Multimeters: so alike, yet so different

I've already published a comparison of two multimeters that differed greatly in both price and capabilities.
Today I'm reviewing another pair, with much more similar specifications, but still noticeably different from each other.

I deliberately chose a pair of multimeters with similar functionality but some differences, and decided to compare them.
Unfortunately, I have to disappoint you: there won't be any precise metrological measurements, at least not in this review. Still, I'll try to check their accuracy, and once I can get access to a metrology lab, I'll test them more properly.

This is the second review in a series about multimeters and everything related to them. There'll also be a third review, probably the last in the series.
That is, if the Chinese seller ships my order.

As I go along, I'll explain some of the ways multimeters differ from one another. Maybe it'll help someone choose the right instrument.
Let me say up front that I don't consider myself an expert on measuring instruments, so there may be some mistakes. I'll describe things from a user's perspective. If you spot any obvious blunders, please correct me or add a clarification.

From time to time, I'll highlight certain terms or features in the text and explain what they are and why they matter.

First, as always, let's look at the packaging.
Both multimeters arrived in almost identical boxes; the only visible difference is the print quality.
HK68B
HONEYTEK HK68B Handheld Digital Multimeter, link to the product in the store, price: $42.24




The UT61E packaging definitely looks nicer.
UNI-T UT61E LCD Digital Multimeter, link to the product in the store, price: $54.14




There is also a mention of the different model variants here. I ordered the version with the highest measurement accuracy in the UT61 series.
Full description of the differences.
The UT61B, UT61C, UT61D, and UT61E can be connected to a PC,
The UT61A has no PC interface, but it does have a non-contact AC voltage measurement function and can measure transistor hFE.
The UT61A, UT61B, UT61C, and UT61D have a display backlight
The UT61B and UT61C can measure temperature
The UT61D and UT61E measure true RMS values (True RMS)
The UT61E has higher measurement accuracy than the other models


Even the packaging is almost identical in size.


The HK68B comes in a plastic tray containing everything included in the package.


The package includes:
Multimeter
Test leads
Temperature probe
Manual


The manual looks like it was simply photocopied and stapled together. But at least it's in English, which is nice :)


This table shows how this model (HK68B) differs from the other models in the series.
The most “advanced” version has a higher count (6000 versus 4000) and a USB port, but its stated accuracy is exactly the same as that of the other multimeters in the series.


First digression.
Number of counts.
This is the maximum value shown on the meter's display. Obviously, the more the better, but there are a few things to keep in mind.
Meters usually come with the following display ranges (roughly speaking):
1999
3999
5999
22000
40000
50000
60000
80000 (UT70D)

It's worth explaining that the first three options will usually have roughly the same measurement accuracy. For example, a voltage of 12.34 V will be displayed the same way on all three meters.
The others can display a more precise result (assuming, of course, that the meter's ADC is designed for that measurement and is working properly), for example 12.345 V.
But there's a catch.
For example, if you measure 34.56 V with a meter that has a 6000-count display, a meter with a 22000-count display will show exactly the same result, because 34.567 is greater than 22.000 and the meter will switch to displaying 034.56 (the leading zero is shown for clarity).

However, remember that the precision with which a meter can display a result and the accuracy with which it can measure are not the same thing.

I'll show this with a real example later. For now, I'll just say that personally, I'd choose either 6000 or 40000 counts, though 4000 is enough for most applications. The 1999-count option is already quite outdated and much less common than it used to be.

Meter accuracy.
When choosing a meter, don't just look at the accuracy in %, but also at how many counts the last digit can vary by.

For example, here are two meters with the following stated accuracy for measuring:
DC voltage — 4/40/400 V (±0.5%+3)
DC voltage — 4/40/400 V (±0.5%+8)

In this example, the first meter is better.

The package included a fairly detailed manual. I've hidden a few photos under the spoiler.
Manual.
As funny as it may sound, sometimes it can tell you something useful, such as little details about how the meter works (if it's a relatively complex one).
So for beginners, I'd recommend looking for either a meter with an English manual or a manual online. It's best to find the manual first, then buy the meter. For experienced users, this usually doesn't matter.

The HK68B came with test leads and a temperature probe.
The test leads are pretty standard, among the simplest and cheapest available.
The probe is a little inconvenient. Personally, I prefer probes with a plug rather than a pair of contacts for connection. The plug type makes a more reliable connection.


