Meter tap of floating transformer output

Let's imagine we've got a fully floating output like the one with transformer. Let's say we need to put a LED meter on this output. Is it better to make a balanced to unbalanced converter and work with it's output or will it suffice to take just one output tx leg and measure it relative to ground?

I'm asking because while observing floated output signal I noticed that the waveform of one leg referenced to ground may look very funky, and to view it properly, the differential probe is needed. I wonder how it translates into levels.

You need to look at it differentially but can use a high resistor value differential for line level metering.

JR

JohnRoberts said:

You need to look at it differentially but can use a high resistor value differential for line level metering.

Click to expand...

Something like this?

It seems a shame to throw a way a floating output with its very useful galvanic isolation, just to add a LED meter. You can preserve the galvanic isolation by using a cheap 10K:10K transformer at its input. And when I say cheap I mean really cheap. I use this one for VU meters:

http://uk.farnell.com/triad-magnetics/ty-141p/transformer-1-1-10k-10k/dp/1610409

Although the frequency response is specified as 200Hz to 15KHz, in practice, when driven by a relatively low impedance source, it easily extends from 20Hz to 20KHz. I have measured a couple and they do have a decent primary inductance so they are capable of good low frequency response. Holger uses them in his analogue VU meter buffer I believe.

Cheers

Ian

Hi ian - I wonder if you can leave out the opamp buffer and go 10K:10K transformer followed by a diode bridge then to some resistors(for calibration) and connect a cheapo DC VU meter? Did you try this when testing?

I like the idea of an all passive solution

Humner said:

Hi ian - I wonder if you can leave out the opamp buffer and go 10K:10K transformer followed by a diode bridge then to some resistors(for calibration) and connect a cheapo DC VU meter? Did you try this when testing?

I like the idea of an all passive solution

Click to expand...

A classic VU meter is just a resistor, a bridge rectifier and a DC meter. it is already floating so you do not need the extra transformer.

Cheers

Ian

ruffrecords said:

It seems a shame to throw a way a floating output with its very useful galvanic isolation, just to add a LED meter. You can preserve the galvanic isolation by using a cheap 10K:10K transformer at its input. And when I say cheap I mean really cheap. I use this one for VU meters:

http://uk.farnell.com/triad-magnetics/ty-141p/transformer-1-1-10k-10k/dp/1610409

Although the frequency response is specified as 200Hz to 15KHz, in practice, when driven by a relatively low impedance source, it easily extends from 20Hz to 20KHz. I have measured a couple and they do have a decent primary inductance so they are capable of good low frequency response. Holger uses them in his analogue VU meter buffer I believe.

Click to expand...

Ian, thanks for your input. This option is quite smart, although unfortunately I don't have enough space to fit additional transformer, even this small.

Ilya said:

Something like this?

Click to expand...

Yes, but you could use even higher value resistors and probably would be better off with a bifet op amp (like old TL07x series.

BTW with the differential you can scale gain down to make the metering easier.

JR

ruffrecords said:

A classic VU meter is just a resistor, a bridge rectifier and a DC meter. it is already floating so you do not need the extra transformer.

Cheers

Ian

Click to expand...

I was thinking that the transformer would be introduced as a way to reduce distortion.  The transformer should isolate the VU meter to reduce distortion in the circuit compared to connecting the diode bridge+VU directly to the audio signal.

> Let's imagine we've got a fully floating output like the one with transformer.

Let's imagine this transformer has a non-floating primary.

Meter the primary. The difference should be one dB or less.

PRR said:

Let's imagine this transformer has a non-floating primary.

Meter the primary. The difference should be one dB or less.

Click to expand...

Unfortunately this is not the case in my particular situation. I thought about measuring at some intermediate stage, but discarded this idea.

Humner said:

I was thinking that the transformer would be introduced as a way to reduce distortion.  The transformer should isolate the VU meter to reduce distortion in the circuit compared to connecting the diode bridge+VU directly to the audio signal.

Click to expand...

At first sight you might think that but in actual fact it just couples the meter to the load almost exactly as if the transformer was not there.  I wondered about this for a compressor  project where I wanted to drive the side chain rectifier from a separate secondary of a transformer. I simulated it and the rectifier added distortion exactly as if the rectifier were connected to the main output.

Cheers

Ian