What Audio Testing Gear Do You Use?

[tranz]

As a spin-off from another thread I thought it would be intriguing to read what fellow CA-ers currently use or in the past have used for testing equipment of their audio gear, or what they would love to own if the disposable cash was available.

 

There are many professional engineers amongst the CA community so it would provide a great insight into the complexity and type of the testing gear preferred by them.

 

And there are many, like myself, that just like to tinker a bit to find explanations for what they are hearing or explain differences in listening experience.

 

Would love to read about and see pictures of your testing gear.

 

Cheers!

[tranz]

In my amateur curiosity fulfillment I have used the following:

 

Tektronix 475A 250MHz with DMM44 Oscilloscope and 250Mhz probes.

 

 

An AM Radio

 

 

Audio Prism Noise Sniffer

 

 

Trifield 100XE EMF Meter

 

 

Decibel Meter

 

 

 

On an old thread I had provided some more details on some of these tools.

 

http://www.computeraudiophile.com/f8-general-forum/helpful-noise-floor-lowering-tools-17667/

 

Cheers.

[One and a half]

The radio and the Fluke are always at home. The Yokogawa and Fluke 195II I have access to, it belongs to work. For it's price I could buy a NADAC DAC with the power supply option and still have change.

 

The AEMC insulation tester is to test custom length power cables after assembly.

 

 

[Keith_W]

Focusrite 2i2 audio interface. USB to computer, phantom power for microphone.

 

 

Dayton EMM-6 calibrated measurement microphone

 

 

Jaycar SPL meter

[sphinxsix]

I use (it's so good I even got two of these..)

[mansr]

Tektronix DPO2002B oscilloscope:

 

Tascam UH-7000 audio interface (modified for external PSU):

 

Dayton Audio EMM-6 microphone:

 

I hadn't thought of using an AM radio to check for interference though I do have one.

 

Wish list:

- High-end scope

- Spectrum analyser

- Function generator

[Jud]

Multimeter and AudioTools app's SPL Graph tool. The latter graphs SPL over an extended period and saves the result, allowing a reasonably good matching of volume between A and B samples.

 

I very much want to get hold of an UMIK-1 calibrated mic for purposes of speaker location when we move to a new home next year (or at least that is the plan).

[Jud]

Tektronix DPO2002B oscilloscope:

[ATTACH=CONFIG]27341[/ATTACH]

 

Tascam UH-7000 audio interface (modified for external PSU):

[ATTACH=CONFIG]27342[/ATTACH]

 

Dayton Audio EMM-6 microphone:

[ATTACH=CONFIG]27343[/ATTACH]

 

I hadn't thought of using an AM radio to check for interference though I do have one.

 

Wish list:

- High-end scope

- Spectrum analyser

- Function generator

 

MassDrop often offers portable function generators and spectrum analyzers from SEEED. I don't know whether these are any good or have the capabilities you're looking for, but perhaps something to look at.

[mansr]

MassDrop often offers portable function generators and spectrum analyzers from SEEED. I don't know whether these are any good or have the capabilities you're looking for, but perhaps something to look at.

 

If I had money to spare, I'd get Tektronix, Keysight, or other professional gear. Those are, unfortunately, hideously expensive for something that would be essentially a toy. The trouble with cheaper brands is that they often look impressive on paper but fall somewhat short of expectations in reality.

[tranz]

Very cool to see. Keep them coming!

[JohnSwenson]

I don't have fancy pictures so ya'all will have to look them up on the web.

 

Tektronix 2467B scope -- 400MHz, 4 channel. $600 on ebay, $13K originally

HP 3585A spectrum analyzer -- 10Hz to 40MHz 3Hz bandwidth. $900 on ebay originally $32K

DS800C 4GHz sampling scope (USB to computer for display), great for some things, useless for others, but only $800 new.

 

Generic Multi-meter (from Sears actually) used to have much higher street cred ones, but they are all gone now, this $13 meter has lasted through decades of hard use, and honestly does just as good as the "big boys" for most use.

 

Pair of Earthworks QTC-1 mics, 9Hz to 40KHz +- .5dB

 

I also have a Tektronix 2246 scope (100MHz 4 channel) which I bought brand new 30 years ago for $2200 when I got my first job out of college and had some real money to spend. These are available for ridiculously low prices on ebay. (I saw one for $97 that said it was working!)

