[Bob Reynolds]

Can anyone give me some advice on how to match output levels with a voltmeter? What sort of test tone should be used?

Thanks,
Bob

[JohnR]

I've not done it, but a simple sine wave would be fine. Or something else that has a steady amplitide.

However you should make sure the amplitude is quite low.

[Bob Reynolds]

I was thinking about the warble tones on one of the Stereophile test CDs. I've found them easier to work with acoustically using an SPL meter, but I don't know what they'll look like on an analog voltmeter.

[Russell Dawkins]

I agree with JohnR. I have used sine and triangle waves from a frequency generator at various frequencies.
To get an accurate reading, you would want to stay away from warble tones and noise - like pink or white - the amplitude fluctuation inherent in such signals would prevent accurate reading of small variations in voltages.

[andyr]

I was thinking about the warble tones on one of the Stereophile test CDs. I've found them easier to work with acoustically using an SPL meter, but I don't know what they'll look like on an analog voltmeter.

Hi Bob,

The reason the warble tones are better than straight sine waves when you use an SPL meter is that they energise the room modes less.  If you're just matching output levels of an electronic device then sine waves is the way to go!     And if you can afford to pick up a CRO ... that will be soooo much more useful than a meter.

Regards,

Andy

[PEB]

A pure sine tone near 1kHz is best for matching output levels of electronics.  Do not use higher frequencies because many DVM's are meant to be used at or below 1kHz.

If an amplifier is involved, either look for a test CD with -20dB recorded signal, or else be sure to set the level to 100mV [0.1V] or lower before invoking the amplifier.  Use common sense: turn the preamp volume all the way down, and then monitor the amp output voltage while you bring up the volume.  (Don't forget to set the meter to AC!)

Keep in mind that there is a risk with long periods of testing with pure sine tones through speakers.  It is possible that a particular crossover component will receive considerable power just because of the test frequency.  Under normal music circumstances, the duty cycle of the crossover parts is lowered.

It is a safer practice to match levels without the speaker connected.  After that, it becomes a room acoustics issue.  Even in a perfectly symmetrical room, a slight difference in speaker/listener distance with left/right speakers can lead to a 1dB channel imbalance.

Phil

[Bob Reynolds]

Thanks to everyone. I have a few questions/comments.

JohnR: Sine wave makes sense, thanks. I was guessing around 1KHz, but unsure of the level -- don't want to drive the amp into clipping.

Russell: That's what I thought would happen with a warble tone -- making getting an accurate reading impossible.

AndyR: Thanks for the comment about warble tones and room modes -- I didn't know that. Also, what is a CRO?

Phil: Thanks for the confirmation. Do you think an analog voltmeter will work or do I need to pick up a digital meter? If so, can you suggest a decent one? Maybe an entry-level Fluke? Thanks for the reminder about starting low and using the AC setting. The comment about duty cycle makes sense and I never would have considered that. I will disconnect the speakers first. I have experienced that furniture placement in a living room can effect balance by 1 - 1.5dB.

OK, so does anyone know of a CD with reasonable sine waves for doing this sort of thing or do I need a signal generator.

I don't own a laptop so software may not be an option. My PC is approximately 50 feet away in another room from the stereo. I have considered using TrueRTA; I have a mic, mic preamp and external sound card. The problem is how to make the connections over 50 feet. Any thoughts on solving this would be appreciated.

--Bob

[Bob Reynolds]

It looks like the Stereophile Test CD #3 has such a track, #27 - 1KHz sinewave (dithered) at -90.31dBFS. It runs for 20 seconds.

http://www.stereophile.com//features/424/index.html

By the way, can someone explain dBFS to me? And the signal is 90.31dB down from what level?

Thanks, Bob

[Russell Dawkins]

dBFS is dB referenced to full scale (0dB) which is the maximum level represented on a CD. A properly mastered CD hits close to 0 dBFS somewhere on the program. If the material is classical, it may go to near zero a couple of times but the average level may be -30 dB. If it is pop the average may be -10 or as little as -4dBFS.
The output from the player or D/A converter is supposed to be 2.0 V.  at 0dBFS but there is wide variability in this. I've seen it range from 1.2 V. to around 3 V.

At -90 dBFS, you would never guess there was a signal there. You would have to run your pre flat out to hear the tone at all.

