I'm a musician and I've been aligning and calibrating some old tape machines with great success. Currently I just use a desktop Mac for checking levels. I run the cables into an audio interface, look at the dbfs meters in software and tweak the pots on the machine until I get the L&R channels evenly matched within 0.1db.
But now I'm getting tired of lugging around machines over to my computer. I have a couple multimeters but they're only single channel. Would a dual channel multimeter be a good fit for my purposes, and if so are there any out there that won't break the bank?
ou have precise vu meters who could do the job instead ?
A dual display meter like the prova-903, pls note the dual inputs are common ground, they are not floating / isolated0.1 db appart means some resolution, not sure an meter will be good for dedicated audio ? or use an portable dual channels scope ?
You calibarte them at 0db with 1khz signal ?
Or a gadjet like this
https://www.ebay.com/itm/184796771209?
ou have precise vu meters who could do the job instead ?
A dual display meter like the prova-903, pls note the dual inputs are common ground, they are not floating / isolated
0.1 db appart means some resolution, not sure an meter will be good for dedicated audio ? or use an portable dual channels scope ?
0.1dB is 1% so it does not require a super accurate DMM to measure the level. The prova-903 looks like a tool that can do the job. It has a wide frequency range for AC measurements. Many multimeters only work well for 50Hz/60Hz. An oscilloscope is not suitable as the DC accuracy is ballpark 1% to 3%.
Keithley makes desktop DMMs that can be used for THD and other audio features. 2015 and 2016 if I recall. They show up on ebay at reasonable prices sometimes, like 300 USD. I don't think it comes with two channels, but you can select between the front and the rear terminals with a clunking pushbutton. So you make a cable and then you can switch between them, and get the things to the required reference level.
Buy a small function generator to supply the signal, them use the same DMM for both channels.
With a stable input signal the calibration will be easier.
IMHO having a dual channel multimeter is very handy when working on audio. Not just for comparing left /right but also input versus output. Having to connect the probes over and over gets tedious quickly.
IMHO having a dual channel multimeter is very handy when working on audio. Not just for comparing left /right but also input versus output. Having to connect the probes over and over gets tedious quickly.
You may be better off with two meters, than one dual channel, but to adjust meters you are best off with stable input signal.
I have taken a look on the Prova-903 meter:
https://lygte-info.dk/review/DMMProva%20903%20UK.html
The prova-903 looks like a tool that can do the job. It has a wide frequency range for AC measurements. Many multimeters only work well for 50Hz/60Hz. An oscilloscope is not suitable as the DC accuracy is ballpark 1% to 3%.
I'm going to have to respectfully disagree on two points.
First, IMO the Prova 903 would likely be a worse setup than simply using two decent multimeters. The only reason to use it would be if you really, really wanted a one-box solution. Aside from the obvious issue of the common ground, reviews (from others on this forum and elsewhere) reveal that it is slow and doesn't meet actually meet its specs--which I believe since it purports to exceed even TOTL meters like the Fluke 289. There's NO WAY it reliably meets a 0.1% ACV spec, even with the 20 counts floor--that is solidly in good bench meter territory. Of course it doesn't have to meet it's spec to do this job, since 1% is probably plenty good for the OPs purpose and it is easy enough to characterize the meters by feeding them the same signal and comparing. Which brings me to...
Second, a proper DSO is a much, much better tool for this IMO, aside from the bulk and line cord. You don't need absolute accuracy for the OPs use, just relative accuracy. That's easy enough to test by simply giving your two channels the same signal. Many decent modern DSOs will be much more accurate than 1%, both relatively and absolutely. Here's a 1.00000V
rms 1kHz signal from a calibrator put into CH1 and CH1 of a Siglent 2354X+ that was just turned on with no warmup and no self-cal.
But now I'm getting tired of lugging around machines over to my computer. I have a couple multimeters but they're only single channel. Would a dual channel multimeter be a good fit for my purposes, and if so are there any out there that won't break the bank?
Get a Siglent SDS1202X-E oscilloscope on sale for $279 or so. It will do what you need and much, much more.
I’m curious about this topic and not an engineer. Wouldn’t one of these cheap meters do exactly what you need?
leader lmv-186a
I own a few models of the Leader family. Nice instruments for analog gear. Also, I’m asking because I don’t know the answer. Is 0.1db accuracy needed in audio? What human can decipher that level? I am a huge fan of micro-accuracy though and admire your desire for excellence and high standards.
With all due respect.
