The Worst Piece Of Test Equipment You’ve Got To Try Hacking

A brand new meter in its blister pack
A brand new meter in its blister pack

I have a fascination with the various online vendors of electronics and other manufactured goods from China. Here are listed the latest wonders from Shenzhen or wherever, which you can have for a surprisingly reasonable price, with the mild inconvenience of a three week wait for the postage.

A particular pastime of mine is to look for the bottom end of the market. Once I’ve picked up the items I came to order I’ll trawl around with the search with low price first and see what can be had for a few dollars. Yes, I take a delight in finding absolute trash, because just sometimes that way you can find a diamond in the rough.

So when I was shopping for a multimeter recently I took a quick look to see what the cheapest model from that particular supplier was. For somewhere around £2.50 or just over $3, I could have a little pocket analogue multimeter, the kind of “My first multimeter” that one might have found in the 1980s. They weren’t too bad, I thought, and ordered one for less than a pint of beer in a British pub.

New Improved Modeli!
New Improved Modeli!

What arrived was promising enough, in a plastic blister pack, the Sunma YX1000A. The cardboard backing proudly proclaimed “New Improved Modeli”, so I had evidently made a wise purchase. On the back it told me there was a multimeter, test leads, battery, and instruction leaflet, but what I unpacked was only the meter and rather lightweight leads. Still, who needs instructions for a multimeter, and I have plenty of AA cells. Unscrew the back of the case, complete with a chip of plastic missing from one of its corners, pop the battery in, and away we go. £2.50 for a multimeter, we’re in for a treat!

This is why moving coil meters need damping.
This is why moving coil meters need damping.

My first task was to hook it up to my trusty 723 power supply, and check that the meter works. Fantastic, the needle swings up to 7.5 volts. Then back to 3.5 volts. Then back to 7, and so on in a crazy oscillation. But it settled eventually on 4V, which was a good sign. Or, at least, it would have been a good sign, had I not been supplying it with 5V.

A quick check with my everyday Uni-T digital multimeter, itself hardly a thoroughbred, and yes, I was giving it 5 volts. Maybe there was a parallax error at work, was the reflection of the needle lined up with the mirrored stripe on the face of the meter? Looking closely, I couldn’t see the reflection of the pointer, perhaps the front of the meter was steamed up. No, in fact the mirror wasn’t really a mirror at all, just a stripe of silver paint to look like a mirror from a distance. This is quality workmanship here, let me tell you.

That's silver paint, not a mirror.
That’s silver paint, not a mirror.

A quick check on the resistance ranges and the ammeter showed a consistent 20% low reading. No problem, users of a YX1000A will simply have to add 25% to any reading they get for the true value. And this gives an unexpected bonus, of a 25% increase in the meter’s range. For free. What other piece of test equipment delivers such value?

Opening up the meter for a teardown, and I found a single printed circuit board, with as expected the selector switch formed by PCB pads. With the exception of a through-hole rectifier diode and trimmer used as the zero adjustment for the resistance ranges, all components were surface-mount. There was no plating on the pads, save for the HASL or similar PCB tinning. Those pocket meters back in the day would usually fail because of oxidation of these contacts, no doubt this one would eventually succumb to the same fate.

Opening up the YX1000.
Opening up the YX1000.

The Sunma YX100A then: A case with a chip in it, a wildly oscillating meter mechanism that evidently has no damping, a stripe of grey paint for a parallax mirror, and a consistent 20% low reading. Even with a slightly tongue in cheek review, it’s fair to say that I have had better multimeters than the Sunma. In fact it’s fair to say that every multimeter I have ever used has been better than the Sunma. It’s possible that to make a multimeter worse than the Sunma would be extremely difficult, but they must have done it, because as they say, this is the “Improved Modeli”. Just how bad was the previous unimproved model?

So, what have I proved here? £2.50/$3 does not get you a decent meter. No surprises there, but I bought it for a laugh so am not disappointed. In fact the review of an awful meter has been considerably more enjoyable than it would have been had it been a good one, such has been the catalogue of dire features to uncover. There are decent pocket analogue multimeters still being manufactured, but if you want one you should expect to pay what it is worth.

