Replacing battery in digital multimeter with AC/DC power adapter

[Frank Sanford]

I would like to do stationary measurements with my UNI-T UT61D handheld digital multimeter for extended periods of time without having to change its nine-volt battery. I figured I could connect the multimeter to a 9V regulated switched-mode AC/DC adapter. Can this go wrong? Will it affect measurement reliability or could it even lead to failure of the multimeter? Do handheld multimeters with optional external power supply exist or is it generally a stupid idea?

[Kleinstein]

An external power adapter is somewhat tricky, as the DMM has no internal insulation between the battery and the measured voltage. So one has to be careful when measuring high voltages. No more CAT 2 or even CAT 3 rating.

With s switched mode adapter there can also be capacitive coupled current to ground. In some circuits this can cause trouble and might even damage a sensitive circuit when you connect the common side to the wrong place.

[alm]

In the majority of the battery powered meters one of the battery terminals is directly connected to the common terminal. So if you would connect a standard power brick with barrel connector to the battery terminals and connect the black lead to a high potential (e.g. mains), this potential would be on the barrel connector and the lead to the power brick. Both are only designed to handle a safe, low voltage, and would not provide adequate protection from dangerous voltages.

I believe Fluke had some ScopeMeters that used an external power supply, but this supply used a special, much deeper, power connector with much heavier insulation.

If you want a mains-powered multimeter, I would look for a cheap bench meter (either new or used).

[Paul Moir]

Generally you don't see it because it ruins the safety of the meter:  there's a connection between the probes and the jack.  So it depends on what you are measuring.  You'll want to use a pretty well filtered / low noise supply so more likely an old transformer type AC adapter rather than switch mode.  Since these are rarely regulated, you would need to make sure the voltage is no higher than a 9v battery or better yet provide regulation yourself.

[Gyro]

Hi, and welcome,

Sorry, no, bad idea...

- As soon as you connect to the battery terminals, you will invalidate the input protection and insulation of the meter.

- Neither of the battery terminals is connected to the common input jack of the meter, the battery is intended to 'float' (based on internal reference circuitry) relative to the inputs. Leakage current from the adapter will affect the operation and accuracy of the meter.

Short answer.. you will, at minimum, get mis-operation or inaccuracy. Worst case, yes you will blow up the meter. Time for a re-think, like a mains powered bench meter.

In the majority of the battery powered meters one of the battery terminals is directly connected to the common terminal.

EDIT: Actually, for most battery multimeters, I don't think this is the case, as above. The meter must handle both positive and negative inputs, so the battery must be biased so that negative inputs fall within the supply range.

[alm]

EDIT: Actually, for most battery multimeters, I don't think this is the case, as above. The meter must handle both positive and negative inputs, so the battery must be biased so that negative inputs fall within the supply range.

So instead of 230 VAC you might have 230 VAC + 2.5 VDC. Not sure if that is much safer. With diodes usually failing short, I wouldn't rely on them protecting you from whatever potential might be on the input jacks.

[Ian.M]

There is no good way to do this - if you mod the battery compartment to get a dc power cable out you've compromised the measurement category rating of the meter.  At the very least, if you want to go down this dangerous road, use a >5KV isolation DC-DC converter inside the battery compartment - don't trust the isolation of an ordinary mains PSU.  However ripple from the external power mod is likely to seriously compromise measurement accuracy so it might all be for nothing

[Gyro]

So instead of 230 VAC you might have 230 VAC + 2.5 VDC. Not sure if that is much safer. With diodes usually failing short, I wouldn't rely on them protecting you from whatever potential might be on the input jacks.

That wasn't my point - the protection is gone regardless! It's the  fact that leakage and noise from the adapter will screw up the internal biasing of the meter (simply because neither battery terminal is connected directly to the common input jack).

[Kleinstein]

The direct connection is a safety problem. So the better way would be to use rechargeable batteries: they usually last for at least 50 hours - so should be good enough for a single use.

The only safe way would be a high insulation voltage rated DC/DC converter inside. The normal cheap DC/DC blocks are not that good in insulation and often produce quite some RF noise and could thus have a negative influence of the accuracy. As the space is still needed for the battery there is usually no way to add this to an existing meter.

[Nusa]

What about inductive power, like many smartphones use for charging these days? That would maintain case integrity.

[Ian.M]

Its theoretically possible, but I'd be concerned about induced voltages in sensitive parts of the circuit, and heating of the shielding.