Test leads.
These leads are rated according to their safety category and carry the corresponding markings.
CAT II 600 V | 4000 V impulse voltage
CAT II 1000 V | 6000 V impulse voltage
CAT III 600 V | 6000 V impulse voltage
CAT III 1000 V | 8000 V impulse voltage
CAT IV 600 V | 8000 V impulse voltage
CAT IV 1000 V | 12000 V impulse voltage

The most common ratings are
CAT III 1000 V
CAT IV 600 V
Despite its 600 V rating, the second option is safer than the first.
I found an image online that explains where each category is used.


There’s actually a pretty good article in Russian from Fluke: *Multimeter Safety Basics*.
But it’s in PDF format, so I’ll attach it to the supplementary materials.

The safety level of the test leads supplied with inexpensive multimeters is a bit of a gray area.
The thing is, good-quality test leads cost about as much as an inexpensive multimeter, so few manufacturers include them—unless, of course, you’re talking about well-known brands like the Fluke mentioned above.
Here’s a photo from another review of mine, where I compared typical supplied leads with some relatively good-quality ones I bought separately.
I don’t think I need to point out which is which.


Even inside the plugs that go into the multimeter, good-quality leads have additional insulation.
These leads are rated CAT III. Below, I’ll show the leads supplied with the UT61 multimeter, which are marked CAT IV, so you can go back and compare them.

Aside from electrical insulation quality, there’s another consideration: whether the cable insulation gets stiff in cold weather. It usually does, at least on all the supplied leads I’ve come across. Even the leads I use get stiff in the cold.

That’s why I don’t see much point in comparing which supplied leads are better or worse: the leads included are usually the most basic kind.
No one is going to include a set of leads costing $10–20 with an instrument that costs just $50–60.


The instrument has a shock-resistant design, with a rubber boot protecting the plastic case.
This design is handy for field work, but less so at home, since it usually makes the instrument bulkier.


Most multimeters have a kickstand for upright positioning. The most compact models usually don’t have one, as it’s generally omitted to keep them small.
Kickstand
When choosing a multimeter, it’s worth checking the kickstand design.
The kickstand should:
1. Lock securely in at least one position—the folded position. Ideally, it should lock in both end positions.
2. Keep the instrument stable, preferably at a fairly shallow angle to the table. For example, I recently reviewed a Mastech multimeter and complained about its awkward kickstand design, even though the instrument itself wasn’t cheap.


Power.
This is another very important consideration.
Most commonly, you’ll find instruments powered by a 9 V battery; AA or AAA cells are less common. Fewer still have rechargeable batteries (usually the more expensive models) or button cells (usually cheaper or specialized models).
A 9 V battery is a poor choice in terms of convenience: it has less capacity and costs more. But if you use the instrument infrequently, that’s not a big deal. A good-quality 9 V battery lasts me about two years.
But instruments powered by low-voltage cells have their drawbacks too. They usually provide a lower voltage at the probes in diode-test mode, which can sometimes be a problem or an inconvenience.

In my opinion, both options have their place, but a 9 V battery will give the instrument a shorter battery life.

This instrument is powered by a 9 V battery. The battery is not included.
The connection is standard for many instruments: wires with connectors. When closed, the cover presses the battery against a glued-on rubber pad, so nothing rattles around inside.
The cover is secured with a machine screw, not a self-tapping screw. That’s good, since a self-tapping screw can lose its grip over time.


Display.
This is a whole topic in itself.
A meter's display should not only be informative, but also easy to use.
As far as information goes, it's simple: the display usually shows everything you need when using the meter (although there are sometimes a few minor quirks). Ease of use, though, deserves a separate discussion.
1. Ideally, the display should have large digits, which makes it easy to read at a glance.
2. The display should have good contrast and be easy to read from different angles, since the meter isn't always positioned conveniently.

Oddly enough, large digits are often found on inexpensive meters, while their branded counterparts are usually more modest in this respect. But in general, digit height can also just be a matter of getting used to.
The digits on the HK68B are about 21 mm high.


The probe sockets are all very familiar and almost identical.
A terminal for measuring current up to 10 A, one for up to 400 mA, a common terminal, and an input for measuring voltage, frequency, capacitance, etc.
The simplest design: holes with contacts inside. But some meters now have shutters that cover unused sockets and retract when you turn the function selector.
In my opinion, this is very handy for beginners—and for everyone else too, though perhaps a little less so.


