Film-Tech Forum ARCHIVE: RMS measurement

This is topic RMS measurement - DTS in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=006006

Posted by Antonio Marcheselli (Member # 373) on 03-29-2004, 02:31 PM:

Hi everyone

Some time ago there was a thread where we speak about using a "true rms" meter to measure the 1000Hz and 31.5Hz tones used by the DTS players. Someone said that a "good" meter is usually able to measure the 1000Hz with no problems.

Since I'm going to check all my dts' output levels during these days and I have to buy a new meter, I found that many of the meter at the shops are rated "40-300Hz". The "TRUE RMS" meters costs 300+ bucks and are rated up to higher frequencies (1000Hz).

Before buying a 300$ meter I was thinking if a true rms meter is really needed to make a correct measurement. I found a 20mV difference between two different "good" meters.

My question is: what kind of difference can I expect between a "true rms" meter and a "good" meter? How do you measure the dts' tones?

Bye
Antonio

Posted by Alan Haigh (Member # 1781) on 03-29-2004, 03:01 PM:

I think the best solution is to buy a cheap meter and borrow a good meter. Set the DTS to 300mV with the good meter and measure this with the cheap meter. Write this number down and set everything to this. Also if you get one back from repair it will be set to 300mV with a good meter.

Alan

Posted by Mark Gulbrandsen (Member # 72) on 03-29-2004, 06:36 PM:

Actually many DMM's are capable of true RMS readings. They do not cost 300.00 either...even new!! You can also find a good used VTVM (Vacuum Tube Voltmeter)and all of them will read true RMS AC voltage very accurately. I have a Hewlett Packard VTVM that actually has a preamplifier tube in the AC probe that makes the meters response flat to almost 5mhz. Just try to find a DMM that'll do that feat.

This one on E-bay is very similar to the H-P meter I have and use mainly for antique TV restoration. Note that this meter is at about 20 bucks right now...pretty cheap.

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=3806943321&category=58275

Leader also makes a good AC milivolt meter which has dual pointers and I also have one and use it for audio design work.

There is another way to do this adjustment which actually has more meaning as to balancing it to your system. and that is to compare the output of your pink noise generator card as you would balance your "B Chain". Once the "B Chain" is balanced then use the SPL to set the DTS level to the same reading as your pink noise card reads. I find this method is actually far more accurate and there is practically no variance in level when going from Anslog to DTS. The only thing you get is the slight blip from the delay in reversion.

Hope this helps out....

Mark @ CLACO

Posted by Gilbert Travin (Member # 2138) on 03-30-2004, 08:05 AM:

If you buy a RMS meter, you must verify its bandwith capability and if it is TRUE RMS. I don't know your DTS system but I think that it delivers white noise signal. For an accurate measurement, the true RMS meter is essential ! The waveform is important : with a sinewave, all meters are all-right, but with a high harmonic waveform (such as square), only the true RMS meter gives the good measurement. In a white noise signal you have a lot of harmonics.
If you notice a little difference between a true RMS and a cheap meter, the solution of Alan is the better : it is a calibration for the cheap meter ! But this method is valid ONLY for THIS measurement !!!! If you need make another true RMS measurement on another system you must start again another calibration of the cheap meter ....

Posted by Antonio Marcheselli (Member # 373) on 03-31-2004, 03:21 PM:

Hi Gilbert

No, the DTS tones are sinewave (31.5Hz for sw and 1000Hz for others channels) and not white noise.

So you think that the measurement are the same between TRMS and a good meter with a 1000Hz sinewave?

What do YOU usually to check dts' levels?

Mark,
Once I check my DTS just received from UK and my "normal" meter says 300mV +/- 5mV.

Bye
A

Posted by Mark Gulbrandsen (Member # 72) on 03-31-2004, 05:35 PM:

A brand new Fluke 110 that reads true RMS AC voltages sells for less than 110.00 U.S.. Not such a big expenditure for an important piece of equipment.

Mark @ CLACO

Posted by Gilbert Travin (Member # 2138) on 04-01-2004, 01:05 AM:

Hi Antonio !