[David Hess]

The danger in using a common floating power supply is that the transformer is unlikely to be designed to sustain high primary to secondary voltages which a multimeter could encounter.  They make special transformers for this.

They make special high isolation power supplies for this type of thing.  I have one that is the size of a big plug-in relay that provides 10s of milliamps and has a transformer with a primary to secondary air gap of like an inch.  The really high isolation power supplies use optical, mechanical (motor and generator), or sound (ceramic transformer) coupling.

This still increases the common mode capacitive coupling but in most meters this is 100s of picofarads anyway.  Bench meters solve this problem internally usually with a separate high isolation inverter or switching power supply.

[Neilm]

In addition to all the safety issues raised above there other potential issues. A battery is a noise free power source. A switched mode supply will have some output noise - this would be conducted into the instrument and would cause noise on internal power rails - there may not be filtering on these as the designer is expecting a battery.

[bitseeker]

Cheap solution: Rechargeable batteries
Expensive* solution: Bench multimeter

* Cheaper way to achieve the expensive solution: Get a GDM-8251A while they last.

[jeroen79]

I believe Fluke had some ScopeMeters that used an external power supply, but this supply used a special, much deeper, power connector with much heavier insulation.

I have a PM97 and it has an external power brick.
It consists of just a mains transformer, rectifier diodes and smoothing capacitor.
It has a longer plug that goes deeper into the meter.
A switching converter in the meter then converts this into the needed voltages.

What does the OP want to use this for?
Low voltage hobby projects?
High voltage professional equipment?
Could the voiding of any certifications become a liability?

[mariush]

A lot of switching power supplies have a Y capacitor between the primary and the secondary of the transformer inside, in order to reduce EMI. So there's some very small amount of current flowing and in some cases you can even feel it if you touch the connector, but it's low enough that it's extremely unlikely to even get more than a buzz or some sparks.
But just the presence of that connection could affect the isolation of the multimeter, or affect it.

And even if there's no such capacitor or you remove it, you'd still have the switching ripple (you'd see the voltage fluctuating at a high frequency if you'd have an oscilloscope), unless the switching power supply has a very good output filter (at least a capacitor - inductor - capacitor)

Linear power supplies ( old style, low frequency,  transformer -> bridge rectifier -> bulk capacitor -> (optional but highly recommended) linear regulator

would be fully isolated and should not affect measurements in any significant way.

later edit:  Another suggestion would be to make sure everything's extra isolated by using one of those digital isolator chips, like the expensive Adum5000 for example: http://www.digikey.com/product-detail/en/analog-devices-inc/ADUM5000ARWZ/ADUM5000ARWZ-ND/1972297
The chip in the link above takes in up to 5.5v DC and outputs 3.3v or 5v at up to 100mA with a lousy efficiency of 33% ... which doesn't matter since you wouldn't use it with batteries. It's more important that it's small and could very well be placed inside the multimeter in the place of the battery and have just a DC In jack on the battery cover or something like that.
The current is also not a big deal, since the multimeter won't use more than maybe 3-4 mA.

[RoGeorge]

Replacing the battery in a DMM with AC/DC power adapter is a bad idea.

If you are doing this to save money, use a 9V rechargeable battery instead.

[MrAl]

Hello there,

I actually did this a long time ago and used the meter for years like that.  The only trick is that the power adapter is double isolated so it does not connect to ANYTHING else except the battery terminals of the meter.  Of course you also have to be more careful if you are working with higher voltage.

The scheme i used was very simple because i was in a hurry one day.  It's just a 9v DC wall adapter and a 9v zener diode with series resistor to limit current.  Dont remember the value of the resistor but the meter only draws something like 1ma max or so.

The wall wart was an unregulated type so the DC output with very light load is higher than 9v thus the need for the zener.  If you have a regulated wall wart that would not need the zener.

I used this to monitor battery voltage for hours and days at a time.  It's like having a mini bench meter.  The meter itself was only 6 dollars and i had the wall wart from something else discarded a long time ago.  I dont buy cheap meters like that anymore though.

And yes, the multimeters usually do not ground one lead of the battery.  Some panel meters allow that but most multimeters i have seen do not ground one terminal of the battery that powers the meter.  So you cant measure the battery voltage with the same meter that it is powering.  I have panel meters that can do this however.

