Buffer amplfier to isolate VU meter

[mrmagnetophon]

I have many tubes, and only 555 and bd139 and 140 transistors to build a buffer amp rather quick.
Any ideas?
-Chris

[mhennessy]

An emitter follower using any transistor should work. How much help do you need with that? The Wikipedia article is a good first step, although don't be put off by all the theory

http://en.wikipedia.org/wiki/Common_collector

Mark

[mrmagnetophon]

ok that works, but it wont even drive my vu meter, and i put the lm386 schematic together to do that and thats even barely moving my meters.
Sigh

[Kat Manton]

You asked for a "buffer amplifier". Thus Mark (and I) understood your question as a request for an impedance-matching amplifier (having a high input impedance, to avoid loading the preceding stage and low output impedance to drive the meter) with unity voltage gain (the input voltage and output voltage are the same.)

This can be realised with an emitter follower, cathode follower or simple op-amp buffer (output connected to inverting input.)

As this doesn't move your meters, either your circuits as built don't work (difficult to believe as they're very simple) or you require an amplifier (with voltage gain greater than unity) rather than the buffer (unity voltage gain) you originally requested...

How much gain do you require, expressed as voltage gain or in decibels? The circuit you require is now not as simple; a specification of gain is required to calculate component values.

Edit: While we're trying to nail down your requirements, it would be useful to know the details of your VU meters. Are these a basic meter movement with a VU scale (which would require external rectification) or actual VU meters (containing an internal bridge rectifier)? What is their sensitivity?

[Refugee]

A VU meter can be tested for an internal rectifier with a non diode testing resistance range on a DMM.

[Kat Manton]

The VU meters I have here (several different types, but all made by Sifam) have a diode symbol, "Ext. 3k6" and "1.734V" on the label.

With the specified 3k6 external series resistor, they read 0dB for +4dBu input; full-scale (+3dB) for 1.734V (+7dBu) input.

(The VU meters I've encountered in the professional gear I've worked on were all Sifam meters of this type, so I don't actually know if this is pretty much the standard or just very common.)

Assuming that's what you've got and the nominal operating level of your system is +4dBu, all you need is a unity-gain buffer to isolate the meter from whatever you're connecting it to; thus the suggested circuits should work.

Thinking about this a bit more, there are several possibilities:

• The meters aren't VU meters; they're failing to read as you're feeding AC into a basic DC meter movement which happens to have a VU scale.
• Your operating level is lower than +4dBu and the meters are reading correctly.
• Your buffer amplifier circuits don't work.

Lots of assumptions; I think we need more information!

HTH, Kat

[mhennessy]

You didn't mention the LM386 in your first post. This has a lot of gain by default - x20, or +26dB - assuming you built it correctly...

My experience here accords with Kat: 0dBu (0.775V RMS) gives -4dB on the VU scale. Sounds like your signal levels are very low, or the meter is faulty. Or, as Kat suggests, they are simple meters without the built in diodes. The "proper" Sifam ones were pricey, and rarely seen in domestic gear...

Have you got the means to measure signal levels? A 'scope would be nice, but even a basic DVM would help. and you can use free software to make signals on your PC if you don't already have a dedicated oscillator...

As Kat says, more information please

Mark

[Sound Man]

Here is a drive circuit based on a pair of transistors - http://www.circuitsonline.net/schake.../vu-meter.html

David

Here is the translation of Dutch Text:-

This circuit can control analog moving coil meters, for use as a VU meter.
The circuit is left connected to the line terminals of the amplifier.
The VU meter works pretty simple. T1 and T2 amplify the signal. The signal is then rectified by the two diodes and applied to the gauge. The capacitors C3 and C4 ensure that the voltage is slightly flattened and the meter responds less quickly.
After building the circuit must be calibrated. For this purpose, the VU meter connected to a tone generator which supplies 0.3V at 1000 Hz. Then P1 is fully open, that is to say that the wiper of P1 is located at the entrance. Then P2 so cut that regulates the meter full deflection. Then P1 is set to 0.5 mA meter (= half rash) indicates.
For a stereo VU meter should the circuit be built twice.