The function selector and buttons for additional controls.
I'll digress briefly here as well.
Function selector.
Meters generally fall into two categories: those with manual range selection and those with autoranging.
In manual mode, you select the measurement range yourself. In autoranging mode, the multimeter's processor does it for you, but you can still select a range manually using the dedicated Range button.
The principle is roughly like the difference between a manual and an automatic transmission. There are fans of both types.
Manual selection may feel more familiar to many people: it's straightforward—you choose a range and work with it.
Autoranging can sometimes be a little annoying because switching takes time while the meter cycles through the ranges it needs.
The switching principle is quite simple: the meter detects an overload and cycles through the ranges until it can display a valid reading without overloading.
It always starts at the bottom. That is, if you're checking a 100 Ω resistor, the meter will settle on the right range almost immediately, unlike when checking a 10 MΩ resistor. The same applies to voltage measurements.
Overall, it's hard to say which is better. One less obvious advantage of autoranging is slightly less wear on the selector.

The additional-function buttons are located above the function selector.
Func — switches between measurement functions within the same selector position, for example AC/DC or resistance/continuity, etc.
REL — Relative measurements (for example, you can short the probes, press the button, and the probe resistance will then be excluded from the reading). A very handy function.
MIN/MAX — Measures maximum and minimum values. I don't use it very often.
Range — I described this mode above: manual selection of measurement ranges.
Hz/% — Measures frequency or duty cycle.
HOLD — Holds the reading on the display.
Backlight button.


Although I was somewhat skeptical about the meter itself, I have to give it credit for its fairly high-quality display, with good viewing angles and excellent contrast.
The meter also has a backlight. Personally, I consider this a secondary feature: if I can't see the reading on the display, I probably won't be able to see the components I'm testing either.
Also, if the meter is powered by a 9 V battery, the backlight drains it quite quickly. On the plus side, the backlight turns off automatically (if I remember correctly, after 15 seconds).


The second beauty is the UT61E multimeter.
The packaging is exactly the same as the previous multimeter's.
This is what's known as the “economy” version. Sometimes the package includes a bag or case for storing the multimeter. That can be handy if you use it on the go, but for home use it's usually an unnecessary expense. Still, if you rarely use the instrument, a bag or case won't go amiss.
Some compact instruments are designed with a top cover that protects the selector switch and display.
It's hard to recommend one particular option; everyone chooses what works best for them.


This multimeter's package contents are a little different.
Multimeter
Test leads
Cable for connecting to a computer
Adapter for measuring transistor and capacitor parameters
Manual
Warranty card.


There's not much to say about the manual: it's almost entirely in Chinese, so it's of limited use to our readers.
The test leads are better here, and longer too. The previous multimeter's leads were about 90 cm long; these are a full meter, with a little extra to spare.
The plugs that go into the multimeter have a slightly unusual shape. I'm more used to L-shaped ones, but it doesn't really matter.
The test leads are rated CAT IV, although they look less robust than even my separate CAT III leads. Still, they're clearly more convenient and better than the ones supplied with the previous multimeter.


The computer connection uses an old-style cable for a COM port. There seems to be a USB version too—the case even says RS232C(USB)—but you can only connect it to USB using a converter :(
The measurement adapter can be used with both SMD components and through-hole components.
Unfortunately, this version of the instrument can't measure transistor parameters. These days, though, it's easier and better to use the well-known Markus tester than to check transistors with a multimeter.


This instrument really is a beauty. It looks great. Let's just say it's a pleasure to hold; it doesn't feel like it was hacked out of a solid block of rubber and plastic with an axe.
The previous instrument looks much cheaper, unfortunately, but that's how it is.




This instrument also has a stand for upright positioning.


It also runs on a 9 V battery.
But the designers clearly overcomplicated the battery compartment a bit.
It's designed so that you first put the battery into the holder, then insert the holder into the multimeter. This prevents the battery from being installed with the wrong polarity.
But what a clumsy design. Installing the battery is genuinely awkward, though fortunately you don't have to do it often.
By the way, this instrument has no backlight, so you'll need to change the battery even less often.
One downside is that the cover is secured with a self-tapping screw rather than a machine screw. I don't understand why they didn't use a machine screw.


The display here is a little different from the previous multimeter.
1. The characters are shorter (about 13 mm), partly because the meter itself is slightly smaller, while the characters are slightly larger.
2. It has what’s called a “dynamic” bar graph. It’s quite handy because it shows changes in the measured value much faster, albeit with much lower precision. If a multimeter has one, that’s a definite plus.
3. The display has lower contrast; you’ll see this later during testing.


The terminals are arranged a little differently, but they’re otherwise identical to those on the previous meter.
The only difference is that this meter measures low currents only up to 220 mA.