 

 

I also have two small Tektronix analog scopes that I picked up for $35 for both at a yard sale. I don't use them in the lab but they are great when doing some remote debugging.

 

If someone wants to start out getting a scope I personally would get the used 2246 (as long as it really works) rather than one of the cheap digital scopes on the market today.

 

I don't have time to go over it now, but probes are very important to get right for your scope. I might give a little mini tutorial tomorrow if I have time.

 

I'm in the process of building two very important pieces of test equipment, because they are insanely expensive to buy them.

 

Phase noise measuring setup, should cost about $1k to build, and do better than the $170K commercial ones.

 

Very low jitter sampling system (ADC). Looking at two different versions, one does 24bit at 1MHz sampling, the other is 16 bit at 250MHz sampling. The 1MHz will have a sample jitter in the 150 fs range, the 250MHz one will be in the 300 fs range. Both store data internally then download to a computer for analysis.

 

 

John S.

[mansr]

I don't have fancy pictures so ya'all will have to look them up on the web.

 

Tektronix 2467B scope -- 400MHz, 4 channel. $600 on ebay, $13K originally

HP 3585A spectrum analyzer -- 10Hz to 40MHz 3Hz bandwidth. $900 on ebay originally $32K

DS800C 4GHz sampling scope (USB to computer for display), great for some things, useless for others, but only $800 new.

 

Generic Multi-meter (from Sears actually) used to have much higher street cred ones, but they are all gone now, this $13 meter has lasted through decades of hard use, and honestly does just as good as the "big boys" for most use.

 

Pair of Earthworks QTC-1 mics, 9Hz to 40KHz +- .5dB

 

I also have a Tektronix 2246 scope (100MHz 4 channel) which I bought brand new 30 years ago for $2200 when I got my first job out of college and had some real money to spend. These are available for ridiculously low prices on ebay. (I saw one for $97 that said it was working!)

 

 

I also have two small Tektronix analog scopes that I picked up for $35 for both at a yard sale. I don't use them in the lab but they are great when doing some remote debugging.

 

If someone wants to start out getting a scope I personally would get the used 2246 (as long as it really works) rather than one of the cheap digital scopes on the market today.

 

I don't have time to go over it now, but probes are very important to get right for your scope. I might give a little mini tutorial tomorrow if I have time.

 

I'm in the process of building two very important pieces of test equipment, because they are insanely expensive to buy them.

 

Phase noise measuring setup, should cost about $1k to build, and do better than the $170K commercial ones.

 

Very low jitter sampling system (ADC). Looking at two different versions, one does 24bit at 1MHz sampling, the other is 16 bit at 250MHz sampling. The 1MHz will have a sample jitter in the 150 fs range, the 250MHz one will be in the 300 fs range. Both store data internally then download to a computer for analysis.

 

 

John S.

 

Interesting that you mainly use analogue scopes. Not that there's anything wrong with that, although I really like some of the features possible with a digital scope, especially all the triggering options.

 

I agree about cheap digital scopes (Rigol etc.) being best avoided.

[Jud]

I don't have time to go over it now, but probes are very important to get right for your scope. I might give a little mini tutorial tomorrow if I have time.

 

Look forward to it, if/when you have time.

 

I'm in the process of building two very important pieces of test equipment, because they are insanely expensive to buy them.

 

Phase noise measuring setup, should cost about $1k to build, and do better than the $170K commercial ones.

 

Obviously very much looking forward to this, and to licensing to interested parties, so we all get the opportunity to look at lots more measurements than we've got today.

 

Very low jitter sampling system (ADC). Looking at two different versions, one does 24bit at 1MHz sampling, the other is 16 bit at 250MHz sampling. The 1MHz will have a sample jitter in the 150 fs range, the 250MHz one will be in the 300 fs range. Both store data internally then download to a computer for analysis.

 

I'm hoping it's the early hour that's making me slow to grasp what this would be used for. John or anyone, intended uses?

[semente]

I am currently using a Dayton Audio iMM-6 Calibrated Measurement Microphone to measure system response at the listening spot and fine-tune loudspeaker positioning.