[mrmagnetophon]

Here is what im trying to do. I have a 3.5 watt amplifier powering the meter, but im getting crosstalk between the channels, im sure its that resistor of 10K anyway to fix that?
Maybe buffer each output and eliminate the load resistor?

[mrmagnetophon]

And yes i am dealing with very low volume levels, as the circuit i was using didn't exactly go together like what i wanted. Sigh.

[mhennessy]

Hi Chris,

You're sighing a lot. And so am I. I can't follow this at all...

In one post you say you have a 3.5 watt amplifier powering the meter, and in the next post you say that the signal levels are very low. In which case, does this suggest that you've implemented a 3.5 watt amplifier working at a tiny fraction of its output power? As you haven't mentioned the gain of the amplifier, or the impedance into which it works, the output power is actually meaningless. Apart from the very real possibility that it's damaged the meter if the meter doesn't include the diodes and series resistor. The voltages need to cause full-scale deflection of a complete and working VU meter have already been given to you.

You haven't addressed any of the points that have been put to you. Specifically:

1. What is the meter? Does it have internal diodes - if not, have you provided them externally?
2. Has the meter survived the experiments so far? If yes, how have you verified this?
2. What is the signal level coming from the microphone preamp under normal conditions?

At this stage, this is all we need.

Sorry if these seem like basic or even patronising questions. But we need to focus on these before we can move forwards.

The mention of crosstalk is separate issue. Your diagram, which is almost impossible to read in any detail, suggest that you have a switch to select which of the two channels goes to the VU meter. A switch won't cause crosstalk unless it's faulty or the output impedance of the preceding amplifiers is unreasonably high. Let's "park" that for now...

Mark

[mrmagnetophon]

1: the meter is an old marion electronic institute meter, has no rectifiers or resistors.
2: yes the meter is fine, i even rebuilt it, i slowly fed it dc voltage, to see if it moves at all, it surely does. (output of portable cd player with rectifier)
3: i cannot measure the signal exactly, but i know its close to line level.

[mhennessy]

Right - so that's your problem - identified by Kat in post #6 - no amount of AC will move the needle, you need to rectify the AC signal!

A bridge rectifier is usually used, and ~3K6 in series is required for the industry-standard Sifam movements, although of course yours may be different. Line-level signals will definitely make a standard VU jump, as discussed in posts #6 and #7. Yours might require some amplification, depending on the sensitivity of the movement, but we can worry about that separately...

You have just about all the answers now; good luck and let us know how you get on

Mark

[mrmagnetophon]

Actually the meter moves when i fed with AC, It stops moving when i feed it DC, yet i rebuilt the meter and it has NO rectifier!
-Chris

[Station X]

Is it a moving coil meter?

[mhennessy]

Originally Posted by mrmagnetophon

2: yes the meter is fine, i even rebuilt it, i slowly fed it dc voltage, to see if it moves at all, it surely does. (output of portable cd player with rectifier)

Originally Posted by mrmagnetophon

Actually the meter moves when i fed with AC, It stops moving when i feed it DC, yet i rebuilt the meter and it has NO rectifier!
-Chris

So which is it?

Those two contradictory messages are ~3 hours apart. Did you change anything in that time? Can you provide pictures of the VU meter?

[Radio Wrangler]

When you rebuilt the meter, did you take any photos of the mechanism? The request above is not for a photo of the front of the thing taken from a distance, but for a close-up of (hopefully) the coil between the pole-pieces of the magnet.

You don't need a small power amp IC to drive a meter. The standard spec for a VU meter calls up 1.7ish volta and 3.6 kilo Ohms. That's 0.000835 Watts Kat posted the numbers a while back. The VU meter was defined by the Bell phone system and called up a meter complete with rectifier and series resistor. The purpose was to put it on a phone line and check the level. Real VU meters have rectifiers. A lot of Japanese audio gear has plain old DC meters with no diodes or resistors in them, but with VU-like scales. They get put in tape recorders, amplifiers etc, with the rectification and scaling done on the main circuit board.

Most of these wannabe VU meters don't meet the specs for the 'ballistics' (timeconstants) of the motion of the pointer, but that may not be too important to you.

You've got a microphone preamp to debug as well as the VU meter. What known working test equipment do you have?... so we can all try to engineer a way to get you going using what's available.

David