Millivolt Meter Recommendation for Audio

[watchnerd]

I'm beefing up my audio testing gear so that I can better calibrate my reel-to-reel tape decks and have a new oscilloscope and function generator. I might use the gear for TT alignment or other experiments (e.g. SUT measuring), as well.

What I'm missing is a really good millivolt meter suitable for audio.

My current multimeter is a Fluke 107, which seems reasonably accurate in the whole 1+ volt range, but I don't seem to get consistent numbers when dealing with .12 - .4 mV level signals, probably due to lack of true RMS capabilities.

Is there something better, more suitable for audio?

Or is the Fluke 107 good enough?

Budget up to $300-$400 would be fine.

Edit , added from @NTK

New table with 10 KHz, 10 Hz and 50 kHz accuracies. Note that only the 6.5 digit Siglent is rated to 10 Hz. The other ones are start at 20 Hz, except Keysight starts at 45 Hz. I kept the calculated numbers and change them to red. They are NOT accurate but represent the accuracies when the signal frequency enters the frequency bandwidth range.

 

[watchnerd]

Can't you measure it with your scope?

You have to use the oscope and the volt meter simultaneously during tape calibration.

You could try to flip back and forth, but it's a bit of a PITA, and I'd rather just pay a few hundred bucks to make things easier.

 

[watchnerd]

For more info, the ones popular amongst tapeheads are vintage meters like:

Leader LMV-189AR

Kikusui AVM-25

 

[NTK]

Benchtop DMM? This is the entry level Siglent, 4-1/2 digits. lowest range is 600 mV, so it should give ~0.0X mV resolution. Next step up is 5-1/2 digit with 200 mV lowest range.
https://siglentna.com/digital-multimeters/sdm3045x-digital-multimeter/

 

[watchnerd]

Benchtop DMM? This is the entry level Siglent, 4-1/2 digits. lowest range is 600 mV, so it should give ~0.0X mV resolution. Next step up is 5-1/2 digit with 200 mV lowest range.
https://siglentna.com/digital-multimeters/sdm3045x-digital-multimeter/

That would match nicely with my Siglent oscope and signal generator.

But I don't get how, if the lowest range is 600 mV, I can measure something like .2 mV?

So the ranges are:

Ranges* 600mV、6V、60V、600V、1000V

Looking in the docs, trying to see what the resolution accuracy is within the 600mV range.

 

[solderdude]

You can turn your PC soundcard into a scope/meter for free and get a tone generator for free.

 

[MakeMineVinyl]

Honestly, in my entire lifetime working with reel to reel tape machines, even in professional studios, I've never used anything other than the VU meters on the recorder. Not to say you can't use an outboard meter, but its likely to just say the same thing as the meter on the recorder. Is there some specific adjustment you need an outboard meter for?

 

[DonH56]

A 'scope is often too noisy to measure mV signals without a lot of averaging. Depending upon what you are doing with it, an analog meter or digital bar graph can be very useful to eyeball peaks.

I have not kept up with typical DMMs so can't help with models. We use Fluke 87's at work and the last high-precision meter I used was an HP/Agilent/Keysight (whatever they are this week) 3458 beast (bench instrument, 8.5 digits, used to trim and measure a 16-bit DAC). I've seen a few relatively cheap on the used market but still well above the $500 range.

 

[watchnerd]

You can turn your PC soundcard into a scope/meter for free and get a tone generator for free.

I prefer having a dedicated device. And it's not like the tone generator was expensive.

 

[watchnerd]

Honestly, in my entire lifetime working with reel to reel tape machines, even in professional studios, I've never used anything other than the VU meters on the recorder. Not to say you can't use an outboard meter, but its likely to just say the same thing as the meter on the recorder. Is there some specific adjustment you need an outboard meter for?

They're vintage machines with sketchy and unknown histories.

I like to test all outputs, find out what is in spec vs not.

Some are undoubtedly overdue for some capacitor replacements. Or board replacements, even.

And I know I have some outputs that are not putting out rated voltage based on what I see on my ADC.

 

[MakeMineVinyl]

They're vintage machines with sketchy and unknown histories.

I like to test all outputs, find out what is in spec vs not.

Some are undoubtedly overdue for some capacitor replacements. Or board replacements, even.

And I know I have some outputs that are not putting out rated voltage based on what I see on my ADC.

Oh, that's more on the repair side. I thought you were needing the outboard metering for routine calibration.

