Vu-Meter to insert

Hi guys, I'd like to build a vu-meter that I could insert on any source, should I go like that :


Or more like that :


The first option looks better right? or maybe none is good..?

saint gillis said:

Hi guys, I'd like to build a vu-meter that I could insert on any source, should I go like that :

The first option looks better right? or maybe none is good..?

Click to expand...

Too complicated. You may use a basic differential amp or an instrumentation amp if you want to use opamps. You may use the THAT line receiver, but you don't need the additional opamps. Google all these terms.
NE5534, why not, but overspec'd for this application.
TL0's are plenty good.

Yes I just picked up the 1st op amp in eagle to illustrate my thoughts...

  I'm just asking about the principle of unbalancing then balancing or simply buffering from the source.. the idea is to get the less distortion possible on my source...

> get the less distortion possible on my source...

Then don't break the Main Line with unbalance/balance chips. The THAT chips are incredibly clean, but why have them at all?

Use a HIGH impedance differential input. Basic idea is that 10X higher impedance does very small damage. Here you want !VERY! small damage, go at least 100X. 1,000X is actually easy. Say most good sources today are 47 Ohm output, twice. A TLO72 can be biased with 1Meg resistors, use two to buffer. That is a 2Meg load on a 94 Ohm source, 20,000X loading.

If "VU" means an actual mechanical meter, you don't even have to un-balance the signal. A meter floats and can be bridged across two buffer outputs. Most "VU-like light-shows" will need an unbalanced input.

PRR said:

A TLO72 can be biased with 1Meg resistors, use two to buffer. That is a 2Meg load on a 94 Ohm source, 20,000X loading.

Click to expand...

So something like the 1st drawing (http://gypsyfarm.free.fr/echange/VU_2.jpg) with tl072s? is it mandatory to bias them? (that means 1Meg resistor between non-inverting input and ground right)

saint gillis said:

PRR said:

A TLO72 can be biased with 1Meg resistors, use two to buffer. That is a 2Meg load on a 94 Ohm source, 20,000X loading.

Click to expand...

So something like the 1st drawing (http://gypsyfarm.free.fr/echange/VU_2.jpg) with tl072s? is it mandatory to bias them? (that means 1Meg resistor between non-inverting input and ground right)

Click to expand...

It may be or may not be necessary. that depends if the point you're eavesdropping has galvanic continuity to ground.
Better assume not, so you have to "bias" them to ground.
But the second opamp cannot work like this, you have to put 4 resistors around it. As I wrote earlier, google differential amp.

Ok I got it! of course can be problematic if it comes from an output transformer...
So should I go like that :


Or more like this :


?

saint gillis said:

Ok I got it! of course can be problematic if it comes from an output transformer...
So should I go like that :

Or more like this :

?

Click to expand...

The second one has a gain of 10x; is it what you want? For unity gain, all resistors should be equal. 1 Meg is ok for TL0, not for 5534.
The first one is overkill.

Yes of course... so 2nd option with all 1meg is better than 1st option?
But just to learn, does the 1st option work fine?

Just remembered this, may be helpful:
http://www.celestial.com.au/~rosswood/diy/passivemeter/passivemeter.htm

I was working out the best way to implement a meter bridge for a recent project of mine. I decided to go passive for simplicity.

I ended up tying the VU's without buffers directly to the + and - outputs but also installed some 4DPT switch's to allow me to defeat the meters(in stereo pairs) when I wanted.

Could be an option to consider.

saint gillis said:

Yes of course... so 2nd option with all 1meg is better than 1st option?
But just to learn, does the 1st option work fine?

Click to expand...

Yes, but as I wrote, is slightly overkill. There are a few motivations for using more than one opamp, particularly if you want additional/variable gain; then , two opamps are good. 3 opamps is gluttony.

Humner said:

I was working out the best way to implement a meter bridge for a recent project of mine. I decided to go passive for simplicity.

I ended up tying the VU's without buffers directly to the + and - outputs but also installed some 4DPT switch's to allow me to defeat the meters(in stereo pairs) when I wanted.

Could be an option to consider.

Click to expand...

VU-meters need a rectifier (most of the time built-in) that introduces some distortion, dependant on the source that is measured. Since the OP wanted a device that he "could insert on any source", a buffer is needed.
The practice of using VU-meters directly across the source originates from the paleolithics of audio, when a fraction of percent distortion could be tolerated because the electronics produced more.
In the old days, the cost of the simplest buffer was a serious issue.
Today, considering the derisory cost of a buffer, I consider connecting a VU-meter permanently across a signal without a buffer as malpractice.

That's true, I understand the distortion that is introduced, In my situation I implemented a defeat switch so I could remove the rectified VU's from the signal when needed.

Granted the cost of the buffers would probably end up cheaper than the switches I used! but my particular situation called for a passive solution.

Those are all good ideas.  I was looking into building a VU meter a while back and found Rod Elliott's tutorial to be very informative, particularly about average vs. peak, which is hard to do right.  See: http://sound.westhost.com/project55.htm

I've have not built this, so I don't know how well it works in practice.  I did find it interesting that one can do this without an opamp, just with some RC damping, although that has to be calibrated.

This comment, "two opamps are good. 3 opamps is gluttony" was very funny, except that good opamps are now so inexpensive you can sprinkle them on your breakfast cereal for that tasty crunch.

Haha!
Also good infos on average/peak metering in Douglas Self 's book Small Signal Audio Design...