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

[Mark Hennessy]

Hello,

A couple of weeks back, we were discussing cheap multimeters, and curiosity got the better of me, so I ordered one of these: http://www.ebay.co.uk/itm/Digital-LCD-Mu...1100534241

It didn't arrive quite as quickly as promised, so it didn't make it in time for my Easter break. But it's here now.

Short version: a lot better than I expected. A worthwhile step up from the horrible £5 versions. Some problems, not surprisingly.

Long version follows:

For some reason, it's impossible to determine the brand and model number from the advert. However, it's a B-Side ADM01 - http://www.bside.hk/bside%C2%AE-adm01.html

Who are B-Side? No idea, but it's made by Mastech: http://www.testingequipment.com/mastech-...meter.html - as we'll see, it's an MS8233D, although it's missing a couple of features of the original (no-contact voltage detection and temperature measurement).

It's a basic 2000-count meter, auto-ranging, 0.5% basic DC accuracy. Uses a PP3 (not included).

Out of the box, first impressions are positive. The manual is reasonable, and the leads look quite good. As you unwrap those, you find that they are very stiff compared to normal silicon test leads, and the probes are hard plastic that is much more slippery than the "soft-touch" finish that Fluke and others use. But, a lot better than the leads you get with the £5 meters.

           

The meter itself is tiny. It's very solid, and the holster fits well. The mode switch is positive in action, but a bit "clicky" and stiff to turn. The 4 buttons are horrible - hard plastic acting on cheap tact switches. The tilting bail is a bit pointless, as it doesn't provide enough stability against switch rotation and button presses, or indeed against movement caused by the stiff leads. On the back, the holster includes moulded sections to store or use the probes.

           

Operationally, it's OK, but a couple of annoyances. First, the combined diode test and continuity mode defaults to the former. That needs to be the other way around as you'll use continuity more than diode test. To change mode, press "FUNC.".

Next, the backlight is good, but it only stays on for 15 seconds! And you have to hold the button for 2 seconds to persuade it to come on.

Pressing any of the switches results in a loud beep if the keypress is allowed. No need for that...

The "FUNC." button switches between AC and DC on the current ranges - the default is DC, which is good. The Fluke 87V does the same, but AC is the default. That still catches me out occasionally.

There is a "HOLD" button, and a "MAX." button. Neither are terribly useful in practice.

One big problem is the sharing of the positive input terminal between milliamps/microamps and the usual modes of V, Ω, etc. This means that if you accidentally turn the mode switch to a current setting, you might end up blowing the internal fuse (or worse). Of course, there isn't really space to have a 4th terminal on the unit, and it is nice to have the microamp range.

Terminals are spaced at the industry-standard 19mm, so adaptors will fit them.

Auto-ranging is pretty quick, but no manual range option. That's a pain. I'd happily swap the "HOLD" button for a range button.

The unit is quick to settle on a reading on all functions apart from Ohms, where it is painfully slow - like 5 to 10 seconds! It doesn't seem to be an auto-ranging problem, as changing from 5k to 10k (using a decade box) doesn't require a range change. I wonder if they are all like that, or perhaps someone fitted the wrong capacitor to mine?

DC accuracy is excellent - well within the specs on the lower ranges (only tested up to 60V). Almost as good on current - and it beeps at you if you exceed the maximum current, which is useful.

       

On DC volts, there is an overshoot issue. The first reading is a couple of volts higher than the correct value. This means that if you measure 18V, the unit will auto-range to the next higher range and show you 18.0V rather than 18.00V. Not good...

AC measurements are the expected average detection, so read high with square waves. Frequency response varies with range - on the 2V range, the -3dB point is 17kHz and it falls away after that as you'd expect. BUT, on the 20V range, it started to over-read at ~4kHz! With a 5V sine wave applied, it peaked at 190V at 170kHz! I suppose it's better to over-read than under-read, but I was quite surprised at this – clearly a problem with the attenuators.

On AC current, it was fine...

Frequency measurements were disappointing - which is a surprise. When we take it apart (next post), we'll see why...