Wes
The prova-903 looks like a tool that can do the job. It has a wide frequency range for AC measurements. Many multimeters only work well for 50Hz/60Hz. An oscilloscope is not suitable as the DC accuracy is ballpark 1% to 3%.
I'm going to have to respectfully disagree on two points.
First, IMO the Prova 903 would likely be a worse setup than simply using two decent multimeters. The only reason to use it would be if you really, really wanted a one-box solution.
OK. Didn't know it didn't meet the specs. It is quite expensive so I expected it would perform better.
Second, a proper DSO is a much, much better tool for this IMO, aside from the bulk and line cord. You don't need absolute accuracy for the OPs use, just relative accuracy. That's easy enough to test by simply giving your two channels the same signal. Many decent modern DSOs will be much more accurate than 1%, both relatively and absolutely. Here's a 1.00000Vrms 1kHz signal from a calibrator put into CH1 and CH1 of a Siglent 2354X+ that was just turned on with no warmup and no self-cal.
No. Just the probe adjustment alone can have a larger than 1% error. Really, for measuring signal levels an oscilloscope simply is the wrong tool -period-. Argueing you can is like saying you can use a hammer to drive screws into wood
No. Just the probe adjustment alone can have a larger than 1% error. Really, for measuring signal levels an oscilloscope simply is the wrong tool -period-. Argueing you can is like saying you can use a hammer to drive screws into wood
A feature you don't know of in these is the new Probe Check when a probes through input is compared against the probe compensation output and a custom attenuation applied to increase accuracy.
bdunham7, can you be educator on this subject seeing you have your scope turned ON ?
No. Just the probe adjustment alone can have a larger than 1% error. Really, for measuring signal levels an oscilloscope simply is the wrong tool -period-. Argueing you can is like saying you can use a hammer to drive screws into wood
What probe? This is a perfect case for using 1X. And this isn't measuring signal levels, it is comparing them. You can test and characterize any imbalance between channels or probes if you want. But as for hammering screws, did you look at the screenshot? It is obvious that the scope depicted is quite adequate for measuring signal levels. Perhaps like a Pozi-Drive hammer?
Scopes do have their calibration issues, especially offsets with small signals. But in many cases IMO they are the
very best available (at reasonable cost) tool for measuring signal levels. Think high-complexity or high bandwidth RMS measurements.
Well, a true-RMS DMM rated for the frequency range to be measured, doesn't need pre-calibration of anything. It works out of the box to give you repeatable AND accurate results. For high frequencies you use a power meter.
@Tautech: no, an oscilloscope is the wrong tool for the job. For starters: what is the long term drift for the calibrator output? It's not specified in the datasheet of the SDS2000X+. Also the amplitude flatness at low frequencies is +/-1 dB which is over +/-10% error (which is 10 times worse than what the OP requires). Need I continue to educate you further or do you get the point?
Normally in audio stuff you calibrate stuff at 0db at 1khz, done tons of real tapes / cassette tapes cd players, dat tapes and other types
Unless it is specified at other frequencies, you have pink noise and white noise generators too
A scope will be very good at 1x, you don't play with dangerous levels, at 10x or 100x is good to get higher impedances on the probe if 1x create some signal load .... on too sensitive circuits like head amplifiers
The best would be a scope with an integrated signal generator, 2 in 1 instrument, It would be the best for trouble shooting
For me "an old era tech" i would use an scope before a meter, or dual input meter ....
But portability wise a good meter would do fine, i really don't know the Prova brand, maybe they are good at 60k counts on each display / channel
Every one here can argue over this and that, it really depend on the OP need(s) and portability wise
I understand from OP that to goal is to align/match both channels output level one against the other. There is no statement about having to match an exactly measured level. We can assume classic analog line audio levels and frequencies.
Couldn't a simple differential opamp with a single mV sensitive (multi)meter on the output be a solution? Too much output on one side is a positive reading, too much on the other side is a negative reading. The reading is proportional to the difference indicating how much the mentioned pots have to be trimmed. Cheap, simple, small, portable. Do you need more?
Going the why-keep-it-simple-if-complicated-is-so-much-more-difficult-route: Genuine question: Can't a modern scope with math function (A minus B) also calculate and show the difference between the signals on 2 channels? The difference would be the representation of the deviation between the said channels. Any Siglent or Rigol does have this on board I believe. Obviously as 2 different channels on the scope are used these have to be matched first -> feed the same signal on both in parallel and any deviation between the 2 is to be taken into account when measuring the outputs of the cassette tape.
I'm having hard time to understand why a single good true RMS (although not necessary) meter won't work here.
