys , simple question.
I have a preamp and a buffered VU meter circuit at the output.
How do I calibrate the VU meter for todays digital levels ?
opinions please .

0VU= +4dBu=1.228/1.23 V.
On the digital side your converter  and daw will show metering in  dBFS.  dBFS is peak metering.  VU is average.  0 dBFS is the maximum level before reaching digital distortion on the converter.  Some converters allow you to trim the level some are fixed. It will say in the manual where  0VU = in the  dBFS side. Personally I prefer 0VU = -16 dBFS It's what was the calibration on the first digital recording stuff I ever did. 

 

[JohnRoberts]

sounds like the engineer is not an engineer, at least not a capital "E" P.Eng. which a lot different that a small ' e' engineer with a diploma or certificate in recording arts. Proper Vu  readings are still critical if that is the only reference...  If used properly, they are just as meaningful as any PPM.  One can learn to translate where an RMS Vu will deflect at any given frequency, transient content or amplitude.
It takes proper training to be able to translate RMS Vu readings.  It is a lost art and has thus become trivialized...sorta like calibrating analogue tape machines....it is critical but often ignored in modern times.

> VU meter at preamp outputThe VU meter is BAD for even guessing at peak levels, which is all that matters with >12-bit digital systems. VU meters are now just decorations.

While 12 bit digital systems are a bit of a straw man argument, I am also of the opinion that old school VU meters are mainly eye candy and not very useful.

Why use a meter that you have to learn the different windage for different instruments to avoid saturating the path. Modern peak meters present that information in a simple easy to interpret manner.

A few decades ago, i pioneered combining peak and VU information together on one display so you get the full story (and lots of blinky lights).

JR

 

[Gold]

0VU= +4dBu=1.228/1.23 V.
On the digital side your converter  and daw will show metering in  dBFS.  dBFS is peak metering.  VU is average.  0 dBFS is the maximum level before reaching digital distortion on the converter.  Some converters allow you to trim the level some are fixed. It will say in the manual where  0VU = in the  dBFS side. Personally I prefer 0VU = -16 dBFS It's what was the calibration on the first digital recording stuff I ever did.

That's a great answer. I also agree that -16dBfs is a good compromise between having headroom and dealing with hot digital levels. Digital unlike analog sounds the same at all levels right up to 0dBfs. Audio that has been mastered in the modern way will sit at a much higher level than  you would use for tracking an instrument.

When vinyl was the release format a recording engineer could record and sequence an album at a target level. If the record sides were relatively short the recording engineer could say to the cutting engineer "cut at 0" and it would mean something. Calibrate the tones on the tape to reference level (0VU) and cut the lacquer disk with no level offset. No such relationship exists between analog and digital levels.

 

[envelope]

While 12 bit digital systems are a bit of a straw man argument, I am also of the opinion that old school VU meters are mainly eye candy and not very useful.

Why use a meter that you have to learn the different windage for different instruments to avoid saturating the path. Modern peak meters present that information in a simple easy to interpret manner.

A few decades ago, i pioneered combining peak and VU information together on one display so you get the full story (and lots of blinky lights).

JR

agreed re: peak vs. Vu, But I thought the member is just asking for actual instructions to perform a a task that he wants to perform, as opposed to  a philosophical discussion on the merits of metering. Sure, RMS Vu meters can be decorative in modern times, but not to everyone. Which ever way someone has been properly trained to monitor levels in the correct way. If someone just wants to see a meter move, why bother with Vu, as analogue PPM meters are far more reactive and fun to watch deflect.
   
(the new Vu buffer/driver with PEAK indicator that JLM produces is great tool for those who are not really familiar with how to read RMS deflections.) 

 

[JohnRoberts]

agreed re: peak vs. Vu, But I thought the member is just asking for actual instructions to perform a a task that he wants to perform, as opposed to  a philosophical discussion on the merits of metering. Sure, RMS Vu meters can be decorative in modern times, but not to everyone. Which ever way someone has been properly trained to monitor levels in the correct way. If someone just wants to see a meter move, why bother with Vu, as analogue PPM meters are far more reactive and fun to watch deflect.
   
(the new Vu buffer/driver with PEAK indicator that JLM produces is great tool for those who are not really familiar with how to read RMS deflections.)

The OP's question had already been answered  (I think) before others added the philosophical veer. Since I spent decades designing meters and received two patents related to peak/VU meters I have strong opinions and have heard all the arguments pro and con.

Since the customer is always right. I also made consoles with mechanical VU meters when that was what the customer wanted (you never argue with a customer who wants to give you money.).  Whether they wanted or not, I also added a peak LED per VU meter to help fill in the missing peak information. 

JR

 

[envelope]

okay, thanks for clarifying John....

To the member who said  that you cannot even guess at the Peak when using a VU RMS meter. You can indeed guess,  and you can estimate quite accurately, if you have been trained to read RMS and PEAK readings. If you run a kick drum through a RMS and PEAK meter at the same time you can see how each deflects. this relationship between the RMS and PEAK reading is a constant,  so if you then remove the peak meter, and only have a Vu you can estimate the Peak value based on the RMS reading when the kick drum 'hits'. It takes a great deal of training, and is a lost art.

