Multi-Multimeters
09/08/2020
A consideration of multimeters I have accrued over the years, on the occasion of buying a new one.
A few days ago I wanted to check an old guitar lead to see just how bad it was. I ran across it in my old gear drawer while looking for something else, and hauled it out.
I remember I put it away perhaps twenty years ago as mysteriously noisy, remaining so despite checking all the soldering on the plugs at either end. Perhaps another look at it might reveal something I missed?
I remember I put it away perhaps twenty years ago as mysteriously noisy, remaining so despite checking all the soldering on the plugs at either end. Perhaps another look at it might reveal something I missed?
What I found was that my trusty old Sanwa digital multimeter was giving spurious readings.
I headed back to my old gear drawer to retrieve my even older Hioki analogue multimeter, seldom used since I got the digital one. The analogue needle display proved to be just the thing to show up how bad the lead was. Even with both plugs weighed down to keep the measuring lead clip connections steady, the tiniest movement of the rest of the lead at any point on its length would produce big and erratic needle swings of increased resistance. A close examination of the lead showed a suspicious green tinge, and an even closer examination with my smartphone camera's fancy new 20X clip-on magnifier (see Left) revealed that underneath the transparent plastic lead insulation, the entire length of the lead's braided outer screen conductor had a horrible green layer of corrosion, copper sulphate most likely. As final confirmation I used the lead to plug a guitar into my amp. The result was horrendous, an awful racket of scratchy, staticky noise, even from just picking up the guitar. The lead is proven to be totally and unrecoverably useless. To finally admit something is unrepairable is a hard thing for me to do. To commit to this conclusion I had to make myself unsolder the plugs from either end for my parts drawer, as they will no doubt prove useful. So......... now I had a useless guitar lead AND a dodgy digital multimeter. Time to buy a new meter, I reckoned. After a bit of on-line searching, I found one with more capabilities than the others, not the cheapest model, but not too expensive either, considering I am not using one every day. The vendor was only about 15 minutes drive away, a lot quicker than an on-line purchase! I am now the proud possessor of a brand-new Chinese-made UNI-T digital multimeter, model UT39C |
The fact that my oldest meter had outlived its successor, and that I now had an even newer meter
got me thinking about my meters in general.
got me thinking about my meters in general.
My Multimeters
The Hioki TX20 Analogue Meter
I found it still in good working order
(Except the Rx1 scale.)
20 kΩ/v
Tolerances 3 to 4%
Made in Japan
The Hioki site says on the Discontinued Products page: - "TESTER TX-20 Before 1980 Not Available".
They still make a similar product. I will presume I bought it in the late 1970s.
It is a very simple design, the manual even has a parts list and circuit diagram. There is no fuse mounting, just a length of fuse wire soldered across a gap on the circuit board. However, in addition to the manual, there is an explanatory sheet with some spare fuse wire stapled to it. The manual has the familiar charming English strangeness seen today with Chinese products, such as: "It is a very delicate instrument, but if correctly handled, it can be used for long years."
I was surprised to find it still had a battery (strictly, a 1.5V cell) in it; I would have thought I would store it without, since batteries can eventually go bad, leaking and causing corrosion which can even render equipment unusable. I must have used it not too long ago, as the battery was perfectly intact, not even flat, and well within usable charge level.
The pot for zeroing the resistance scale caused a bit of erratic needle-swing, but settled down well after a few wiggles back and forth to scrape off the suface oxidation.
This meter served me well for about 15 years and is still in working order, excepting that in the course of comparing the meters I found the RX1 scale is reading very low. (See test results below.)
I found it still in good working order
(Except the Rx1 scale.)
20 kΩ/v
Tolerances 3 to 4%
Made in Japan
The Hioki site says on the Discontinued Products page: - "TESTER TX-20 Before 1980 Not Available".
They still make a similar product. I will presume I bought it in the late 1970s.
It is a very simple design, the manual even has a parts list and circuit diagram. There is no fuse mounting, just a length of fuse wire soldered across a gap on the circuit board. However, in addition to the manual, there is an explanatory sheet with some spare fuse wire stapled to it. The manual has the familiar charming English strangeness seen today with Chinese products, such as: "It is a very delicate instrument, but if correctly handled, it can be used for long years."
I was surprised to find it still had a battery (strictly, a 1.5V cell) in it; I would have thought I would store it without, since batteries can eventually go bad, leaking and causing corrosion which can even render equipment unusable. I must have used it not too long ago, as the battery was perfectly intact, not even flat, and well within usable charge level.
The pot for zeroing the resistance scale caused a bit of erratic needle-swing, but settled down well after a few wiggles back and forth to scrape off the suface oxidation.
This meter served me well for about 15 years and is still in working order, excepting that in the course of comparing the meters I found the RX1 scale is reading very low. (See test results below.)
The Sanwa CD-720C Digital Meter
It has lasted pretty well.
