Analog meters

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

dogears

Well-known member
Joined
Nov 15, 2017
Messages
548
Location
Texas
I've never done a lot with analog metering, and I have a prototype going that I want to add meters to.

One question I have - if I want to grab some random  DC meter, not even one listed as "VU" -- are these typically slow? I see some spec sheets that say something like 3 second max response time.  I can add an external diode bridge and cap to adjust timing, but obviously that won't help if the meter itself has some kind of super-slow ballistic behavior. 

Does anyone know of a source for decent edgewise DC ammeters? Ive talked to Weschler and Hoyt but Hoyt wants a minimum 50 qty buy and Weschler couldn't answer my response time question...
 
Hi John,

Right, I'm not trying to hit the IEC standard. Just reasonable response time, useful visual indication.  There are a lot of industrial DC meters that the datasheet says something like 50% overshoot max, 2.5 sec response time max. That seems like it's pretty much useless considering a VU meter is suppose to have 1-1.5% rise time and 300 ms response.  But, I'm not sure if those datasheets are very cautious "max" or a "typical".

I'm looking for an edgewise meter, just having trouble understanding datasheets. They're all pretty limited in terms of information.
 
dogears said:
Hi John,

Right, I'm not trying to hit the IEC standard. Just reasonable response time, useful visual indication.  There are a lot of industrial DC meters that the datasheet says something like 50% overshoot max, 2.5 sec response time max. That seems like it's pretty much useless considering a VU meter is suppose to have 1-1.5% rise time and 300 ms response.  But, I'm not sure if those datasheets are very cautious "max" or a "typical".

I'm looking for an edgewise meter, just having trouble understanding datasheets. They're all pretty limited in terms of information.

I actually got two different patents covering simultaneous peak/VU metering. Even expensive  VU meters are average responding so of limited value (IMO). That said I have put 20+ real (expensive) VU meters in a big dog console because the customer is always right even when (you know).  ::)

I would not attempt to design a non standard meter just from electrical specs. Other than for a starting point.  Build a prototype maybe with a few different cheap meters, and look at them with audio wiggling the meter movements.

JR 
 
Yeah, I'm not trying to do anything nonstandard. At least, I don't think I am.

I'm trying to show gain reduction, which seem simple since my sidechain control voltage is linear DC to dB GR. Just need a resistor to scale the DC current meter to 0-20 dB GR ( control voltage constant * dB range = full scale). Since my control voltage is post-time constant control, ideally I'd think your meter would be instantaneous GR. So faster is better.  It seems like all of that is useless if the mechanical action of the meter itself is 2.5 or 3 seconds!

 
Just some general advice on this topic, based on my experience over the years:
1.  Real VU meters use a combination of mass added to the rotor (the part that sits in jewel bearings in a d'Arsonval type moving coil movement) and a controlled resistance driving the movement (about 3.6 k if I recall correctly and a small copper-oxide rectifier). I think it would be very difficult to recreate this with and ordinary meter movement and external R-C networks.
2. I can't imagine a basic meter movement taking 2.5 to 3 seconds to settle unless it has severe overshoot and swings back and forth a lot before it finally settles ...
3. The "ballistics" of any basic d'Arsonval type meter (current, not voltage) can be "damped" in the same way as a loudspeaker. It's why such meters arrive with a shorting clip across the terminals ... it keeps the needle from flailing about during shipment. You can virtually eliminate overshoot by driving such meters through a low-impedance resistor network. As with speakers, little is gained, damping-wise, by making the source resistance less than its own DC resistance (10-20% of the meter resistance will give all the damping possible). Driven from such a low-resistance source, most current meters (200 uA full-scale is common) will respond quickly and have very little or no overshoot ... which sounds ideal for indicating GR.

Bill Whitlock
 
Build a prototype maybe with a few different cheap meters,

Bit of an aside, and though not edgewise, I recently purchased some bulk, very cheap (koss) VU meters for aux bus, and I must admit they are exceeding my expectations. Super fast (too fast!) and sensitive. 

They are most definitely nonstandard, and Joe Malone advised me on a buffer to supply the extra current to they wanted.  But in terms of getting a read of the audio signal, they appear to indicate transients and low signal better than any of the larger, more expensive ones tested.  ???
 
I am repeating myself, but I do that a lot.

A mechanical meter will never be fast enough to catch and indicate brief transients unless the electronic front end has active hold circuitry to allow the mechanical meter to catch up to where the audio was.

The primary argument for "true VU" meters, besides the obvious eye candy appeal, was that some professionals cut their teeth on VU meters so were accustomed to interpreting the results and didn't want to change or deal with non-standard mechanical meters. There are less of them every day.  Back in the day, brief transient overloads were well tolerated by recording mediums like magnetic tape. Recording engineers from experience would apply windage or different offsets when tracking solo instruments using slow VU meters.

Fast forward to now with digital media that is less tolerant of overload (but you wouldn't know it from some releases). LED meters can be as fast as we dare, and faster than our eyes can follow so for intelligibility narrow transients need a little hold. 

Even back those decades ago when I gave customers their precious true VU meters, I added LED clip indicators along side each meter to provide some awareness of when their audio path was saturating, or about to.

For eye candy, I wouldn't worry about replicating true VU ballistics because IMO it is an archaic (and expensive) reminder of times past, with about zip utility for working in modern digital mediums.

I now return you to your ongoing discussion.

JR
 
Recording engineers from experience would apply windage or different offsets when tracking solo instruments using slow VU meters.
Interesting. I ass-umed a strict convention was in order, a 'language' if you will, that did not require translation when other people came to the studio.
 
A VU meter was never for absolute precise level indication. It’s a hybrid level and loudness meter. 