All you need to do is to get a reading of 1.228Vrms (i.e +4dBu) at 1kHz sine wave on both channels. Most likely, this will be the reference level on most pro gear (there are some exceptions depending on the country).
You just have to repeat the process for each channel.
I'm having hard time to understand why a single good true RMS (although not necessary) meter won't work here.
All you need to do is to get a reading of 1.228Vrms (i.e +4dBu) at 1kHz sine wave on both channels. Most likely, this will be the reference level on most pro gear (there are some exceptions depending on the country).
You just have to repeat the process for each channel.
It is quite possible that some adjustments affect both channels (think about left/right balance) so you'll want to see two readings at the same time. IMHO the OP's question is valid and he/she really needs to see two readings in order to do the job efficiently. Otherwise the question would not have been asked. I used to do some audio / telecom work every now and then and the dual channel AC voltmeter I have did come in handy several times.
Well, a true-RMS DMM rated for the frequency range to be measured, doesn't need pre-calibration of anything. It works out of the box to give you repeatable AND accurate results. For high frequencies you use a power meter.
@Tautech: no, an oscilloscope is the wrong tool for the job. For starters: what is the long term drift for the calibrator output? It's not specified in the datasheet of the SDS2000X+. Also the amplitude flatness at low frequencies is +/-1 dB which is over +/-10% error (which is 10 times worse than what the OP requires). Need I continue to educate you further or do you get the point?
I'm not sure how much this argument helps the OP, but my thoughts were that DSO is good enough for the job and since he already has multimeters, expands his capabilities. A pair of reasonably matched inexpensive multimeters might do as well, although the ability of the scope to find issues other than signal level might be an eye opener.
As for your thoughts generally on scopes and signal levels, my statements are based on practical experience not spec sheets. My SDS2354X+ is a year old and the calibrator output measures exactly 1.500V within my ability to measure. It's probably good enough to sub for a DMM-Check. As for meters being accurate out of the box, it isn't relevant to the OP's situation, but when you get into high crest factors your 'out-of-the-box' DMM will fail (to be accurate) quite quickly and worse, you won't realize it. On the scope you'll be both be able to tell and to measure signals with crest factors of 10, 20 or even 30. There aren't many DMMs that will do that.
The problem with measuring accurately is that your measurement system has to be designed to reach certain accuracy levels. Anything else is just anecdotal evidence based on luck. How can you be sure if the OP buys an oscilloscope that it has the same 'accuracy' as yours? You simply can't promise that or even pretend it will be. The specs say that loud & clear. It is similar to buying 1% resistors. A whole bunch of them will be better than 1% and some will be right on the edge. But you don't know what you get upfront and for as long as the resistors are within 1%, you can't claim anything.
For sure an oscilloscope can be used to look at the waveform of a signal and see whether it is very distorted but small distortions are better measured using a distortion analyser.
A feature you don't know of in these is the new Probe Check when a probes through input is compared against the probe compensation output and a custom attenuation applied to increase accuracy.
bdunham7, can you be educator on this subject seeing you have your scope turned ON ?
I don't see the Probe Check feature anywhere. Comparing CH1 1X and CH2 10X there's very, very little difference. The probe does load the calibrator down a bit, makes a about a 500ppm difference. My DMMs load it down even more, which you'd expect. So the calibrator source impedance isn't very low.
Fluke model 45. I don't know if they still make them or not but they've been around a while and I'm sure you can find one on E-Greed. Like most Fluke meters, it's a very good meter and not a hunk of junk.
https://www.tequipment.net/Fluke45.html#description
A feature you don't know of in these is the new Probe Check when a probes through input is compared against the probe compensation output and a custom attenuation applied to increase accuracy.
bdunham7, can you be educator on this subject seeing you have your scope turned ON ?
I don't see the Probe Check feature anywhere.
Really, well OK let's show you some new tricks.
Channel menu> Input attenuation> Probe check......it's in the latest firmware and not sure when it was implemented as can't yet find it in Release notes.
Really, well OK let's show you some new tricks.
not sure when it was implemented as can't yet find it in Release notes.
I'm on 1.3.9R6 and haven't updated because I didn't see anything worthwhile in the release notes. I don't have that feature.
Really, well OK let's show you some new tricks.
not sure when it was implemented as can't yet find it in Release notes.
I'm on 1.3.9R6 and haven't updated because I didn't see anything worthwhile in the release notes. I don't have that feature.
Best your read the V1.5.2R1 release notes again. Sure there's a small bug in the time region setting but it can be worked around with a UTC setting or I have a beta without that bug.