• Please note that Audiokarma will be offline briefly for updates early on Monday morning, September 28th.

Digital multimeter basics: know your test equipment

Ray Gianelli

AK Subscriber
Subscriber
Anyone who's ever seen a pic of my bench knows that I have a lot of test equipment. Most of it is rarely used, but essential when needed. My DMM is by far the most used tool I own, easily used for over 95% or more of troubleshooting. Which is why I spent more on my DMM then I did for my oscilloscope. It's that important. However, for our purposes troubleshooting audio gear calibration isn't necessary.

A DMM is often specified as having 3 1/2, 4 1/2 or even 5 1/2 or more digits. WTF is a 1/2 digit, you say? Good question. The leading (Most Significant Digit AKA MSD) is either a zero or a one. We don't need a really serious meter for audio troubleshooting, but a good quality meter is recommended simply for the trust factor if nothing else.

I'll briefly cover all the functions of a basic DMM. Most are self explanatory.

AC Volts: Good for checking AC into the device under test. Also good for verifying the secondary output on the transformer. And a good test of the DC outputs for ripple; should be in the millivolts if the filters are good.

AC voltage function is also good for measuring the voltage drop across an 8Ω dummy load, with some caveats I'll cover below.

DC Volts: Everything in you audio device runs on DC. So once past the power supply DC is what you're looking for, or possibly finding where you shouldn't. Handy from input to output, and all the other stages as well. Your go to function for troubleshooting a live chassis.

DC Millivolts: This function shines for setting bias and offset, which are both specified in millivolts.

Diode Test: We use the diode function for testing semiconductor junctions in both diodes and transistors. Invaluable for repairing amps.

Continuity: Audible continuity is a not often used but is a lifesaver in certain situations, like the repair I recently did on an NAD:
https://audiokarma.org/forums/index.php?threads/case-history-nad-3400.864486/

Resistance: Verify the resistance of... resistors. If your meter has good resolution you can use the resistance function to help zero in on a shorted component, such as a shorted bypass cap on a power supply rail.

Current: The current function isn't used often. Most of the time we calculate the current by measuring the voltage drop across a known resistance. I have used it to test high voltage capacitors for leakage. Note that you have to put the positive test lead into another jack on the meter. Be sure to move it back before taking any voltage measurements or you'll blow the fuse in the meter, and they're usually something that you won't be able to find locally.

Some meter will have other functions, such as frequency, capacitance or HFE. Personally I prefer my Peak Atlas ESR meter for caps, since it gives me ESR and value in a single test.




It's vital that we not only know our test equipment's attributes but also its limitations. To illustrate that point let's look at 2 multimeters from extreme price points.

First up we have the Harbor Freight 7 function DMM:
7 function.jpg

At $5.99, that's less than a dollar a function!

At the other end of the spectrum we have the Keithley 2015THD:
Keithley 2015THD.jpg


This was a $4,000 meter when new. Fortunately it's of an age where it can be had for around 10% of that. This is a specialized DMM that was made for audio work. Unfortunately its internal sine wave generator isn't of low enough distortion for some of the amps I see. It's perfect for adjusting the distortion in an FM tuner though.

I mentioned above about measuring the voltage drop across an 8Ω dummy load. If you put a 1 KHz signal into an amp and measure its voltage drop across an 8Ω dummy load with both of these meters we'd get 2 completely different readings, with the Keithley being higher than the Harbor Freight model. Why?

This is where you need to know your test equipment's capabilities and limitations. The specs for both meters will tell the tale.

The Harbor Freight specs:
HF DMM.png

You can see that the Harbor Freight DMM's AC response is only rated to 450 Hz. Its response will fall rapidly after that. Now look at the specs for the Keithley:
2015THD.png

This meter can be trusted to give accurate readings across the 20-20,000 Hz range. Now, could I live with only the Harbor Freight if I had to? As long as I didn't ask it to perform tasks it wasn't designed for, yes. Knowing what your test equipment can and cannot do is crucial to getting good results.

Another gotcha that can trip you up is knowing your DMM's input impedance and when it can be an issue. Most DMM's have an input impedance of 10 megohms. It doesn't usually become an issue in solid state amps, but can in tube amps where resistors in the megohm range are more common. Measuring across a 10 megohm resistor with a meter having an impedance of 10 megohms will give you a grossly erroneous reading.

Most of our voltage measurements are going to be well within the capability of our DMM. But what if you're working on high voltage power supplies, such as you might find in an electrostatic speaker? Your DMM, and even more importantly the probes, are probably rated to 1,000 VDC. So how do you measure something in the kilovolt range? Use a high voltage probe, such as this:
Fluke high voltage probe.jpg
The alligator clip goes to ground, and the dual banana goes into your DMM. It essentially forms a voltage divider with the input impedance of your DMM. I have the probe shown above, left over from my days repairing CRT TV's. This probe is good to 40 kilovolts.