Level and loudness are two different things. Most people looking at meters want a loudness display.  A VU meter gets you partway there.

A PPM is strictly a level meter and is quite misleading as a loudness meter.  For most users a PPM will need more interpretation than a VU meter.
 
boji said:
Interesting. I ass-umed a strict convention was in order, a 'language' if you will, that did not require translation when other people came to the studio.
More like a learned windage for what works when tracking different solo instruments with different crest factors (peak to average ratio) meter differently. Percussive instruments (like drums) will under-report their peak level when feeding a slow average meter, so could clip sooner with lower average meter readings than expected.

Peak reading meters do not require learning any new language to use them, they just show actual clipping headroom for any source they track. 

That said peak only meters can have less utility for eyeballing a mix as they don't truly represent loudness. Thus my preference for Peak+Ave, but I am repeating myself (I do that a lot).

JR

 
these are the copper oxide rectifiers that MrCMMR was talking about,

these will match up with an old school meter that does not have them installed internally (DC meter)

perhaps a precision rectifier could be substituted, like an OP90 or similar low offset opamp with all the fixins installed as periphery components needed to implement said task,  did that sound too pedantic?  :D
 

Attachments

  • r.JPG
    r.JPG
    142.8 KB · Views: 19
boji said:
Interesting. I ass-umed a strict convention was in order, a 'language' if you will, that did not require translation when other people came to the studio.
Which is about what you get when using standardized VU-meters. A good SE knows that bass can be close to 0VU, Kick drum about -6 and claves should barely be allowed to move the needle. All this with tape as the media. VU meters and tape are partners in crime. Digital wants peak meters.
 
abbey road d enfer said:
Which is about what you get when using standardized VU-meters. A good SE knows that bass can be close to 0VU, Kick drum about -6 and claves should barely be allowed to move the needle. All this with tape as the media. VU meters and tape are partners in crime. Digital wants peak meters.

A perfect example of this was a studio where they had one of the Soundcraft 2" 24 tracks (about the size of a mini-fridge) with peak meters and later upgraded to a Studer 820.  The engineers said that cymbals sounded terrible on the Studer.  They were printing them with the peaks lighting the LEDs and the HF was self-erasing at that level.
 
Some really interesting descriptions there, Thanks Bill W, and  John R    :)

I'm just thinking my way thru it for a while!

....

The very finest VU meters I have used, the ones from mci tape machines (early ones) with the large motor are the triplett us made ones.

They have an internal bridge of sorts, using some ancient glass diode type (red and green painted bands on clear glass diode body)

Anyway, these things are seriously fast and precise ballistics ...  on the e-gtr (clean) and e-bass they respond to the smallest detail, even though they are 'averaging meters'  as has been said.

I have used them with a variety of buffers, including with the mci active buffer pcb that mechanically couples to the meter, providing balanced buffering, with adjustment and some rails for lights (x2).

The best meter with a fine buffer circuit - the other one I use is the jlm vu meter buffer module (now in stereo)  :)

...

I never got to make a really good led ladder based metering pair  for the main mix.  I think there are some better solutions around for the diy person nowadays, as compared to 15 years ago when I started diy in earnest (and wound down my main gig quickly).

I use the TC ELectronics FInalizer a lot for the great 'led array'  metering ...  and nowadays, as I have it on my lounge TV monitor, I use the 'sample rate convert' function alot too.  It really cleans up the hokey digital outputs on the TV,  'media box' and 'dvd recorder box'

..

Oh yes, and for 'edgewise' vu meters with proper motor assembly and ballistics, there is some us made 'simpson' ones that do get around from time to time. These are classic with the curved jewel plastic display casing that sits 'proud' of the facia front  :)

Just superb in the movement too (for a 'vertical needle'  edgwise type!) - large and 'see-able' with some good strong backlighting (not included!)

It really is the 'transparent' vu meter  ...  the whole case is super clear and light transmissable (for such a substantial magnetic assembly)  and  so lighting up is pretty easy      (light er up, boys!)

...

They (triplett, simpson)  do ship with the 'terminator' as I recall,      on a 'springy wire' link,    ..  one replaces it with the supplied little 3K6, 2W ceramic wire resistor    ..  or something like that  ..  in series.

I know when I record the 'clean stringed instruments',  direct or mic  ... 

- they go into my    'la-2a  build'    that has  proper  'gr' and 'vu'  metering  on seperate  'mci / tripletts'  with pink led-bar backlighting    ...    the tripletts get dancing in opposition to each other  ...      I don't really want for any better metering.
 
In the analog chain,  I use the  'triplett's  for metering  ...    then that goes to a Finalizer  for the conversion to the digital domain.

With native digital, like the TV, dvd-recorder, network media box  etc  I use the Finalizer  with it's  3 types of digital inputs and sample rate conversion  ..  if required, it really cleans up  digital clocking bs  with consumer gear  :)

From there, it goes to a Motu 2408 Mk3 with a pci-e (version 3, circa 1995 I think) to the PC.

I have a network going, and a home cloud and what not.

An old Yamaha 03D is the next other hub  in it all ...  THEN with my my    'diy audio' builds  :)

..


The recording media is nowadays most often to the windows PC    ...  the 'dvd recorder' is good for TV (almost never) ..  with a quite fine 'dub audio' capability ....

But the best one is my Tascam DR-40  ....  using the 'bal line inputs'    and  if at a picnic  ..  the  'dual front mics'    :) :)

...

It's all in my 'lounge room' now ..  with the wood heater    :)

My old studio room is a storage space with a crackin' but seldom used Tama kit ... with 2 mics, an interface and my old work laptop. 

Not much use thesedays  ...  all the action is  in the 'lounge room'  with the Tascam , DAW and REW.
 
Back
Top