Operating the meter.
This one also uses six buttons, but they handle more functions, and the operating principle is somewhat different.
Hold — freezes the reading
Range — manual range selection
REL — relative measurement mode
Peak — this button wasn’t on the previous multimeter. When this mode is enabled, the meter displays the peak value rather than the RMS value of the signal.
The yellow and blue buttons are function keys that activate the mode marked in the corresponding color on the selector dial.

Speaking of the selector: it gets a solid 5 out of 5. The action is smooth and quiet, with a distinct click into each position.


And here are a few photos with a matchbox for scale.
The photos clearly show that the UT61E is smaller and more compact, and it looks neater, too.
But you can also see that the display is a little harder to read.


It’s almost the same size as my old Mastech 890.


The stand angle is the same on all three multimeters, so switching from the familiar Mastech will feel natural.


A little about the meters’ operating modes.
I’ll hide this section under a spoiler, since it’s not particularly interesting and is mostly just for general information.
Operating modes


A brief introductory test, with a comparison thrown in.
In this test, I'll compare some of the meters' characteristics, such as the voltage and current at their terminals in different operating modes.
I'll take measurements by comparing the readings from the two multimeters, which makes things easier to see, and explain the differences as I go.

The second test is more important. It provides a fairly simple way to check how accurately the instrument's internal voltage reference (VREF) is calibrated.
I'll use a well-known little board for the test. I've already reviewed it and recommend buying one: it's very useful and relatively inexpensive. It also gives me more confidence in the measurement results.


For the next test, I used a selection of my components with fairly high specified accuracy (for components, that is). There are more accurate ones out there, but this is what I had on hand.


But first, I noticed an oddity with a fairly serious multimeter from another review of mine: the Mastech MS8240D.
The issue is easy to see when measuring low-capacitance capacitors :(
The instrument always displays two digits after the decimal point, even though it has a resolution of 22000 counts.


I put the measurement data into a table.
For the Mastech MS8240D, two measurement values are shown.
The first is the measured value; the second takes the instrument's measurement error into account.
Interestingly, although the HK68B's display tops out at 4000, it can show values above 8000 in capacitance measurement mode.


Unfortunately, the test socket only lets you check capacitors, although there's a version of this instrument that can test transistors.
I first tried to test a transistor, but after an unsuccessful attempt and failing to find how to enable the corresponding mode, I decided to check the manual. Alas, this instrument can't do that.


Testing DC current measurement accuracy.
I kept this test simple. Knowing the readings from the Mastech MS8240D for specific values, I set the current so that its readings matched the values I got from the metrology measurements. The test does have some measurement error of its own, of course, but it's more than adequate for measuring current.
The set current values, in order:
1 mA, 10 mA, 100 mA, 200 mA, 1 A, 2 A.
The last two tests have no measurement-error correction.


Here are some more tests I ran to evaluate one of the features of the meters under review: TrueRMS measurement.
TrueRMS
This function lets you accurately measure voltage and current with non-sinusoidal waveforms.
A basic meter usually rectifies the input voltage/current to get the peak value, then divides that value by 1.42 (the ratio between the RMS and peak values of a sinusoidal signal).
This method works very well for measuring sinusoidal signals, but is completely unsuitable for measuring signals with other waveforms, such as triangular, sawtooth, or square waves.
To measure a signal accurately, you first need to integrate it properly (convert it to its mean value), and only then calculate the result.
By the way, the simplest and very accurate method is a thermoelectric converter: heat an element and measure its temperature. The higher the temperature, the higher the voltage. In a sense, this is “hardware” TrueRMS.

I wouldn't say this function is essential or that you can't live without it, but if a meter has it, that's definitely a good and useful thing.

I'll hide the rest of the tests behind a spoiler, but first I'll show the difference between a meter with a 4000-count display and one with a 22000-count display.
As you can see below, when measuring voltages up to 400 mV, the basic meter still measures in mV, while its more accurate counterpart requires switching to a range for measuring voltages up to 600–1000 V.


Lots of tedious—very tedious—photos and measurements.

Frequency measurement.
A useful feature, but I wouldn't call it very important. Of course, it depends on what you use the multimeter for, but I've only needed it a few times over the years.
The thing is, early multimeters didn't measure frequency very accurately. I may be mistaken, but I think they used frequency-to-voltage conversion and then measured that voltage.
Modern multimeters use a more accurate method of measuring frequency, so these tests are more of a formality.
For the test, I first applied 4 MHz and 8 MHz signals, and the measurements showed no problems. A difference in the last digit is also normal, so all the multimeters passed this test without any issues.
But when I applied a lower-voltage square wave from an analog output (that is, one that was actually synthesized), I got some ambiguous results.
At first, all the multimeters showed “the weather on Mars,” but after tweaking the generator settings a little, the UT61 started working correctly. The others weren't so straightforward, and it took some fiddling to get accurate readings.
The UT61 won here too: it started showing correct readings sooner than the other two multimeters, although logically it should have behaved more or less like the Mastech. But things aren't that simple.