A few other accessories that I find indispensable are the oft mentioned banana to minigrabber clips, shown here:
20190406_114833-1.jpg

And my favorite for probing into tight places:
20190406_114807-1.jpg
These came as a pair, and allows you to get into tight spaces without fear of shorting something else, as only the tip is exposed. While it's usually a non issue, they are rated for 300 VDC, while the probes are good to 1,000 VDC. Be aware of these limitations as they are a safety issue.

As always, if I've forgotten something or made an error please post or contact me directly.
 
Register to hide this ad
Nice write up Ray. Use a Fluke 12 at home. Agilent U1242a at work. Like the HV probe. Would love to have that Keithley. Use a HP3468, Fluke 45 at work.
 
I got a few of the HF ones for free with coupons a while ago, ha ha. That's $0 per function.
I need to find some of those mini-grabber clips.
 
I got a few of the HF ones for free with coupons a while ago, ha ha. That's $0 per function.
I need to find some of those mini-grabber clips.
Got a few sets from Amazon. Having more than one set allows you to set the bias on both channels at once, providing you have 2 DMM's.
 
Nice write up Ray. Use a Fluke 12 at home. Agilent U1242a at work. Like the HV probe. Would love to have that Keithley. Use a HP3468, Fluke 45 at work.
You get to use some nice meters at work. What do you do for a living?

I like the Fluke 45. The dual display is nice for checking power supply outputs AC and DC at the same time.
 
I appreciate this post and the knowledge sharing, and I hope you keep doing this. Thanks Ray.
 
I appreciate this post and the knowledge sharing, and I hope you keep doing this. Thanks Ray.
Nice to know it's appreciated! But to be honest, I learned long ago that by sharing your knowledge you will actually increase your knowledge. How? Either by someone asking you something about the topic that you might be fuzzy on, and by researching it you fill in those blanks. Or as happened recently, you'll be wrong and corrected. Admittedly that stings a little, but in the end you're better for it. That's why I always end these tutorials with the request that if you have anything to add or find an error please post or PM me.

In the end it's all about learning.
 
Something worth a mention on those old HV probes, they're only reasonably accurate when working into whatever input impedance they were designed for. Typically 10M for a VTVM, which agrees with many DVM. If you happen to try it on something with a 1M load it will not be accurate. You can re-scale them if you know the value of the resistor and do some math magic though. Lets be honest, nobody wants to do math just to take a measurement unless there is no choice.
 
As mentioned earlier, calibration isn't necessary for the work we're doing. Below are a few shots of my Keithley 2015THD bench meter and my Fluke 177 handheld. The Keithley was calibrated 20 years ago, and the Fluke left the factory without calibration. Fluke only calibrates them and provides a certificate if you pay extra.

With 50 volts input:
20190407_084330-1.jpg

With 10 mV input:
20190407_084851-1.jpg
Certainly good enough for audio work!
 
Something worth a mention on those old HV probes, they're only reasonably accurate when working into whatever input impedance they were designed for. Typically 10M for a VTVM, which agrees with many DVM. If you happen to try it on something with a 1M load it will not be accurate. You can re-scale them if you know the value of the resistor and do some math magic though. Lets be honest, nobody wants to do math just to take a measurement unless there is no choice.
+1000

That probe works fine with my 177, but the Keithley has an input resistance stated as > 10 GΩ until you get to the 100 volt range, where it is the expected 10 MΩ. So the Keithley is a good tool for checking across a 10MΩ resistor in the tube amp example above, but not for using the high voltage probe.

It goes back to what I said at the beginning. You need to know your test equipment's attributes and limitations.
 
Nice write up Ray! One thing most people don’t realize about hand held meters is that most are not rated for AC voltage accuracy above 500Hz (yes, even Ray’s Fluke 117). This becomes problematic in audio work when testing the frequency response of amplifiers where 20KHz is the minimum required and 50KHz is the norm. I discovered this the hard way and now have a few bench meters with much higher ratings as well.
 
Ray I work for a small electronics company who produces temperature controllers, temp sensors, customer builds. Been there since 82. The Fluke 12 will read capacitance up to 10000 mfd. Still using the Tek 922 scope at work. Had 2 of the Tek digital scopes die in 2017. TDS 210 lost the backlight. The other lost the trace. Would't let me take a look at them.
 
Last edited:
Ray I work for a small electronics company who produces temperature controllers, temp sensors, customer builds. Been there since 82. The Fluke 12 will read capacitance up to 10000 mfd. Still using the Tek 922 scope at work. Had 2 of the Tek digital scopes die in 2017. TDS 210 lost the backlight. The other lost the trace. Would't let me take a look at them.
Very cool! At one time in my life I was an instrument and control tech for a large water utility. I learned about devices to measure and control flow, pressure and temperature. Learned about pitot tubes, orifice plates, thermocouples and RTD's.

My boss realized that I could repair a lot of things in house that they used to send out. So he let me take a shot at things first. I had a pretty good success rate, he looked good and I got to stay inside an air conditioned building instead of working outside in the Florida heat. I call that a win win!
 
That's a win for sure. The controllers I test use T/C, RTD, Thermistors as temp and control sensors. Used to test temperature controlled ovens.
 
Back
Top Bottom