I look at this awful meter though and see an entertaining opportunity. Here is a case with an AA battery holder and a moving coil meter, for relative pennies. What can you do with a Sunma YX1000A to make it better? Would any of you like to drop one on your next Banggood order and have a go at hacking it? Perhaps you could fix it and make a good multimeter from it, or maybe you’ll gut it and make something completely different. Put it on Hackaday.io and tell me about it, I’m sure we’d all love to see. I can’t promise any prizes but I’ll see what I can do to grant fame and fortune to your work.

133 thoughts on “The Worst Piece Of Test Equipment You’ve Got To Try Hacking

    1. Always assuming it’s not just a moulded screw head to pretend it’s got an adjustable zero. After all, the parallax mirror is fake.

      One of the things the meter did get right was that it had been zeroed.

      1. You have indeed done us a service.

        The most important part of an analogue multimeter is the meter movement – if it’s crap, then there’s not much hope for the device. This one seems… crap.

        Re the selector contacts on the pcb – I have an old Radio Shack analogue multimeter that’s still running fine after 20+ years of occasional use. If the maker bothered to put some contact grease on it, or if you add some, they can last.

        1. That’s where ironically a cheap electronic multimeter would probably have faired better than this cheap analogue multimeter.
          One can be had for £3.46 on Banggood, ID #1119906

          Does it have the flair of an analogue one? Nah. Does it work? Probably alright.

      1. I doubt there was much profit in it! And don’t blame the Chinese for landfill, in fact there is a lot of trash that’s being shipped to China because they’ll take it apart and try to recycle what proves to be too expensive to do so everywhere else!

    1. The same way that most moving coil meters work – it takes some of the power from the circuit you are measuring (the battery, in units like the one Jenny bought, typically use the battery for resistance measuring) – it’s why a lot of coil meters often have a lower impedance than their digital equivalents as they’ll be leaching power to drive the coil directly (rather than sampling and buffering the voltage)

      1. Really all a lot of these cheapie meters are doing is rectifying the AC to DC then measuring the DC voltage with an offset to give a semblance of measuring RMS. If AC was being run directly into the meter movement, it either wouldn’t move at all, or swing with the AC cycle + then – till it pegs the bottom of the meter movement. Likely no analog meter movement actually has a fast enough response time to do that unless you’re measuring an AC signal in the low single digit hz though.

        1. Your right that the diode introduces an error which would be significant at very low AC voltages and for that reason they use a germanium diode that has a very low froward voltage drop.

          The other part of your comment gave me a chuckle – “to give a semblance of measuring RMS”

          The laws of physics (conservation of energy work and momentum) in relation to the movement of the needle dictate that the reading is precisely RMS.

          Digital RMS meters were created to emulate the RMS behavior of moving coil mechanical meters.

          1. The moving coil averages the force (in effect but not technically true) that is opposed to the spring and that is not RMS as there is no squared relationship to cause the “S” in RMS.

            The needle deflection as a result in changes of force is governed by s = ut + at^2/2 or the displacement is equal to the initial speed times time plus a half of the acceleration times time squared. The acceleration is the force from the coil decided by the moving mass (a complex equation in itself because of the rotational movement). This provides the squared relationship that causes the meters reading to be RMS.

          2. At steady state, where a meter will actually be used, there is no acceleration, and no velocity. The meter deflection is proportional to the average current through it, plain and simple.

          3. no matter how small there will be both acceleration and movement.

            If you believe there is no movement then people would be buying meters with glued in place needles.

        2. The scale on cheap analog meters is so calibrated to give the RMS voltage when the AC input is a sine wave… which is like 90% of the time we’d use it: the wall socket, or the output of a power transformer. But simply reading a half-wave rectified AC signal with a DC analog meter is not true RMS.

    2. Electromagnetism.

      It can read everything except ohms without a battery. Pass current through a coil and it creates a magnetic field. If the coil happens to be on a swivel around a magnet, poof, deflection. No battery required. Diode helps when measuring AC.