That wraps up the tests. Now on to the most interesting part: taking them apart.
Especially since I have two patients on my bench today, not just one :))))

Also, for ethical reasons, and so as not to traumatize the delicate psyches of multimeter rights advocates, I'll hide this part behind a spoiler.


And here are comparison photos of the insides of both multimeters under review.







Right at the end, I remembered that I'd forgotten to check how it works with a computer.
Using it with a computer.
In 95% of cases, maybe even 99%, this isn't essential, but there's at least 1% of the time when it can be very useful. For example, when using the multimeter as a logger. And I wouldn't put the option of turning it into a huuuuge display on the back burner either.
So it's not essential, but in some cases it's almost indispensable (some multimeters have a built-in logger, but they still end up being connected to a computer).

A cable is included, and it connects with this simple little procedure.
Remove the insert from the top of the meter and replace it with an identical one that has a cable.
Since the other end of the cable has a COM connector, I decided to make the test a little more challenging and connect it through a COM-to-USB converter. My computer has COM ports, but they're becoming less and less common.
But don't confuse a COM-to-USB converter (more precisely, an RS232-to-USB converter) with an RS232TTL-to-USB converter. The former has a negative-voltage generator for proper COM port operation (although that's not used here, since the cable only works in one direction). Also, USB-to-RS232TTL converters have inverted signals.
Still, I don't see any particular problem with connecting an RS232TTL-to-USB converter either; you'll just need to use a soldering iron :)


First, I tried running the software for this meter on my main computer, which runs Windows XP, and got another blue screen.
In my previous review, I complained that this software wouldn't work with the Mastech, and suspected it was because the software wasn't made for it.
I tried running it on a Windows 8.1 tablet, but the software displayed one measured value and then reported an error as well. :(


I didn't give up there. I went looking for alternative software for that multimeter, and found some.
It's less functional than the original, of course, but it works and lets you display the image full-screen. I'll include this program in the supplementary materials.


Since the original software wouldn't work, I started packing the multimeter back into its box, feeling a little disappointed.
But then I unexpectedly found a CD in there. I have no idea how I missed it at first.


Naturally, I immediately checked what was on it (on a computer without a disk drive, this is a real hassle—you have to cobble together the whole setup again. When will software finally come on flash drives?).
The disc had version 4.01 of the software; before that, I had tried version 2.0.
I installed it, launched it, and everything worked perfectly. So, be careful when unpacking :)
I’ll also upload everything that was on the disc to the additional materials section.


So, what can I say in conclusion?
Honestly, I liked the UNI-T UT61E. It’s a good, solid multimeter, though it does have a few drawbacks.
Its pros: excellent measurement accuracy, a convenient design, a very handy mode selector, and the ability to connect it to a computer.
Its cons: it could offer more features if it made full use of the installed processor. The 9 V battery and lack of a backlight are only minor drawbacks, since they’re not deal-breakers, though I would have preferred it to run on AA batteries.
250 V fuses.

The HONEYTEK HK68B was a bit of a disappointment, but first, the pros:
Good voltage and current measurement accuracy, and True RMS works correctly, but unfortunately only over a very narrow frequency range. It comes with an external temperature probe, has a large, high-contrast display, and a sturdy case (thanks mainly to the rubber boot), plus a backlight.
The cons: capacitance measurement accuracy is clearly lacking. The frequency range for AC voltage and current measurements is limited to 600–1800 Hz. It also runs on a 9 V battery. In my opinion, this meter is a bit overpriced.

My take: in my search for the “ideal” multimeter, the UT61E came close. I liked its relatively compact case and well-balanced feature set, along with its high measurement accuracy. As I mentioned above, though, it does have a few minor drawbacks. But if I had to choose between the HK68B and the UT61E, I’d definitely pick the latter.
I’m not saying the HK68B is bad; I just think you can find better options for the same price.
What’s missing from the UT61E? Maybe inductance and ESR measurements. I guess I’m asking for a lot now :)

That’s about it. I hope my review helps you choose a multimeter, whether for work or “for home and family.” I’ve tried to describe the criteria for choosing the right multimeter as I see them. I may have got some things wrong, so I welcome comments, additions, and corrections.



The product was provided by the store for review.