50Hz gives -0.01kHz (yes, minus!)
500Hz gives 0.46kHz
1kHz gives 1.00kHz
5kHz gives 5.04kHz
18kHz gives 18.08kHz

(5V RMS sine wave)

Sensitivity was better than the 0.8V spec – at 1kHz, it worked down to 75mV RMS.

The diode range is limited to around 1.5V, so it won't light an LED.

Continuity is non-latched, and works at around 10 ohms and less, not the 100 ohms quoted in the manual.

Current consumption from the PP3 varies between 2.12mA and 2.68mA. However, the buzzer takes about 23mA, and the backlight takes 15mA. The low-battery indication comes on at 6.1V.

Next post: teardown...

[Mark Hennessy]

First step - remove battery cover/tilting bail, and then the holster (easier said than done!):

   

Inside, a single PCB. Quite reasonable at a glance:

   

Undo 4 screws to remove the PCB:

   

A closer look at the top and the bottom of the PCB:

       

So we have ceramic fuses - albeit rated to 250V. The meter is supposed to be CAT II 600V, but that rating only applies to the voltage input - the current connections are labelled as "250V max" on the front.

No isolation slots routed in the PCB, no shields between the naff terminals - what one expects at this price point. The mode switch is also as expected - no gold plating at this end of the market! There are the usual ball bearings and springs to provide the detents. Note the covering of silicon grease around the switch contacts.

An even closer look at the PCB:

   

Looking at the input circuitry, we can see that a spark gap is missing, despite being mentioned at least twice in the manual. There is a pair of PTC thermistors for protection (not sure how useful they'd be, given the size of them and how long they'd take to warm up). At first glance, the spacing on the PCB looks reasonable, but there is a massive howler on there - look for R1A and R1B. These MELF resistors are 5M each, and are in series to give 10M - and the other end of this goes straight into the IC. The lower ends (green bands) are only separated by about 2mm, so if a spark jumps this, the energy will blow up the IC. Fail.

Moving to the right, we can clearly see the current shunts (R49, R48, etc). Note that VR4 calibrates the 10A range.

Moving up, look at the 2 SO-8 packages. These are doing the frequency measurements. There is a TL062 dual op-amp and a 7555 timer! The circuit is a F to V converter, calibrated by VR3. Explains why the frequency measurements are so poor...

Looking at the top half of the PCB:

   

Top-left we have a bunch of unpopulated components. I'd bet that most of these are for the contactless voltage detection.

The main IC is an FS9952 in COB form:
  FS9952 Datasheet.pdf (Size: 453.34 KB / Downloads: 11)

The crystal is 4MHz. Not sure what U2 is, but guessing it's a voltage regulator - the chip runs on 3V. The MELF resistors (R2-5) are range-switching, working with the 10M seen earlier. Note the poor hand soldering on the tact switches and wires. Note also the flux residue. I did scrape some of it away around the +Ve input socket.

From a look at the PCB and the datasheet, it appears that VR1 adjusts the voltage reference in the IC and VR2 adjusts the AC calibration. D1 is part of the rectifier, which makes use of 2 op-amps inside the IC.

So, not as bad as feared based on previous experiences at this end of the market, though I reserve the right to change my mind once I've used it for a bit! I'm not sure I'd use it on the mains - and definitely not in my consumer unit - and I'd also be cautious with HT, given the problem with R1A/B.

It'll be really interesting to get hold of one of those RS meters - unfortunately out of stock until July now. Apart from the different LCD panel and the different positions of the range switch, these look to be almost identical - I bet it uses the same IC and basic design. I note the RS version has a range button, and temperature rather than the pointless frequency option. Incidentally, if you can find it, the ADM02 also has temperature rather than frequency.

Hope this is of interest,

Mark

[Murphyv310]

Hi Mark.
What an excellent review, very informative and concise. I'd happily pay a tenner for the meter even considering its limitations its a handy meter to have, especially around transistorised equipment.

The only thing that you have missed from the review and its not a criticism, you didn't do a drop test! Only joking

[Mark Hennessy]

Thanks, Trevor

Yes, it's astonishing value for money. Drop test? There's an idea

I reckon it would do OK - certainly a lot better than the £5 versions. The holster is quite compliant, and the case is quite chunky. No-doubt an involuntary drop-test will happen sooner or later - I'll report back...