 

 

But I am not convinced that the Nexus 5's sound card is up to scratch and will probably replace the iMM-6 with a more conventional computer-operated measuring system.

 

 

Left speaker, continuous pink noise, 1/3 octave smoothing

[PeterSt]

Interesting that you mainly use analogue scopes. Not that there's anything wrong with that, although I really like some of the features possible with a digital scope, especially all the triggering options.

 

You can multiply the bandwidth capacity of a digital scope / analyzer with a factor of about 300 in order to get the same possibilities of an analogue (storage) scope. So if you want to compare a 1GHz analogue scope (the highest bandwidth Analogue I think) with digital, you're up for a 300Gs.

Sell your boat first is not sufficient.

 

With an underrated digital scope (like a 40Gs Tektronix I ever had and which so-called could measure jitter up to 200fs) you only get "math" which isn't even reliable (to the creativity of the designer).

[PeterSt]

Both store data internally then download to a computer for analysis.

 

Brrr ... something I hate !

Or put differently : audio analysis is to be real time. Move a cable and such and see what happens.

Well, at least most often it is about such a thing. Think unexpected result and see whether all is (positioned etc.) right.

 

Peter

[mansr]

You can multiply the bandwidth capacity of a digital scope / analyzer with a factor of about 300 in order to get the same possibilities of an analogue (storage) scope. So if you want to compare a 1GHz analogue scope (the highest bandwidth Analogue I think) with digital, you're up for a 300Gs.

Sell your boat first is not sufficient.

 

With an underrated digital scope (like a 40Gs Tektronix I ever had and which so-called could measure jitter up to 200fs) you only get "math" which isn't even reliable (to the creativity of the designer).

 

Both analogue and digital scopes can do some things better or cheaper than the other. Digital scopes can also do things that are simply impossible with an analogue one, and even doing it poorly is often better than not doing it at all.

[semente]

Forgot to mention the VDH polarity checker:

 

[JohnSwenson]

I don't have fancy pictures so ya'all will have to look them up on the web.

 

Tektronix 2467B scope -- 400MHz, 4 channel. $600 on ebay, $13K originally

HP 3585A spectrum analyzer -- 10Hz to 40MHz 3Hz bandwidth. $900 on ebay originally $32K

DS800C 4GHz sampling scope (USB to computer for display), great for some things, useless for others, but only $800 new.

 

Generic Multi-meter (from Sears actually) used to have much higher street cred ones, but they are all gone now, this $13 meter has lasted through decades of hard use, and honestly does just as good as the "big boys" for most use.

 

Pair of Earthworks QTC-1 mics, 9Hz to 40KHz +- .5dB

 

I also have a Tektronix 2246 scope (100MHz 4 channel) which I bought brand new 30 years ago for $2200 when I got my first job out of college and had some real money to spend. These are available for ridiculously low prices on ebay. (I saw one for $97 that said it was working!)

 

 

I also have two small Tektronix analog scopes that I picked up for $35 for both at a yard sale. I don't use them in the lab but they are great when doing some remote debugging.

 

If someone wants to start out getting a scope I personally would get the used 2246 (as long as it really works) rather than one of the cheap digital scopes on the market today.

 

I don't have time to go over it now, but probes are very important to get right for your scope. I might give a little mini tutorial tomorrow if I have time.

 

I'm in the process of building two very important pieces of test equipment, because they are insanely expensive to buy them.

 

Phase noise measuring setup, should cost about $1k to build, and do better than the $170K commercial ones.

 

Very low jitter sampling system (ADC). Looking at two different versions, one does 24bit at 1MHz sampling, the other is 16 bit at 250MHz sampling. The 1MHz will have a sample jitter in the 150 fs range, the 250MHz one will be in the 300 fs range. Both store data internally then download to a computer for analysis.

 

 

John S.

 

I forgot two more:

 

Intronix Logicport 34 channel logic analyzer. It connects over USB for display. Very useful for digital analysis. A pain to setup, all my recent stuff is dense SMD stuff, I have to glue a pin header onto the board and solder thin little wires onto the chips (not easy for 0.5mm pitch chips), but when nothing else can do, it is well worth it.

