Review: £10 multimeters - Now includes the new RS-14!

[EinsteinAGoGo]

Most of us have no need to go prodding around busbar chambers or even our own consumer units.

Certainly, and, if you're looking for mains voltage, in a disboard, etc. you only really need to check 'is it there, or isn't it'
And you'd be a lot safer with a Martindale, or something similar, than any DVM.

(Random example)

http://www.ebay.co.uk/itm/331468636870

And I wouldn't want to make any measurement across energised bars carrying any sort of load.

[Mark Hennessy]

I'm not sure what the difference between the Martindale and a Fluke 87V as they have the same Cat ratings - apart from the fact it doesn't have a current range, so you can't try to probe a busbar while accidentally set to A or mA.

That said, the Fluke 87V beeps at you if you have a probe in either current socket when set to non-current functions, which helps to avoid that particular "brain fart". But if you do make the mistake, the HRC fuses in a Fluke stand a much better chance of opening safely than the 250V types fitted to cheap multimeters. And the design of the case and PCB is again intended to reduce the chance of injury to the operator.

Ultimately, I wonder if a non-contact detector is the safest way for that sort of work?

[ed731pdh]

From experience, clamp meters tend to do better with higher voltages or amps.

[Dick.1]

Very interesting thread Mark.

Thanks for posting

Dick.

[Mark Hennessy]

More cheap multimeters...

Here are 3 multimeters from Aneng. I believe the OEM is Zotek, whoever they are. Versions of these are available with either Aneng or Zotek branding - I've ended up with Aneng...

They all share several characteristics, so it makes sense to mention these in one post.

First, they are true RMS. Of course, this is astonishing at this price point. The RMS conversion appears to be digitally implemented inside the chipset, which is unusual as most multimeters today continue to use the "analogue computer" approach. But, once you've done the basic design, it's cheap to mass produce.

The chipset in question seems to be the DTM0660 - datasheet here: http://www.kerrywong.com/2016/04/03/dtm0...ranslated/ - if you search around, there's a lot of chatter about this chipset. It can be hacked to improve/alter the functionality of meters equipped with it, it seems. I've not tried anything like that as yet.

The calibration is done using NVRAM rather than pre-sets. That's like a proper meter! It does have the drawback that you can't easily recalibrate it should the divider resistors drift with time, so it'll be interesting to see how they hold their calibration. Needless to say, at this price point, there's no fancy thick-film networks! But right now, they all seem to be bang-on. They all meet their spec with ease, and agree with my existing Flukes within a couple of counts.

Two of them are 6000-count (the third is 9999-count). All meters in this thread so far have only been 2000-count!

They all run from 3V, so a pair of AAA or AA cells. All have display backlights (with annoyingly short time-outs, like the others reviewed above), they all have displays with excellent viewing angles, they all measure frequency and duty cycle and capacitance, they all have pretty fast continuity response. I wouldn't recommend any of them for high energy use, but they should be fine on the electronics workbench. Some testing has been done - e.g., https://www.youtube.com/watch?v=HrcxnbkkhYg - and they're not at all bad for their price.

They don't suffer the over-shoot issue that I highlighted earlier. The auto-ranging isn't quite as quick as the earlier meters, but the settling time is consistently short.

Even the manuals are OK. But very small print!

In short, much as I still like the Bside ADM02, I'd recommend any of these over that now. There is just so much more functionality for hardly any more money.

I'll review these in the order I purchased them in the next posts...

[Mark Hennessy]

The Aneng AN8002 (aka Zotek ZT102). About £12 delivered.

   

This meter is tiny. Honestly, it's hard to describe it. Physically, it feels very solid indeed as the case is quite thick. But, it is hard plastic. The pictures led me to think that there was a holster, but no; it's solid plastic. It bounces well, being quite light, so the lack of a holster or other padding probably isn't a big problem, but a rubber holder would make it less slippery on the bench. There is the usual tilting stand, and on a carpet tile, you can operate the range switch with one hand if you use a spare finger to steady the meter. The range switch is quite stiff, but positive in its action. The two buttons are a rubber membrane type, and feel nice.