[Mark Hennessy]

By way of an update...

Don't buy an ADM01. Instead, buy the ADM02

What's the difference?

The ADM02 swaps the frequency measurements for temperature. Now, if the frequency measurements on the ADM01 were any good, then this might be a trickier decision. However, as I've said above, they're not worth the effort frankly.

The ADM02 is the same price - £10 give or take a few pence - and is available from numerous eBay sellers. Mine took a long time to arrive, but that's to be expected.

Externally they are almost identical. A standard thermocouple is supplied. The test leads are the same. The supplier said that batteries aren't included, but mine did arrive with one fitted. However, it was completely dead!

Internally, things are slightly simpler. The op-amp and 555 timer used by the Hz function is gone and things are slightly more relaxed in that area of the PCB:

   

As I said earlier, I think these are outstanding value, even with the reservations raised earlier. A definitely step up from the sub-£5 models.

[Mark Hennessy]

And now, the multimeter that started all this - the "RS-14".

The B-Side meters were picked up on the basis of them being a close visual match for the RS-14. Well, as far as you could tell from the pictures on the RS web site. However, the RS-14 is different from B-Side meters. But they have a lot in common...

First, it costs £14.20+VAT at the time of writing - 17.04 all in. In percentage terms, that's quite a step up from £10, but when you think what an entry-level Fluke would cost, it's still ridiculously cheap!

       

What are the differences?

Firstly, the case doesn't have the separate holster that the B-Side units come with. Instead, the blue sections are that soft-touch plastic that is so common today, and that helps to give the unit a more expensive feel. However, it doesn't add any strength to the case, and I reckon the B-Side meters would perform better in a "drop test" because of the shock-absorbing properties of the holster. And soft-touch plastics can become tacky with age. The underlying plastic in both cases is much the same thickness, and both types of meters feel equally solid - commendably so given the price point.

Side by side, the RS-14 is slightly smaller, although the difference is mostly because of the holster. Not visible here, it's thicker, but that most definitely is because of the holster (that has the probe storage sections moulded into the rear).

   

Earlier I criticised the rotary switch of the B-Side for being a bit stiff (so therefore hard to change mode with one hand while the unit is stood up on its tilting bail). However, the RS-14 is much worse. On my hardwood bench, it is impossible to change mode without it sliding about the place. On a carpet tile it's better, but you still need to use spare fingers to hold the damn thing still while turning the switch. Really, you need to have it flat on its back. The switch has a rubberised coating to help with grip, and that's very necessary!

Also, when on its back on a hard surface, it slides about all over the place. The only non-slip elements are at the top, so offer no resistance against the rotation of the mode switch. If only they'd put raised non-slip bumps on the bottom as well. The B-Side is much better here.

However, the buttons are much better. Rather than the cheap tact switches, they've used a membrane keypad. Good...

In terms of functionality, the left button does the same with both - changes between AC and DC on current, switching between C and F, switching between diode and continuity. You get the idea. But, the next one changes range with the RS-14. That's good news. You might only need manual ranging occasionally, but when you need it, you need it!

The next button is a combined backlight and Hold button. That's less good - I'll get on to that.

Finally, there is the Max function. As I said before, not especially useful...

The LCD is much bigger, giving larger digits. But the viewing angle is rubbish compared to the B-Side types. If you lay the meter flat on its back - which the stiff switch and hopeless tilting bail more or less obliges you to - then you can hardly read the display because the "off" segments are lit up! It's worse if you're off to the side.

   

The probes are better. They are the same length, but much more flexible. Not silicon at this price, obviously. They are dual-rated, depending on whether you use the protective covers that obscure most of the probe - in place, they are Cat IV 600V rated (and Cat III to 1000V). Remove these and they are Cat 2 1000V (the same as the B-Side types).