 

LeCroy Mercury T2 USB protocol analyzer. It can be quite useful when a USB interface is not working right, but it is a pain in the neck to pour over huge packet logs trying to find where something went wrong.

 

John S.

[JohnSwenson]

Interesting that you mainly use analogue scopes. Not that there's anything wrong with that, although I really like some of the features possible with a digital scope, especially all the triggering options.

 

I agree about cheap digital scopes (Rigol etc.) being best avoided.

 

So far the combination of logic analyzer and analog scope has been quite useful. Only a couple of times I have wanted a modern digital scope, not enough to spend the money on one.

 

One interesting combination is to use the trigger out from the logic analyzer into the trigger in of the analog scope.

 

John S.

[JohnSwenson]

I'm hoping it's the early hour that's making me slow to grasp what this would be used for. John or anyone, intended uses?

 

The ADC system is a frontend for any digital measurement system such as the oft mentioned AP system.

 

Since we are talking about measuring DACs with jitter in the sub ps range, we need an ADC that has even lower sample jitter in order to try and find out if small changes in the DAC jitter produce measurable changes in the output.

 

By making it a frontend to a computer you are free to use any analysis tools you want rather than being stuck with the standard tools built into things like the AP. For example you can use things like a wavelet transform to find out how the spectrum varies with time.

 

I really hope that as I learn what to look for I can start finding some correlations between hearing the "it sounds more real" and what is actually coming out on the wire.

 

John S.

[jabbr]

I

 

I'm in the process of building two very important pieces of test equipment, because they are insanely expensive to buy them.

 

Phase noise measuring setup, should cost about $1k to build, and do better than the $170K commercial ones.

 

Very low jitter sampling system (ADC). Looking at two different versions, one does 24bit at 1MHz sampling, the other is 16 bit at 250MHz sampling. The 1MHz will have a sample jitter in the 150 fs range, the 250MHz one will be in the 300 fs range. Both store data internally then download to a computer for analysis.

 

 

John S.

 

Rutgers' phase noise circuit, or another? If another, what?

 

and essentially you are building your own 24 bit scope but 1mhz gets you to 150fs?

[JohnSwenson]

Earlier I promised to write a short piece on scope probes, here it is. If you are an experienced test equipment user, you can skip this, it is for those who just bought their first scope and have no idea what those things on the ends of the cables are.

 

A scope probe is a device for getting the signal in your Device Under Test (DUT) into the scope with the least amount of distortion possible. It seems like a simple task, but it turns out to be not so simple. I'm not going to go into all the things that can go wrong, this is supposed to be a simple article after all. But the gist is the scope probe has been found to be the best way to do this for most applications.

 

First up, how to use one. The probe will have a BNC plug on one end, this plugs into the scope, the other end kind of looks like a pen, with a sharp metal tip. There is also a short wire with an alligator clip coming out of the middle, called the "ground clip". Hooking this ground clip up correctly is one of the most important topics for beginners, so I will spend some time on it.

 

The most import thing to ALWAYS remember is that the ground clip is connected to the chassis of the scope, which is connected to the ground pin of its AC plug. Whatever this ground clip is connected to is shorted to the AC ground pin. If this is something other than a ground point in your DUT, very bad things can happen, frequently involving sparks and smoke (maybe even fire). Damage to either the DUT or scope is also highly likely.

 

So be very careful what you connect that ground clip to.

 

You have to connect it to something, if you don't you won't get a trace that has any bearing on what you think you are measuring, it just has to the RIGHT thing.

 

For audio devices the shell of the RCA jacks are usually a good place. Maybe the metal chassis, but maybe not. Power amps get dicey, neither of the loudspeaker wires my be connected to ground so just connecting the ground clip to the black post may result in the infamous sparks and smoke.

 

It's usually a good idea to connect to a ground point near what you are testing, that is why the ground clip wire is short. (in some circumstances it is way too long, I'll get to that later).

 

When testing modern PC boards with lots of parts you need to make sure the clip doesn't touch something it is not supposed to touch (see sparks and smoke above).

 

The above is not to try and scare you from using a scope, just to impart a sense of deliberateness to it, take the time to figure out where a good ground point is before connecting that clip. If you don't know, you can try using a DVM to see what kind of voltages exist in the circuit. If your meter shows voltage between the chassis and what you want to use as a ground point, it probably is NOT a good ground point to use with a scope.