 

[abbey road d enfer]

0VU= +4dBu=1.228/1.23 V.
On the digital side your converter  and daw will show metering in  dBFS.  dBFS is peak metering.  VU is average.  0 dBFS is the maximum level before reaching digital distortion on the converter.  Some converters allow you to trim the level some are fixed. It will say in the manual where  0VU = in the  dBFS side. Personally I prefer 0VU = -16 dBFS It's what was the calibration on the first digital recording stuff I ever did.

That's a great answer. I also agree that -16dBfs is a good compromise between having headroom and dealing with hot digital levels.

Please allow me to disagree. As much as I agree with the fact that recording at about -16dBfs is a safe operating area with modern systems that have about 20-bit effective resolution, I don't think a VU-meter calibrated to indicate 0VU for -16dBfs is THE answer. As many have pointed out, the ballistics of VU meters do not allow useful indications, except on near-sinewave signals. The "lost art" that allowed " One... learn to translate where an RMS Vu will deflect at any given frequency, transient content or amplitude" is no help in a digital world. However trained /experienced the operator, there would always be the same uncertainty that would lead to over or under estimating, with the same resulting risks of overload. The only proper answer is the use of an adequate meter.

When vinyl was the release format a recording engineer could record and sequence an album at a target level. If the record sides were relatively short the recording engineer could say to the cutting engineer "cut at 0" and it would mean something. Calibrate the tones on the tape to reference level (0VU) and cut the lacquer disk with no level offset.

I've never met such a situation for any commercial release. For demos, yes, but not for a full-format 33rpm.
The inadequacy of VU-meters in lacquer mastering was perfectly understood a long time ago; in 1973, when I first dealt with cutting lathes, there were PPM's and peak limiters. No VU meters.

 

[Gold]

I've never met such a situation for any commercial release. For demos, yes, but not for a full-format 33rpm.
The inadequacy of VU-meters in lacquer mastering was perfectly understood a long time ago; in 1973, when I first dealt with cutting lathes, there were PPM's and peak limiters. No VU meters.

I have a few engineers as clients who send in sequenced reels that are a 1:1 transfer. They are very good engineers but it is certainly possible. These records are released on labels.

The transient information you get from a PPM is not very useful for setting levels on vinyl disk. Program with a large crest factor could be cut at exactly the same RMS level as a peak limited version of the same. In other words the transient information doesn't alter the possible RMS level (except in cases at the margins). The unlimited version with transient information may cause increased distortion on playback which is the opposite case to digital. I like VU's and find them more useful disk cutting than PPM's, although I use both.

I have a Neumann SP66 cutting console from 1968 or so. It has giant VU meters, light beam PPM's (which are very touchy) and a correlation meter. Some of my distaste for PPM's may be from my first exposure to them as poorly calibrated light beam meters.

 

[andre tchmil]

all very useful explanations guys, thanks
We live to learn

 

[opacheco]

".................Personally I prefer 0VU = -16 dBFS It's what was the calibration on the first digital recording stuff I ever did. "
[/quote]

Sorry but what is the scientific fundament of this adjustment??.....What is the criteria in this selection??, is this arbitrary?

Opacheco.

 

[JohnRoberts]

".................Personally I prefer 0VU = -16 dBFS It's what was the calibration on the first digital recording stuff I ever did. "

Sorry but what is the scientific fundament of this adjustment??.....What is the criteria in this selection??, is this arbitrary?

Opacheco.

AFAIK this is arbitrary.  Back in the old school analog days a path would have a fixed 0VU using either a +4dBu or -10 dBV (voltage) standard , and headroom or clipping point would vary from product to product, with 20 dB or more of headroom considered good.  Operating in the digital domain clipping or saturation is fixed, not as a voltage but as digital full scale. In an ideal world we would aim for the same 20 dB of headroom, while the old school 20dB was above an average 0VU so likely less than 20 dB in peak metering terms.

-16dBFS sounds OK to me... Hotter might limit digital headroom and cooler could affect digital noise floor, but try not to lose much sleep over this.  Note this XdBFS is a relative internal level not a standard analog voltage. So you may need to align this digital 0VU to your analog 0VU (or not).

JR

 

[BYacey]

In digital recording, the clip indicator on the converter is king.
I've never found a VU meter very useful when recording to digital, except to perhaps compare signal levels  of similar wave envelopes.  I find it easier to increase the gain  until the clip light starts to flicker on the converter, and then back off the gain 15 to 20db.

Drums are the most problematic, due to the very fast transients that a mechanical VU meter ballistics have no hope of following.

 

[pvision]

Most of the studios I worked at used 0VU = +4 dBu and, if they had BBC-spec PPMs, 0VU = PPM 4

Tape machines would be lined up so that 0VU at the console gave 0VU on the machine and usually that would be the Dolby calibration level as well

According to the tape you used, and the level of tape compression you wanted, the outputs would be calibrated with a test tape so a 250 nWeber tape would read 0VU - or whatever level you chose - and the record side setup for unity gain

So, like digital systems now, there was a standard operating level within the console & outboard, but the recording medium level might alter with each project

Nick Froome