      Resistance needs a voltage/current source. Use in same circuit as above, but without external voltage. Full scale is 0 resistance, adding resistance == less current == less magnetic field == less deflection.

      1. Yup, this is how all of the analog meters worked, DC/AC voltage, current all worked without battery. The battery was only needed to pass a current though a resistor to be able to read its resistance. Unless you had one of the super high end FET analog meters back in the day that were transistorized.

      1. In my Basic Electronics Lab, one assignment was to make various measurements using 3 different VOMs, each with a different Ohms/Volt rating (the lowest was 1KOhm/Volt, I don’t recall the other 2). It was an eye opener to record the different readings each gave for the same voltages! This was before 10MOhm/Volt DVMs were common.
        Hmmm, maybe one of the meters we used was an RCA VTVM…

    1. I’d consider it a target for modification in to a simple RF power meter.
      No digital electronics to misbehave thanks to RF, multiple ranges and a diode detector supplies it’s own voltage to the meter anyway.
      then just some measuring against a proper power meter and making a new meter scale with Tonne software’s Meter or MeterBasic.
      Thanks to the selector you’d even have multiple ranges.
      I should attempt to find my cheapo moving coil meter. I think I paid 3.6eur for it.
      Add in two 100ohm resistors or a proper 50ohm RF termination, 1n5711 shottky and a 100nF cap. Rest is measurements and calibration.

      QST had an article on building QRP measurement equipment from harbor freight DMM’s too.

  1. Ooooh, ooooooooh! The review has quality matching the meter. Did you do that deliberately, or did the meter infuse your neurons with its innate je-ne-sais-quoi? You identified it both as a Sunma YX1000A and a Sunma YX100A, and you gave the inaccuracy as both 25% and 20%. So much to enjoy, and I didn’t even have to spring for £2.50 or any other amount of money, and instead simply allocated a bit of my copious free time. (Should be out feeding the horses.)

    Your review has provided me nearly as much mirth as the fabled advertisement for the write-only memory that appeared in an electronics magazine around 1970. Hold on a moment while I look under the rocks and through the Google …

    Got it! Here’s the whole WOM story: http://www.sigwom.com/?page_id=17

      1. How was the movement’s linearity across the scale? Were all measurements off by the same amount? if so, then it would be a resistor issue and the movement’s maybe not that bad. The meter overshoot could be damped by a capacitor, probably across the meter coil, at the expense of speed.

        $50 of work to make a $5 meter useful…

        These are “kitchen drawer” meters, destined to sit unnoticed til someone needs to check some batteries or a dead lightbulb. Our kitchen drawer meter is also a 2k ohm/volt meter, but unlike the Sunma, the movement is nicely damped, and it reads within 2% of my DMM. And the mirror is real. i recall it cost $10 on sale. So cheap meters aren’t all crap…

      2. meh. Grab a bottle of typing correction fluid (is there still any in left the stores?)and a pencil.
        Now pull the clear cover and just “re-calibrate” the dial face to match the needle deflection! ;)

  2. Most of the stuff I buy fom china via ebay has issues. When it does I ask for a refund and 99/100 I get it.
    I wouldn’t be doing this if they didn’t feel the need to lie about the product to sell it.
    But because that’s seemingly part of chinese business practise, I’m happy to take the compensation for the mis-selling.
    They seem more bothered about negative feedback than the fact that the item was crap to start with.

  3. I’m a big fan of bare meter movements for things, but you might as well start out with a reasonable-quality damped one, available for roughly $5. I just ordered a few center-zero units, and when they arrive I’ll try to remember to post something fun.

    Center-zero meter movements are perhaps my favorite vestigial analog instrument. Paired with a low-ohm shunt resistor, there’s nothing better for getting an intuitive feel of current flow in a simple circuit. Pretty much everything else can be done better digitally now, but not that.

    This textbook, by the way, is beautifully written and has an entire chapter about metering circuits, which is excellent for any noob to work through, given such a movement and a handful of resistors: https://www.ibiblio.org/kuphaldt/electricCircuits/DC/DC_8.html

    1. My 1953 Willys Jeep had a center-zero ammeter. It had no shunt, and no coil at all. The battery cable came through the firewall, looped through the back of the meter, and back through the firewall. The meter read directly the magnetic field around the wire. I was astonished when I figured it out, even more astonished than when I figured out how the speedometer worked.