   

Safety is the big selling point with this meter - it's rated to Cat III, 600V. Of course, this means the Cat IV rating of the probes is moot, as the "weakest link" sets the overall rating. The B-Side meters are only Cat II 600V, so theoretically, the RS meter is safer. To be honest, having done an internal inspection, I'm unconvinced. I think the RS-14 is less likely to damage itself when subject to a voltage surge. But in terms of damaging the operator, which is what the Cat ratings are really about, I'm not sure. Both meters use the same sorts of fuses for current, and both have very similar construction using similar materials. So without independent testing, the jury is out for me. Frankly, if you need to work on an incoming supply, you should be using something rather better than sub-£20 multimeters.

What do these meters have in common?

They both use the same IC. So some of the annoyances of the B-Side meters are equally present in the RS-14. Such as the stupid backlight behaviour. You have to hold the backlight button for 2 seconds to persuade it to come on, and it only stays on for 15 seconds! But with the RS-14 it's worse, as they've joined the backlight and Hold pins to the IC, meaning that a brief press of Hold engages the hold function. But the long press needed to get the backlight on also engages Hold, meaning that once you've got the damn thing to light up, you have to press the Hold button again to get it out of Hold mode. You'd get use to it, but I was surprised at how annoying that was.

By way of example, the Fluke 87V combines the backlight function with the Hi-res mode. A brief press controls the backlight, and a long press engages Hi-res. As you press the button initially, the backlight comes on (or more precisely, it changes to the next available state - on dim, on bright, or off). But if you hold the button long enough to put it into Hi-res mode, the backlight reverts to the mode it was previously in because it's obvious that you didn't want to change the backlight; you wanted to change resolution. Might seem trivial, but as someone who does a bit of UI design, it's the sort of detail I notice and appreciate when it's obviously been thought about.

The backlight is fine, but not as bright as the B-Side models. Interestingly, the ADM02 is considerably brighter than the ADM01, but that's almost certainly because they aren't too strict about the type of LEDs they purchase for these!


Teardown

Removing the battery access door also gives you access to the fuses, although I'm not sure what sort of tool you'd need to remove them! And look carefully through to the PCB, near to the battery wires. Yes, that is a pre-set resistor there - mind the lead doesn't trap itself in there and alter the setting!

   

Internally, they are quite different:

   

The first thing I noticed was the 10A current shunt. There's no preset, unlike the B-Side types, and "calibration" was performed using a pair of side-cutters. That surprised me!

Next, there's a fair few leaded components. All the divider resistors are TH types, mostly stood up on end (and need straightening up!). The obvious improvement over the B-Side meters is the use of leaded components in the 10M input divider - remember how I pointed out the possibility of a spark jumping the 1.5 or 2 mm between the end-caps of the MELF resistors (R1A and R1B)? Here, that prospect has been eliminated (see R27 and R28 - the 5M, 0.5% types - though I note the B-Side resistors are 0.25%). Thumbs up!

   

Also, something has actually been fitted in the position labelled SG. It's a MOV - an 820V type. Nothing stopping you retro-fitting one of these to your B-Side if you like!

Like the B-Side variants, the main IC runs at 4MHz. The one electrolytic capacitor is made by Chang - quality

       

The meter itself is a DT-914, which appears to be made by CEM


Operation

In use, the RS-14 has the same "overshoot" issue as the B-Sides. So, set your PSU to 18 volts or thereabouts and dab the probes on the output, and it'll jump to 18.0V rather than 18.00V.

But fortunately, it doesn't suffer from the same AC over-reading as you increase the frequency. Both my B-Sides do this, so it's not a fault with the first one I bought. The RS-14 just falls away as you'd expect - the -3dB point being approx 7.5kHz (well, on the 2V range - I didn't check the others exhaustively).

Interestingly, the RS-14 has an AC mV range that is easy to miss if you don't spot it in the manual. It will never auto-range into that range, but if you press the Range button enough, you'll get it. Because there is no Range button on the B-Sides, that mode is not available (even if it actually exists).

Annoyingly, the Diode/Continuity function defaults to Diode, just like the B-Side meters. It's non-latching again, which is fine if that's all you're used to - but once you've used a Fluke or similar, it's tough to go back. The threshold is 50 ohms. Amusingly, the RS manual says both <100 ohms and <30 ohms! The ADM01 was measured at 10 ohms, but the ADM02 is 40 ohms (the manual says <100 ohms). Clearly, this is not a precisely controlled parameter!

