• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

dBm or dBV Audio Voltmeter

ECC33

Member
2008-03-08 1:29 am
Hello All,

I have been looking into getting an audio voltmeter with a dB scale for frequency response measurements.

Some have got a dBm scale; 0 dBm = 1 mW/600Ω, and some have got a dBV scale; 0 dB = 1 Volt.

What, in real-world terms, will be the difference and which is the one to get?

Many thanks for any help!
 
ECC33 said:
Some have got a dBm scale; 0 dBm = 1 mW/600&, and some have got a dBV scale; 0 dB = 1 Volt.

What, in real-world terms, will be the difference and which is the one to get?

db(m) measures power ratios:

db(m)= 10log(Po / Pi) (Where P is in milliwatts)

db(v) measures voltage ratios:

db(v)= 20log(Vo / Vi)

Therefore, a change of +/- 3.0db(m) is a halving or doubling of the power, and a +/- 3.0db(v) change is by either sqrt(2) or sqrt(0.5).
 
ECC33 said:
Hello All,

I have been looking into getting an audio voltmeter with a dB scale for frequency response measurements.

Some have got a dBm scale; 0 dBm = 1 mW/600&, and some have got a dBV scale; 0 dB = 1 Volt.

What, in real-world terms, will be the difference and which is the one to get?

Many thanks for any help!

Because the dB has no specific value and is only a reference, it must be referenced to something. The internationally accepted standard is now 1 milliwatt in a 600 ohm terminated line which is 0.775 volts for 0 dB(m). (The old standard, no longer used, was 500 ohms.) Almost all professional AC voltmeters use the 1mW/600 ohm standard.

Some less professional meters will reference 0 dB to end of scale at 1 volt as a convenience. This will let you measure differences in levels expressed in dBs (dBv), but you must always remember that the true reference for power levels is actually lower. (0.775v) This could lead to mistakes in some cases. So for best all around use, I'd stick with the 1mW/600 ohm 0dBm reference.

Since decibles are a logarithmic function, the dB divisions on a regular linear scale meter movement are compressed at the low end making readings slightly difficult. Because of this, some meter movements are made non-linear by magnetically biasing the movement toward the low end. Doing so makes the dB scale linear and easier to read. Unfortunately the voltage scale now becomes non-linear and a little harder to read. So it's a trade-off that you must choose between.

Still other meters simply put the dB scale at the top most part of the face so that the compression is more spread out and easier to deal with. I prefer Hewlett-Packard meters because they're easy to get on the used market and are the most accurate because HP individually calibrates each printed scale and movement together, as one, at the factory when made. And I assume that you want an analog meter and not a digital. Digitals are ok for absolute measurements, but are very nerve racking for watching changes.

Victor
 
AndrewT said:
I wouldn't use a meter to measure frequency response.
You may find that the meter is less accurate than the device being tested.
Compare input to output!!!!!

Even great meters like the HP3478a and HP3456a limit their bandwidth to prevent RFI from screwing up your measurements. Analyzers like the Tektronix DA4084 and AA501, Boonton 1120 are good to a bit above 100kHz.

Most hand held DVM's are limited to a few kHz.
 

EC8010

Ex-Moderator
2003-01-18 7:57 am
Near London. UK
Here we go again! dBm is history. Once upon a time there genuinely were 600 Ohm audio transmission lines (miles long) which were driven from 600 Ohm and terminated in 600 Ohm. Studio centres even used 600 Ohm internally. Meters had dBm written on their scales, and has been correctly pointed out by other posters, 1mW into 600 Ohms requires 0.775V. The thing is, such a meter read voltage, not power, and people didn't like all this 600 Ohm nonsense anyway, so a new standard was adopted; dBu. 0dBu is 0.775V and takes no account of impedance. Studios (the few that are analogue) drive from a low source impedance into a high load impedance, and even better, many high load impedances. This couldn't be done with the 600 Ohm nonsense - you needed a distribution amplifier with as many outputs as loads.

The trouble is, the instrument makers didn't see the joke, and persisted in labelling their scales dBm when they were really measuring dBu. Rant over.

My audio test set uses a nice analogue meter and can easily measure frequency response to 0.02dB from 10Hz to 300kHz. As pointed out earlier, DMMs are best for measuring mains; although the more expensive ones go up to tens of kHz.
 
ECC33 said:

Next question, any recommendations please?
DMMs have got their uses, but I was looking for an analogue meter.

Which meter have you got?

These are the analog meters I would recommend: Hewlett-Packard...

400E or EL (L=log meter movment) good to 10Mhz
400F or FL good to 4Mhz & has a swichable low-pass filter.
400G or GL good to 4Mhz
3400A true RMS good to 10Mhz.
Option 1 on any of these is dB scale on top.

427A multi function AC/DC/volts/current/ohms. (1 Mhz) Battery only.
option 1 gives both AC line operation or battery (22v)

For any of these meters, be sure to get a late version. Stay away from the old blue/gray color and get the mint green/tan color. Especially for the 3400A. The older choppers and thermal couplers inside (3400A only) get flaky with age.

Kikusui and Leader from Japan makes a decent instrument.
Boonton 93A is very good.
Ballantine 323 or 3015A not bad either.
Some of the very old Ballantine 300 series (tubes) were nice meters in their day. They're real cheap now but need restoration.

Victor
 
I like the HP400E- very trouble free but not true rms. The HP3400s are also good, true rms, but not nearly as reliable. I don't use either one as much these days, relying on digital bench meters, or the scope. AC analog meters are great for peaking things up or measuring nulls, but the inability to see just what you're measuring is a great weakness. Is it signal, or is it noise? There was also a Heathkit AC voltmeter, very similar to the HP400, that was actually quite good- the IM-5238.
 
Conrad Hoffman said:
The HP3400s are also good, true rms, but not nearly as reliable.

A dead HP3400 thermopile is about as useful as a dead flashlight battery. HP used to have an "exchange program" for the thermopiles. I wouldn't say they are "unreliable", as much as they can't be abused willy nilly. Precision is limited to a couple digits, unlike an HP3478 or HP34401.

I have an "in-cal" HP function generator and the Fluke 8920A and HP3403C are about equal.

I don't know why prices for the HP3581 wave analyzer have spiked so much -- these have a very good RMS detector and are very acurate to 50kHz -- and will measure down to some tens of nano-volts -- but there's a Genrad model out there as well.
 

ECC33

Member
2008-03-08 1:29 am
Thank-you all again, for your help.
Much appreciated:)

The thing is, I've got a distinct disadvantage to most of you; I'm in the UK; so finding anything remotely useful over here can be either highly unlikely, very expensive, or both!

There is a chap on ebay selling HP3400s for £45. I do like Heathkit stuff, I've got some here, but it's harder to find over here.

Is true-RMS really necessary?

The Fluke 8060a looks like a candidate; according to tubebuilder:-

http://www.tubebuilder.com/testeqreviews.html#fluke8060


After browsing the line-up of the FLUKE meters, I came across the 8060a. This instrument is perfect for making audio measurements. Applications include making direct gain measurement in dB for amplifier stages. It measures the 3dB roll-off points indicating the frequency and voltage at these points. Very high accuracy (0.05% basic). One caution with these meter; do not measure voltage when you are set-up to measure current. Check where your leads are plugged into before use. No real damage occurs, but it is very awkward to replace internal fuses.

Anyone own a Fluke 8060a?