      1. I’m guessing the speedo works the same as virtually all mechanical speedos: a magnet connected to a flex cable from either the transmission or a front wheel spins next to an aluminum disc connected to the needle, with a helical return spring. This is actually a rudimentary induction motor.

  4. Well for $3 it’s pretty good. I wonder what we in the West could produce for that price. Maybe a color paper page with a multimeter image :) ?

    Thank you China. You have good and bad stuff as everyone else. It’s not your fault that we cannot stop buying stuff even though we are already full like a Titanic.

  5. Very amusing piece. Doubt I’d ever consider buying test equipment directly from China. Have had mostly good experiences buying components, mostly v cheap ones like ne555 and lm358s. Have had a few dodgy dc2dc converters, though had a few decent ones too. The stuff is so cheap it’s not a big loss if some of it is well off the mark, assuming you’re only using the stuff for fun.

    1. I just bought an AC and DC clamp-on loop-current DMM from China for about US$40, and it’s actually pretty great. Nicely rugged, and with useful functions. It’s now my main meter for auto and boat DC service.

    2. I would never buy ne555 or lm358 chips from China. I would forever have to worry before using either type chip again whether I was using one of the sketchy ones. NOT worth the money you save.

  6. Silver stripe for a Mirror… nice! I’m guessing the producers of this toy have never seen or handled a real meter in hand but have only seen pictures of a meter. At least they didn’t paint the needle! Perhaps it would be better if they did? Reminds me of a joke, at least a broken clock tells the correct time twice a day.

  7. I had a digital meter I got for about the same price at Harbor Freight, wasn’t terribly accurate, hated AC, and once I was checking ripple on an alternator with it, and a spike fried something inside, as after that the display would just show random gibberish.

  8. I know you are after cheapies, but how about a review of the higher end stuff e.g. 5-8 digit VMs in the GBP 200-400 range? I’d love to know if they give fluke a run for their money!

    1. So would we. But we buy our review models, we don’t take freebies or loans except in extremely exceptional circumstances. It means you can trust our reviews. And several hundred $ is probably out of budget, sorry.

  9. Bahh!! that’s a perfectly good piece of test gear – you just didn’t correctly specify what you’re testing.

    Is it still on your desk? – Gravity detector!!
    Is it wet? – Rain detector!
    Is it melting? – Overtemperature detection!

      1. Stereo Review (magazine) once proposed a test for phonograph cartridges.
        ERC (Elasticity Rebound Characteristic) : the height of rebound (in millimeters) of a phonograph cartridge dropped at STP from 1 meter onto a level slab of marble.

      1. LOL I had a “friend” that never returned tools. He then asked if I had a cut off saw so I made an arbor adapter to put a 19″ TCT circular saw blade on a 4″ angle grinder with no guards and left that in his car port. Soon after that he returned all my tools. Perhaps we was stupid enough to turn it on. It would have pivoted uncontrollably as it spun up.

  10. How about making this into an article/lesson on meter calibration? Show how that’s done, then perhaps a couple of drops of silicone oil on the movement pivot will damp it enough to be usable.

  11. Whenever I see these I always think of using it to make an analog FET voltmeter. That was a common DIY project before cheap DMMs appeared. And still a useful piece of kit if you need to monitor a slowly changing voltage. You could make a pretty good piece of gear for RF testing for field day with rather modest additional effort.

  12. I had an identical meter in the 1970’s. It was made in Japan (China wasn’t industrialized then) and it was relatively cheap and also a reasonable quality budget meter.

    Well at least with this one you got your £2.50 worth of … entertainment!

    1. I seem to have bought this exact meter about a month ago – because I _couldn’t_ buy one at Harbor Freight. Within the last year or so they seem to have discontinued their analog meters and now only have the digital variety. Nothing wrong with digital, but analog has it’s place too. :-)

      I’ve yet to get it out of the package, but after reading Jenny’s article, I’ll have to lower my expectations…

  13. Oh, I actually had the old “MODELI”, or at least one with the same name and design, when I was a kid!
    Bought it in a vietnamese market for about 2$ equivalent!