 

[watchnerd]

Oh, that's more on the repair side. I thought you were needing the outboard metering for routine calibration.

Soup to nuts maintenance work, calibration, alignment, overhaul.

 

[watchnerd]

A 'scope is often too noisy to measure mV signals without a lot of averaging. Depending upon what you are doing with it, an analog meter or digital bar graph can be very useful to eyeball peaks.

I have not kept up with typical DMMs so can't help with models. We use Fluke 87's at work and the last high-precision meter I used was an HP/Agilent/Keysight (whatever they are this week) 3458 beast (bench instrument, 8.5 digits, used to trim and measure a 16-bit DAC). I've seen a few relatively cheap on the used market but still well above the $500 range.

Those 87s are about the same price as the desktop models, it looks like.

8.5 digits on the 3458, jeeze....what was the use case?

 

[sergeauckland]

Honestly, in my entire lifetime working with reel to reel tape machines, even in professional studios, I've never used anything other than the VU meters on the recorder. Not to say you can't use an outboard meter, but its likely to just say the same thing as the meter on the recorder. Is there some specific adjustment you need an outboard meter for?

For routine line-up of bias and EQ, I agree, the VU meters are all you need, but you can't measure distortion, wow and flutter and noise using the VU meters.

I still swear by my Ferrograph RTS-2 test set which has most of what you need also for amplifier maintenance.

As far as audio millivoltmeters go, I've been very happy with my Levell TM3A. It goes from 500V down to 15uV. The TM3B is the same with a bigger meter.

S

 

[DonH56]

Those 87s are about the same price as the desktop models, it looks like.

8.5 digits on the 3458, jeeze....what was the use case?

Data converters and low-level instrumentation systems. Last time was for a 16-bit DAC design of mine (not audio) where we needed uV (or less) accuracy and stability over the full range of the DAC. When doing precision work there are a lot of times it came in handy, and was great for troubleshooting things like power/ground plane drops. Also used it for leakage measurements (current and voltage) when the pA meter or fV leaf voltmeter wasn't available. For an 8.5-digit meter it was fairly fast for ATE use though we only used it for low-volume (very) precision measurements when required for production.

Consider we were aiming for 1/8-lsb trimming with a 1 V full-scale so needed 2 uV accuracy and stability. It was a nightmare, 3458A or no...

 

[MakeMineVinyl]

For routine line-up of bias and EQ, I agree, the VU meters are all you need, but you can't measure distortion, wow and flutter and noise using the VU meters.

I still swear by my Ferrograph RTS-2 test set which has most of what you need also for amplifier maintenance.

As far as audio millivoltmeters go, I've been very happy with my Levell TM3A. It goes from 500V down to 15uV. The TM3B is the same with a bigger meter.

S

One thing to keep in mind is that the absolute value of record/playback distortion is not as important as the fact that the distortion consists completely (or close to completely) of odd-order harmonics, with an absence of even order harmonics. If there is anything more than a trace of even order order harmonics, this indicates that the bias oscillator is not outputting a symmetrical waveform and/or that the heads are in need of being demagnetized.

 

[sergeauckland]

One thing to keep in mind is that the absolute value of record/playback distortion is not as important as the fact that the distortion consists completely (or close to completely) of odd-order harmonics, with an absence of even order harmonics. If there is anything more than a trace of even order order harmonics, this indicates that the bias oscillator is not outputting a symmetrical waveform and/or that the heads are in need of being demagnetized.

Indeed, and that's one benefit of an FFT based distortion measurement as it's immediately obvious what the harmonic content is. It's possible to work it out from a distortion residual trace on a 'scope, just a lot less obvious. As to heads needing demagging, this never happened in the studios I worked in as machines were demagged very morning and every time a test tape was brought anywhere near a machine. It was a practice I used at home when I still used tape. Including a line-up check of at least bias and azimuth before every recording. A right pain, and one reason I went over to PC recording as soon as I could. My first PC soundcard (Digigram) cost about the same as I'd paid for my Nakamichi.

S

 

[Digital Mastering System]

HP EL400

 

[sergeauckland]

HP EL400
View attachment 81039

That looks remarkably like the one I had in the first lab I worked in.
Mine was a Philips PM 2454. I wonder who made it and who badged it, Philips or HP?

 

[solderdude]

https://nl.aliexpress.com/item/33021370646.html?spm=a2g0z.12010612.8148356.15.acda69c27VOTqy