[Seekonk]

I do a similar thing with a one of those free DVM I use as a monitor.  I use a DPDT relay (I have boxes of these from mfg evaluations) and a push button to activate it.  The meter really only needs to operate when I look at it.  So, a capacitor is connected to the supply in the deactivated position to charge. Pressing the push button switches that capacitor to the meters power.  Totally isolated and no batteries needed. Just a 2200uF charged to 9.5V will run the meter for 30 seconds.

[med6753]

My Fluke 8021B has a jack for external power but I've never used it. I checked the manual and it does indeed warn to use only a Fluke A81 external supply and not a standard "calculator" type supply as it does not provide adequate common mode rejection.

[jeroen79]

Replacing the battery in a DMM with AC/DC power adapter is a bad idea.

Then how are bench multimeters working safely on mains AC?

[SeanB]

They typically use a transformer designed for low primary to secondary capacitance, generally with the windings on separate bobbins on the cosre, and also rely on the transformer being rated for 2kV between primary and secondary side, easy to do in as 50/60Hz magnetic core that has been high pot tested to survive that in manufacture. Then they use a generous air gap ( there tends to be a lot of spacing between mains side and secondary side, even around the mains switches) to give the isolation, and this gives them the ability to have 500VDC rating between ground and the mains.

There are a few meters like that, good for being powered on all the time, and you also get ones with a built in battery allowing both mains and battery operation.

[David Hess]

Or they use a normal off-line power supply and a separate inverter to provide the floating supply and galvanic isolation which has the advantage of using a smaller transformer core which is easier to design for high isolation and low capacitance.  Gate drive transformers are suitable.

[RoGeorge]

Replacing the battery in a DMM with AC/DC power adapter is a bad idea.

Then how are bench multimeters working safely on mains AC?

Of course, replacing a DMM battery with an AC/DC adapter will not lead to instant death, but this kind of modifications might have side effects in a wide range, starting from incorrect measurements up to burning the house, or worst. In general, any modifications is not a good idea, unless one know very well how to do it.

Bench multimeters are safe when powered by mains AC because they were designed to work that way. Various safety certifications required by the industry were implemented by professionals. Moreover, the final product was tested against various safety standards and regulations.

[lordvader88]

Ok then what about using an AC/AC transformer, and then regulating it(after FWR of course) with an LM317 or LM723 ? I can add a fuse and current limit it too. I have a little step down transformer from a radio with low voltage outputs.

I am only boarding little 12V circuits, which themselves run off a SMPS, so shouldn't that be fine for now until I get a nice rechargeable battery ?

[mariush]

You could use a digital power isolator, like Adum5010 or Adum5000 for example : http://www.analog.com/media/en/technical-documentation/data-sheets/ADuM5000.pdf

Adum5010 : https://www.digikey.com/product-detail/en/analog-devices-inc/ADUM5010ARSZ-RL7/ADUM5010ARSZ-RL7CT-ND/3821204

It converts 5v to 5v or 3.3v with around 30% efficiency, but considering a multimeter only uses a few mA you don't really care.  If the multimeter normally runs on 9v battery and shows low battery with just 5v, you could have a tiny boost regulator on a circuit board to boost the voltage to around 6-7v, or whatever the threshold is for low battery.

Or you could have something like this isolated 5v->5v converter which easily fits in the 9v battery compartment : https://www.digikey.com/product-detail/en/murata-power-solutions-inc/NXE1S0505MC-R7/811-3040-1-ND/5047366
 
And if you want to go a step forward you could add a super capacitor after the digital isolator to charge up and keep the boost regulator powered for tens of minutes after you disconnect the power adapter (but note that it may take a couple of minutes for super capacitor to charge enough to use the multimeter and then a few more minutes to charge to 100% while you can use the multimeter still plugged in the adapter).

dc in (5v ) ---> isolated converter / digital isolator 5v-> 5v -> boost regulator to 5v .. 9v

dc in (5v) ---> isolated converter / digital isolator 5v --> lithium charging IC ( MCP73811 for example) --> tiny 100-400mAh lithium battery -> boost regulator to 5v..9v

dc in (5v) ---> isolated converter /  digital isolator 5v or 3.3v -> ldo to 2.5v -> super capacitor (10F+ 2.75v) --> boost regulator 

[lordvader88]

Money is the problem, the 9V battery died and I don't have $10. I have 3 other meters, + multi-tester, and a little LCD voltmeter, so at least I'm not suck, but I only have 1 9V that barely works in the little 1, and the 3V 1's dial is shorting somewhere so its a hassle change dial setting. So for now I have to use a combo rather than just my main DMM

Also I already breadboarded an adjustable LM723 for another transformer, so its ready