Ok !
Generaly, the AC capability of a meter is : sinewave and 50/60 Hz.
The 31.5 Hz sinewave is probably correctly "readed" by the "cheap" meter. BUT, it is possible that the frequency of the 1000 Hz signal (sine) is out of the bandwidth of the meter and you notice a difference with a "good meter".
The "high quality" true RMS meters incorporate a microprocessor that compute exactly the RMS value by analysing the signal waveform and frequency : they are more expensive !

Posted by Antonio Marcheselli (Member # 373) on 04-02-2004, 04:15 AM:

Thanks to all

I will search for other brands of meters since the one I found costs about 350$!

Bye
A

Posted by Mark Gulbrandsen (Member # 72) on 04-03-2004, 04:29 PM:

Antonio,

I thought I'd drag this back to the top as I had to do this this past week and it works fine..... have not done this in some time and completely forgot about it.

If you have an accurate scope you can take the peak to peak AC voltage and divide that by .707 for the RMS AC voltage. This is only true for a sine wave though... not pink noise.

Mark @ CLACO

Posted by Gilbert Travin (Member # 2138) on 04-05-2004, 04:13 AM:

Hello Mark !!!!

Hummm, i beleave that :

peak to peak value = 2 * sqrt(2) * RMS value, for a sine wave (of course !)
where sqrt(2) = square root of 2 = 1.414

or : RMS = (p-to-p)/2.828 = (0-to-peak)/1.414

Of course : 0.707 = 1/sqrt(2) [Big Grin]

Posted by Alan Haigh (Member # 1781) on 04-06-2004, 03:03 PM:

Today I measured my cheap Metex meter against an Expensive Neutrik analyser. All outputs measured 300mV on the Neutrik.

For 1kHz tone - 280mV
For 31Hz (sub channel test tone on DS5) - 304mV

I think they are usually set to be accurate in the 50 or 60 Hz area. It's interesting though isn't it?

Alan

Posted by Matt Zeiner (Member # 1914) on 04-06-2004, 03:51 PM:

FWIW, the newest version of the Ivie IE-33 software calibrates the line input to 1V and they added a voltmeter function that reads true RMS up to 20 volts across the entire audio spectrum.

Posted by Mark Gulbrandsen (Member # 72) on 04-06-2004, 05:04 PM:

Still, the best way to calibrate the O.P. levels is to do it with a mic in the auditorium and with the DTS playing through the processor you are going to be using. Run the Pink Noise on the setup disk and calibrate it to match the levels you set AFTER you have calibrated the B chain levels. You will normally find that the levels may not fall at exactly 300mv on each channel when you are done due to slight variances in components in the processor. Few if any processors use 1% tolerance parts in but a few very critical locations. However you will find that when you are finished doing it this way it will match the analog optical level exactly making reversiions much less noticable.

As for Alan reading steady state tones of any frequency with a standard DMM he will ultimately find it a waste of time. Standard DMMs generally do not have flat enough frequency response to be at all meaningful except at the frequency the manufacturer states it is calibrated FOR....on many cheap meters they do not even state that. You will find that meters that are capable of reading true RMS generally have VERY large input coupling capacitors and are calibrated refrenced to 1khz.

Matt, My IVIE PC-40 has RMS AC and is extremely accurate... about + or - .05% deviation from 1 volt from 20 to 20khz! Thats far better than either my HP AC VTVM or my Leader dual pointer RMS AC voltmeter which are about 1 to 2 percent of full scale accuracy.

With the availability of inexpensive new RMS capable DMMs and plenty of truely great used meters that can be had for mere pennys why not just do it with the right equipment in the first place....... Any other way, excepting the use of a scope along with math conversion formulas is really a waste of time. And that last method can be time consuming!

Mark

Posted by Gordon McLeod (Member # 33) on 04-06-2004, 07:56 PM:

Mag systems are set up with steady state tone and the DA 20 is in effect fixed output based on bus level

Posted by Alan Haigh (Member # 1781) on 04-07-2004, 09:34 AM:

I think I'm fine now, actually. 1kHz off the disk at 280mV and sub at 304mV. completely accurate and not a waste of time -- as long as I use the DS5 disk. Also this tone method is by far the most accurate way. There will always be errors acoustically due to diffence in reference pink noise levels (Dolby is not 300mV RMS pink noise). By setting DTS to 300mV it matches the DA20.