The slow resistance readings noted with the B-Side is also present. It's a bit hard to describe, and it's a bit inconsistent. It's not related to the auto-ranging, which is fast. I hooked up a couple of decade boxes to a B-Side and the RS-14, and played around for a good 10 minutes - on average, they're basically as bad as each other, but the RS-14 is perhaps slightly quicker.

Verdict

So, which should you buy?

Forget about the ADM01 with its useless Hz range - it's a 2-horse race between the ADM02 and RS-14.

Of the negatives I've raised, most of them are common to both - they both use the same IC and are electrically very similar.

The RS-14 is more expensive, but has the backing of RS. It's Cat III rather than Cat II, for what that's worth in the context of sub-£20 meters which haven't been independently verified. The RS-14 is slightly smaller, but slips around on the bench. The probes are better - quite usable. A working battery is included. The buttons are nicer.

Against that, the ADM02 is cheaper, but just as well built (apart from the probes, which really are too inflexible for day to day use). I prefer the separate holster, and I think it results in the ADM02 being the more rugged of the two.

On the bench, the main problem with the RS-14 is the display. The larger characters do not compensate for the poor viewing angles when on its back. But the tilting bail - combined with the over-stiff rotary switch - cause a problem. A solution might be a separate stand - perhaps a few scraps of timber to hold the meter at a suitable angle, with sides to stop the meter rotating when you operate the switch. Fine if you like woodwork and you have the time, I suppose...

So on balance, I'd buy the ADM02 unless the slightly smaller size of the RS-14 tips the balance for you. At this price point, both are disposable, but both represent a worthwhile step up in quality, accuracy, functionality and safety from the sub-£5 meters out there.

Will I buy a batch of RS-14 meters to replace the 30 year old Fluke 75s that we have in our Funds lab at work? No. We're better off hanging on to the old Flukes. But I might buy some to replace the AVO 8s that we still use (because some transistor experiments require the measurement of ~50uA, which the Fluke 75 doesn't do brilliantly).

Again, either of these offers so much more than the sub-£5 meters

[ppppenguin]

My RS meter arrived yesterday. Thanks for the review.

One very obvious difference compared to my Fluke 87 is that the low current ranges use the same socket as the voltage etc. This means you can blow the low current fuse a bit too easily by accident. You can argue that an AVO had the same arrangment but the cutout was easily reset. Just need to take due care. At least the current ranges are the other side of the off position so you can reach them by accident and the 10A range is first so even if you overshoot the off position you've got a margin of safety.

[Mark Hennessy]

Yes, the shared socket is one of the first things I mentioned in the other thread, although I don't think I've mentioned it again in this one. Obviously, the price to pay for the small size, where there isn't really space for 4 input sockets. And you have the same problem with the super-cheap meters as well...

[Old Sparky]

How would either of these meters compare with a M3800, one of which I've used for several years. These, whilst not being auto-ranging, feature both AC & DC Amps ranges up to 20A, the latter having a seperate socket, plus the usual AC & DC Voltage ranges, Resistance to 20Meg, diode test/continuity bleeper, and trasistor checker. The back has to be removed to change the battery and gain access to the fuses, but that is also true of the Fluke 77, one of which I still have but no longer use. The M3800 is about the same size as a Fluke 77.

[ppppenguin]

Mark made the point about comparing these sub £20 meters with the cheapest Flukes. Assuming the RS one genuinely meets its Cat III 600V rating (and I doubt that RS would be willing to put its good name to it otherwise) then it's OK for most tasks that most of us are likely to do.

Most of us have no need to go prodding around busbar chambers or even our own consumer units. Anyone who's messing around with HT over 300V really needs to be well aware of the risks, the DMM being just one small part of the problem. TV fans will encounter boost voltages up to 800V or so on some colour sets with enough power to do some real damage. There's enough line scan energy around there to trouble most DMMs so it's time to dig out an AVO for that. If you deal with TVs having mains derived EHT then you need the right kit, right attitude and a strong sense of self preservation.

If you're Nick and play with kit that can easily go bump in the night then you need an even stronger sense of self preservation.