 

Now that I have scared everybody, lets talk about different probes, and what makes them different. The most important difference is 1X or 10X. You see this a lot but many people don't know what this is. Remember I was talking about the probe being a way to get the signal into the scope with the least distortion possible, 10X is all about that.

 

If you just feed a wire into a scope the input impedance and capacitance of the scope and the wire itself will distort the signal. At audio frequencies it is not too much, but when dealing with digital systems the distortion can be huge. A long time ago it was figured out that s a simple RC network can balance out this distortion, BUT it results in the attenuation of the signal. This is what the 10X is about, the RC network in the probe attenuates the signal at the tip by 10 times. Or to put it another way the signal at the tip is 10 times greater than the signal going into the scope (hence the 10X). It does this to radically decrease the distortion of the signal.

 

One problem with this scheme is that every scope has a slightly different input circuit which means a one size fits all probe will not work. So all 10X probes will have a little hole somewhere and a plastic adjustment tool so you can tweak the network to make it work with your scope. In order to facilitate this all scopes come with a "calibrator" output which outputs a square wave, you hook the probe up to this and adjust the probe until the edges of the square wave look clean, sharp edges, no ringing or overshoot.

 

BTW the 10X is not some constant of the universe, it is done that way to make it easier for humans to do the math in their heads. Which brings up doing the math in your head. This is always a pain in the neck and has lead to MANY a misinterpretation of measurements.

 

To get around this scope manufacturers have come up with several ways to do it for you. One method gives you a switch on the scope, you set it for 1X or 10X depending on what type of probe you have. The scope then changes its readouts so the voltages match that at the tip of the probe, so you don't have to worry about it. Tektronix has a neat way to deal with this, their probes have an extra spring loaded pin in the BNC plug that contacts a ring around the jack, when this pin is grounded the scope automatically changes the readouts, this means you never have to worry about the situation, 10X probes automatically tell the scope what they are. BUT you have to have the special Tektronix probes, if you use a generic probe without the pin, you HAVE to do the math in your head.

 

You can get probes in either 1X or 10X. If the 10X is so much better, why ever use a 1X? The 10X attenuates the the signal by 10X, so if you are measuring a very low level signal a 10X probe can leave it buried in the noise of the scope input amplifier. A 1X probe may let you see such a signal where a 10X would not. In addition you just may not want to deal with the mental gymnastics of a 10X probe if your scope doesn't have a way to deal with it for you. If you are dealing with audio signals the 10X probe is not usually necessary, so you might just want to use a 1X.

 

Some probes have a switch that lets you set the probe to be either a 1X or a 10X, with the Tektronix system these are great, but for other scopes it is way to easy to forget which way the probe is set and and get your measurements way off because of it.

 

Scope probes don't last forever, they are moderately fragile devices, so take care of them. When you buy a used scope, assume the probes that come with it are useless, just buy some new ones and don't even think about using the used probes, you have no idea how good they are.

 

There are many cheap probes on places like ebay, I don't recommend them, they are usually very poor performers and will just mess up your measurements but may be all right for just audio measurements. My preferred place to buy probes is probemaster.com, they have very good probes at reasonable prices. They even have Tektronix compatible probes with that extra pin (these models have RA in their part number).

 

One more important tip, many probes come with a spring loaded clip the pushes over the probe body, get extras of these, you always loose them!

 

Remember I was talking about the ground clip wire? For fast digital signals that short wire is way too long, it will radically distort the signals, EVEN with a 10X probe. There is a neat solution for this on many modern probes, it is a little gold spring that fits over the top of the probe and has a little stiff wire sticking out that will be the same length as the probe tip. You find a ground via on the PCB near where you want to probe, stick this pin in the via and then move the tip around to the place you want to probe. The spring gives you a few mm of movement which is usually enough to find a ground via on a digital board. This gives you a ground wire a couple mm long, which completely cleans up the distortion caused by the ground wire length.

 

Well that was a little longer than I wanted it to be, but I hope this helps some people starting out with scopes.

 

John S.

[tranz]

Thank you for sharing JohnSwenson!