    But, amusingly, it was much better than this one. The battery was included for start, and the paralax mirror was actually there.
    The measurements were reasonably accurate, considering I was a kid, only working with low DC voltage.

    When I finally decided to upgrade I opened it. Incredibly it was the same design as this one, down to the diode. But traces were somewhat wider, almost as if the designer did it by hand, also the pads had some minimal plating and there were no smd parts.

    So I guess those were the “IMPROVEMENT”.

  14. Less moolah == more better.. this one is only £2.15 -> https://www.ebay.co.uk/itm/AC-DC-1000V-Voltmeter-250mA-Ammeter-1K-Resistance-Meter-Analog-Multimeter/222210305994?epid=25005530104&hash=item33bcc42fca:g:q4sAAOSwyLlXpJEs
    I’m almost tempted, but I can get a crap digital one for only £2.38 -> https://www.ebay.co.uk/itm/DT830B-LCD-Digital-Voltmeter-Ammeter-Ohmmeter-Multimeter-Volt-AC-DC-Tester/282577230333?hash=item41caea59fd:m:mETGN7yRMa3jdgoEmhctRew
    .. choices, choices… Maybe we need a hackaday.io page for hacks for the DT830B and all its crappy variants too.

    1. I got a DT830D (similar to more common B version, but with a speaker for continuity test and a crappy frequency meter) for less than $3 from Aliexpress. I still use it as a second multimeter and it’s pretty good for the price (even better than the B).

    2. Some people really need to get a link or URL anonymizer.

      At least your links don’t contain your full family genealogy like some that I have seen but eventually if you keep doing this you are going to give someone access to your website logins. Even posting *** links like this can be dangerous if your account is linked to a financial payment method. Thankfully you were not logged in when you copied the link. Hint hint.

      You can even “trim” links manual if you recognize the “human readable” part of “Search Engine friendly URL’s”

      You links become –
      https://www.ebay.co.uk/itm/222210305994
      and –
      https://www.ebay.co.uk/itm/282577230333

      Off topic, I know! But I have seen links so bad (here) that if hackers were car thieves they would be equivalent to driving your car to the car thieves house ensuring it has a full tank of gas (fuel), signing over ownership and registration and giving a bottle of champagne as a bonus.

      Not as bad at other sites where are compelled to ask – “what the go here! your URL doesn’t include your ***** (genital) size!”

  15. Harsh whingy comments!
    We are not the target market for stuff like this, well not all of us. This is a ideal pos meter to leave in your glovebox or behind the seats of the van, so when something goes wrong electrically hundreds of miles from home, its raining and your trying to diagnose why the fuse keeps popping beneath a streetlight so you’ve got headlights that work, even the crappiest meter is a joy to have to hand.
    Sometimes you just need a little needle to wave randomly, and suggest at values, do a little continuity testing and have so little actual value you dont feel guilty about abandoning it into that magic space behind the seats where tools go to die. Plus its so obviously crap, its unlikely that someone will steal it.
    Just dont use it on your bench for anything requiring anything more. Job done.

  16. I appreciate the test, although it could have been more throough..And Ken’s reminder if RadioShack’s Mixronta, $3.98 in ’66… as stated, ran for decades. So adjusted for infkatiin it might be a $15 meter now… and dejusted for deflation, the reviewed meter might be a buck. So of COURSE you have to fiddle with the resistors. I would have liked to see Gilligan’s Professor make an equal meter on the island, and I’ll ignore the lack of portability to which coconuts would have limited him. It’s hard to find a meter if this type and footprint. There us a place for a slightly better meter than the above at a low price.

  17. Another cheap meter that can be had, is a DMM from Harbor Freight, using their coupon “make any purchase, get this meter free”. Unfortunately, I don’t have it in my hand right now, so I can’t tell you what model it is, I bought a small pack of heatshrink tubing (less than $2)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s