10 MΩ
Tolerances 1 to 2%
Made in Japan
It is a bit hard to date this exactly. It came as part of a nice toolkit in a briefcase, a couple of which were supplied with the NEC telephone exchanges I worked on installing and maintaining in the late Eighties to mid-Nineties. The toolkit was still in my vehicle when I moved to data wrangling in an office, and has stayed with me ever since. I will guess around 1990.
It features auto or manual range selection, a data hold function which locks readings temporarily, and a continuity tester with beeper of such high frequency I can no longer hear it. A nice touch is the mount for a spare fuse, inside the case.
The cell holder terminals are a bit corroded, as I had some cells go bad once. I am suspicious that there might be some damage I can't see that has finally made it go faulty. The fault causes incorrect readings that slowly fall, see test results below.
I would guess there is either a high resistance short to some capacitor, or a dodgy capacitor, but as there is nothing visible on the circuit board, and no capacitors look swollen (a common fault), the actual fault would be hard to find.
It has lasted pretty well.
10 MΩ
Tolerances 1 to 2%
Made in Japan
It is a bit hard to date this exactly. It came as part of a nice toolkit in a briefcase, a couple of which were supplied with the NEC telephone exchanges I worked on installing and maintaining in the late Eighties to mid-Nineties. The toolkit was still in my vehicle when I moved to data wrangling in an office, and has stayed with me ever since. I will guess around 1990.
It features auto or manual range selection, a data hold function which locks readings temporarily, and a continuity tester with beeper of such high frequency I can no longer hear it. A nice touch is the mount for a spare fuse, inside the case.
The cell holder terminals are a bit corroded, as I had some cells go bad once. I am suspicious that there might be some damage I can't see that has finally made it go faulty. The fault causes incorrect readings that slowly fall, see test results below.
I would guess there is either a high resistance short to some capacitor, or a dodgy capacitor, but as there is nothing visible on the circuit board, and no capacitors look swollen (a common fault), the actual fault would be hard to find.
The UNI-T UT39C Digital Meter
Has yet to prove itself.
10 MΩ
Tolerances around 1%
Made in Guangdong, China
The meter has a few extra functions the others lack: - a transistor testing socket, capacitance testing socket, frequency measurement, and a temperature probe, but I can't see me using those much. (I have a cheap pistol-grip infrared thermometer which is pretty handy for most temperature measurements.)
The test probes have the added safety feature of a shroud around the ends which plug into the meter, preventing accidental contact with high voltages at that end.
All thse fancy functions must be power-hungry, as it requires a 9V battery to run it. The one that came with the meter went flat while I was doing the tests below.
Has yet to prove itself.
10 MΩ
Tolerances around 1%
Made in Guangdong, China
The meter has a few extra functions the others lack: - a transistor testing socket, capacitance testing socket, frequency measurement, and a temperature probe, but I can't see me using those much. (I have a cheap pistol-grip infrared thermometer which is pretty handy for most temperature measurements.)
The test probes have the added safety feature of a shroud around the ends which plug into the meter, preventing accidental contact with high voltages at that end.
All thse fancy functions must be power-hungry, as it requires a 9V battery to run it. The one that came with the meter went flat while I was doing the tests below.
A big design fault is the range selection switch. A meter can be damaged by connecting to high voltages when set to readings on the resistance scale, (see the Hioki Fault section below!) yet the grip/pointer for the switch is inexplicably symmetrical, and you can see from the picture above that it would be quite easy to misread which end is the "pointing" one. I would have made the non-pointer end squared off, and added a white inlay to the marker on the pointy end.
In fact, I have done the white inlay, using some white correcting fluid, scraped back when dry (See Left). This seems one of those cases where the product design team's non-technical desire for "a nice look" has trumped the technical team's functional design requirements. |
Comparison Tests
The Hioki Fault
I had a think about the Hioki being faulty on only one range. A quick look at the circuit diagram (See Right) shows that the only difference to the meter's circuit when the RX1 scale is selected, is that a resistor called R13 is in circuit, whereas for the RX10 range there is R14 in circuit, and for the RX100 range there is R15.
Therefore, the most likely cause of the fault is R13, which the parts list (See Below) tells us has a resistance value of 100Ω. I took the meter's back off and had a look, and sure enough, we find the 100Ω R13 looking decidedly dodgy, with a blackened crusty bit on the exterior (See Below, Right). The UNI-T meter tells me it is open-circuit. That is, it may as well be made of air. I would guess I have at some stage tried to check a car battery or Mains voltage with the meter still set to the RX1 range. |
This fault looks decidedly repairable. Finally, one day, 23/08/20, THIS LINK became active, where a reader who might be interested can see an account of the repair, because I got round to doing it, and it was successful.
The Sanwa Fault
This fault doesn't seem as if it would be easy to find, and if in the main IC would be unrepairable. However, there is a chance it has something to do with the corroded cell holder terminals. Finally, one day, 31/08/20, THIS LINK became active, where a reader who might be interested can see an account of the repair, because I got round to doing it, and it was successful. |