   

Going around the range switch (options available by pressing Yellow in brackets):

V (AC, Hz, duty cycle, and back to DC)
mV (AC and DC)
Ohms (continuity, diode test, capacitance, and back to resistance)
Hz (% and Hz)
A (AC and DC)
mA (AC and DC)
Temp (C or F)

The blue button is Hold (why bother?), or a long-press switches on the backlight. At least the behaviour is sensible (the long press doesn't engage Hold in the process, unlike the RS14).

Being a small meter, the mA socket is shared with Volts, Ohms, etc. The 10A range has its own socket.

The best available resolution is 10uV, 10uA, 0.1 ohm, 1pF and 1 degree. When it comes to frequency, it's 0.001Hz. Astonishing, it had no problems going to 80MHz (yes, mega-Hz). Joe's video (linked above) suggests it'll go higher, but beyond that frequency the 17dBm (~1.6V) from my RF generator isn't enough. When I get the time, I'll properly characterise the sensitivity vs frequency, but suffice to say at 100kHz, it needs -20dBm (~20mV). In the video linked above, Joe Smith found it works up to 200MHz, though no level was given. Pretty good, really.

The meter comes with a storage bag and a set of probes (as usual, removable caps which change the CAT rating). The probes aren't terrible - better than the Bside ones, but not exactly Fluke. I've used them with no problems though. The meter is available with more accessories - see later - but that does bump the price up somewhat. The sockets aren't great - the same style as the Bside - but they take "proper" probes fairly comfortably (they'll stick up about 6mm, but make reliable contact).

Inside, it's striking just how little is in there. They use MELF resistors for the main divider and shunt, there isn't much in the way of protection beyond a PTC for the resistance current source, but Joe Smith found that the meter was quite robust for its class (it passed tests the Fluke 87V didn't, if I've understood correctly). Note the fuses, which are 10mm long. I'm not sure how easy these will be to source - apparently you can get wire-ended fuses in that size, so you buy these and snip the wires off. At least they're ceramic, but they're certainly not HRC types!

   

Note the 8-pin EEPROM next to the COB. And the switch (SW3 RANGE). Yes, it does change the range. It's a shame it's not on the front panel. Operationally, this is really the only criticism - it's not often you need to switch to manual ranging, but when you do, you do.

The electrolytic is a Rubycon. No expense spared! Or it's fake - who knows

   

There's nothing on the front of the PCB beyond the display and the switch tracks (looks like gold plating). You can see the OEM model number (ZT102) on the silk screen, along with some iffy hand soldering. Not visible in the image is a small amount of grease for the switch contacts.

In short, I've been using this meter for a couple of months now, and I really like it - I love the sheer amount of functionality in such a tiny instrument. I especially like the large LCD with its excellent viewing angles (it only starts to dim when viewed at 90 degrees, but on the bench or leaning on its stand, it's perfect). At this price, I've no hesitation at all in recommending it.

Be careful when buying as some sellers also offer the AN8001 (no temperature) or the AN8004 (no temperature and only 2000-count, despite costing more than the AN8002). Check the drop-down menu before clicking "Buy it now"!

More to follow...

[Mark Hennessy]

The AN860B+ (aka Zotek VC17B+). About £17 delivered.

       

This is a much bigger meter. It's nearly the size of a Fluke 87V (see picture for scale). It has all the functionality of the AN8002, but with some nice additions:

• Separate jacks for current.
  
• 20A rather than 10A
  
• Micro-amps range (100nA resolution)
  
• Manual ranging
  
• Relative mode
  
• Max/min
  

This is still a 6000-count meter. In terms of resolution, it's the same as the AN8002 apart from uA.

Note there's no dedicated Hz position. To get decent sensitivity, you go for mV, press "Select" to change to AC, then press "Hz %" to switch to frequency. It's not as good as the AN8002. Certainly, 100kHz at -20dBm was no problem, and 1MHz is feasible at the same level, but it starts to roll away (5MHz needed -15dBm and 10MHz was tricky at the levels available on this generator).

BTW, I haven't tested the AC volts extensively, but the -3dB point of all of these is in the low kHz region. More later.