[andyr]

Thanks to everyone. ... AndyR: Thanks for the comment about warble tones and room modes -- I didn't know that. Also, what is a CRO? ...

Hi Bob,

Sorry, CRO = Cathode Ray Oscilloscope ... aka as a "Scope"!    

Regards,

Andy

PS:  Yeah, I've got Stereophile Test CD #3 too ... I recommend it.  I don't know that I'd use the "dithered" 1Khz tone, though (mainly bcoz I have no idea what "dithering" does to it!!).     There are plenty of other useful tones on the CD!

[Bob Reynolds]

dBFS is dB referenced to full scale (0dB) which is the maximum level represented on a CD. A properly mastered CD hits close to 0 dBFS somewhere on the program. If the material is classical, it may go to near zero a couple of times but the average level may be -30 dB. If it is pop the average may be -10 or as little as -4dBFS.
The output from the player or D/A converter is supposed to be 2.0 V.  at 0dBFS but there is wide variability in this. I've seen it range from 1.2 V. to around 3 V.

At -90 dBFS, you would never guess there was a signal there. You would have to run your pre flat out to hear the tone at all.

Thanks Russell.

It came to me later that FS probably meant full scale. For a digital medium like CD, I assume that is FFFF (hex), i.e., all bits set to one -- that would give the largest amplitude.

So the challenge for the recording engineer is to have the maximum level of the performance as close to, but less than 0dBFS. That way he captures as much of the dynamic range of the performance as the medium can carry. Is that correct?

I've seen the balanced output of some CD players hit close to 4V, so I assume the -90dBFS is playing it safe.

[Bob Reynolds]

Hi Bob,

Sorry, CRO = Cathode Ray Oscilloscope ... aka as a "Scope"!    

Regards,

Andy

PS:  Yeah, I've got Stereophile Test CD #3 too ... I recommend it.  I don't know that I'd use the "dithered" 1Khz tone, though (mainly bcoz I have no idea what "dithering" does to it!!).     There are plenty of other useful tones on the CD!

Hi Andy,

I assume that today there are other display devices for scopes. Being an old guy, CRT is the only way I think of a scope.

I've read the word "dither" thousands of times and I have some notion of what it means, but if you could shed some light on the term I'd appreciate it.

I've used the 1/3 octave warble tones before, but found they weren't "fine" enough for bass level. Does anyone here use the TrueRTA software?

[Davey]

Bob,

Yep, 0dbFS is the maximum level signal that can be "recorded" to a CD.  Most CD players with normal, unbalanced connections will output approximately 2 volts RMS with a 0dbFS signal.  Balanced outputs usually have 6db higher signal levels which explains the 4 volts you mentioned.

Track 27 on the Stereophile CD is not appropriate for what you want to do.  It's designed for a totally different purpose.  The Stereophile CD3 doesn't have a track appropriate for this.

I would suggest you (and anyone else) download a copy of http://www.goldwave.com
You can generate whatever type of test signal you want and then burn to CD to master your own "test disk."

Regarding dithering......Wikipedia or Google or ??? is your friend.

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

Cheers,

Davey.

[Bob Reynolds]

Hi Davey,

Can you explain why track 27 on the Stereophile CD3 is not appropriate? That way I'll have some idea of what signal to generate.

Thanks,
Bob

[Davey]

Bob,

Russell explained it already.  A -90dbFS recorded track is essentially silence.  A track like that is for examining low-level D/A performance of a CD player.  That's not what you're interested.  Actually, the Stereophile CD3 isn't much of a "test"ing CD IMHO.    Their Test CD2 is better for this type of thing.

Anyways, what you want is a midband tone (1khz is fine, 400Hz is fine too) recorded at somewhere between 0dbFS and -20dbFS.  That will give you a nice solid signal that will indicate consistently on a standard voltmeter.

Cheers,

Davey.

[Bob Reynolds]

Thanks to all. I will attempt to construct a test signal this weekend. I've narrowed down the choices of Fluke digital multimeters. So soon I should be able to give it a go.

-- Bob

[Bob Reynolds]

Just a small update. The M&K pro web site has an area of setup test tones -

http://www.mkprofessional.com/wav/index.htm

They list a 1KHz tone at -20dBFS for level matching between channels.

I use their MPS-5310 subwoofer and LFE-4 bass management controller, so I should have remembered using that when I did the setup.