The probes are the same as the AN8002. Better probes work, but still stick out from the front panel (I might've thought they'd go in further as it's a larger case). It's worth saying that this meter gets a separate rubber holster, and it's pretty nice. Still the same sickly shade of blue - the thing I like least about both these meters, frankly. Like the Bside (and many others), the holster has space to store or operate the probes in the rear. This meter runs on a pair of AAs (the AN8002 uses AAAs). The viewing angle isn't quite as good as the AN8002, but is really quite OK unless it's laying on its back at the rear of the bench (much better than the RS14).

Inside, it's even less exciting.

   

Note that they've used the more usual form of SM resistors rather than the MELF type. The main input resistor is formed from 7 1.5M types. As before, there is limited protection (just a PTC). The fuses are 20mm, but glass. Luckily the case is strong, with good overlap between the two halves, so hopefully all the glass fragments will be contained should the worst happen. The sockets are the same cheap type as the AN8002 (and the Bside).

Here, the PCB is layed out for a QFN packages as well as the COB, which would make life easier for those who wish to hack it. I didn't take a picture of the front, as there's nothing to see, plus I couldn't be bothered to desolder the backlight LED.

In summary, this is a great meter for the price, but as I already have a Fluke 87V, that's what I'll use! If you don't have one, I'd recommend this for bench work (low voltage, low energy only). Safety aside, it has much of the functionality of the Fluke, and it's really quite nice to use.

[Mark Hennessy]

The AN8008 (aka Zotek ZT109). About £15 delivered (including accessories).

   

Finally, the newest kid on the block. This meter is the same size as the AN8002, but has a number of improvements:

• First, the colour scheme is better. Trivial, I know, but I really like it!
  
• It's a 10,000 count meter (9999)
  
• It has uA. And it has 10nA resolution!
  
• It has improved voltage resolution (1uV in the 10mV range). That's bench-multimeter territory!
  
• The temperature function has been replaced by a square wave output mode. That's of dubious merit, frankly...
  
• There is manual ranging!
  

Really, this is a seriously good bit of kit, and it's still tiny!

Having just checked, the frequency counter still gets to 80MHz.

The square wave output is 3V peak to peak - I haven't yet checked to see if it varies with battery voltage. It's from a source impedance of around 2k, and it's AC coupled, so symmetrical about ground. It's pretty clean. By pressing "SEL", you can choose from a range of frequencies from 50Hz to 5kHz. As I say, I'm not sure I'll ever use it, but perhaps it'll be of interest to the "makers" with their Arduino projects? The waveform is very clean and the frequencies are all spot-on.

Inside, it's very similar to the AN8008, but there are layout differences. Component quality is the same, but soldering is much better in this example. I note there is another switch (SW3), but no idea what it does yet. This one has an extra TO92 device, which is a precision voltage reference (ICL8069 - though note the incorrect silk screen!). Is that better than the internal one in the COB, I wonder? Otherwise, it's the same small ceramic fuses and MELF resistors.

       

I've only had this meter for a couple of days, but already I love it. I didn't think it would be possible to improve on the AN8002 without spending a lot more, but for an £3, you get improved performance, and the extra accessories. Perhaps in time, cheaper versions minus the accessories will become available?


Conclusion

These reviews have been dashed off in a hurry, so I'm the first to admit that I've not tested them as thoroughly as the Bsides and RS meter. However, I have actively used the AN8002 for several months now, so I'm sure that more rigorous testing won't bring any surprises. Also, it turns out these meters have been quite comprehensively investigated by others - just search - and no serious problems have surfaced in that time. So I'm confident about recommending these to those on a budget. Again - and sorry to repeat myself - these are for low energy bench use only. They claim 600V CAT III and 1000V CAT II (the AN860B+ actually claims CAT IV 600V). These haven't been independently verified (obviously, as this price point). In terms of construction, all these meters are much of a muchness, but they are disposable. With care, they should last well enough.

Hope this is of interest,

Mark

[ppppenguin]

Again it's just astonishing how much you can get for your money. With meters like these and scopes from Rigol many of us don't need to pay more.

It seems the only reason for buying Fluke or other high quality kit is now safety in high energy environments.

[Nick]

It claims "True RMS" too - have you checked this with maybe a sawtooth, square, triangle or even dodgy-messed-with pseudo-sine-wave?

Thanks for the excellent (as always) review - very tempting DMM...