Digital Multi-Meter not sensitive enough?

[ptolemy]

Hi guys,

I am using Equus 3320, which came highly recommended; however I have a problem with it.

It's not too precise. I have a direct drive sst-50 that I know goes near 4amps with imr18650 and 3+ amps with other cells, but it's barely going over 2.5a. Same happened in several other past lights.

I assume that either the unit is not sensitive enough or equipment to handle 2-5amps with .1 precision or the probes/connecors (I use default ones) are not made for it.

I seen some folks use 14gauge wires and plug 1 end directly into DMM and other to light, and they appear to get much bette readings. Is that the issue? Or is there are better probes/units out there (within a certain price range)?

Anything else can be done to achieve a preciser readings? After all, if it's not that precise, then it's useless

Thanks for the help:)

 

[Curious_character]

There's always a voltage drop across an ammeter -- it's that voltage drop, in fact, that the meter is actually measuring when it shows you the current. Some meters have more than others. In general, you'll get less drop on the highest scale, usually 10A. The problem with the drop is that it drops the voltage to the circuit being measured, which in turn changes its current. The effect is particularly strong when the supply voltage is low like it usually is with flashlights.

What I do to measure current with minimal drop is to run the current through a piece of wire with known, low resistance and measure the voltage drop across it. The limiting factor is the resolution I can get. For example, #23 copper wire has a resistance of 20.3 ohms per 1000 feet, or 20.3 milliohms per foot. So a tiny bit less than 6 inches of it has a resistance of 10 milliohms. Run current through it, and you'll see 10 millivolts across the wire per ampere of current. That's the voltage drop, 10 mV/A. My cheap DVM reads to 0.1 mV resolution on the 200 mV scale, so I can resolve 10 mA of current -- 1.080 A +/- 0.005 reads 10.8 mV. You can check and calibrate the system by putting it in series with another ammeter, which should read the same current regardless of the circuit voltage.

If you use this method, be sure you're measuring the voltage drop only across the wire itself, and not including contacts at the ends.

c_c

 

[Popsiclestix]

The effect is particularly strong when the supply voltage is low like it usually is with flashlights.

It's not the low voltage that's the issue for the ammeter here. It's the high current that you're asking it to measure. DMMs measure currents at low voltages and low currents quite accurately. (ie. integrated circuits). They are likewise as inaccurate at high voltages and high currents.

I seen some folks use 14gauge wires and plug 1 end directly into DMM and other to light, and they appear to get much bette readings. Is that the issue? Or is there are better probes/units out there (within a certain price range)?

Thicker probes are a start. The resistance of a wire is inversely proportional to its thickness. If you want to get really fancy, you can make probes out of silver wires for that last 5%. IIRC, the resistivity of Silver vs Copper is something like (1.7 Ω*m vs 1.6 Ω*m).

You may also want to try magnets soldered onto end of wires. If your probes are as sharp as mine (I can do surgery with them :ohgeez:), they won't measure anything beyond a few hundred of milliamps (if that) with any sort of precision.

An alternative solution is to buy a clamp ammeter. They typically start at $50 and prices can skyrocket from there depending on the features. These work by measuring the magnetic field generated when current passes through a wire. Thus it does not directly introduce resistance into the circuit.

 

[ptolemy]

thanks for the replies so far, but i think either I or you guys over complicating it.

i simply want a fairly repliable way to measure amps being pulled by different lights. if it pulls 4amps anything +- .10 amp (ie 3.90 or 4.10) accurance is fine for me, but not +-1amp...

so, i am just trying to improve on it.

i guess let me ask different question:

would a $200+ fluke multimeter like 87-5 with default probes/etc measure it more accurately than equus 3320 or same issues would arise>?

thanks!

 

[Mr Happy]

It's not too precise. I have a direct drive sst-50 that I know goes near 4amps with imr18650 and 3+ amps with other cells, but it's barely going over 2.5a. Same happened in several other past lights.

Direct drive is the problem here. With direct drive the current is incredibly sensitive to the circuit resistance. A typical meter with typical probes will insert a lot of resistance into such a circuit, relatively speaking. More resistance equals less current flowing in a direct drive arrangement.

The current measurement you are seeing is likely quite accurate; it simply indicates that less current is flowing than you expect as a result of the meter resistance affecting the circuit.

What you are seeing is known technically as the "burden voltage" of the meter. Even very expensive meters have this phenomenon.

What you need to do is what others have suggested: either use a DC clamp meter, or measure the voltage drop across a calibrated low resistance element of the circuit in question.

 

[ptolemy]

Direct drive is the problem here. With direct drive the current is incredibly sensitive to the circuit resistance. A typical meter with typical probes will insert a lot of resistance into such a circuit, relatively speaking. More resistance equals less current flowing in a direct drive arrangement.

The current measurement you are seeing is likely quite accurate; it simply indicates that less current is flowing than you expect as a result of the meter resistance affecting the circuit.

What you are seeing is known technically as the "voltage burden" of the meter. Even very expensive meters have this phenomenon.

What you need to do is what others have suggested: either use a DC clamp meter, or measure the voltage drop across a calibrated low resistance element of the circuit in question.

ah, i think i understand the problem :) thank for spelling it out for me

with that said, why do people use direct wires into probes instead of using the provided leads...i assume either leads aren't rated for high current or wires provide less resistance?

 

[LedTed]

The absolute best way of making this sort of low resistance measurement is via a SMU through utilization of custom connectors and a gold platen.

But, that would be moving into the realm of very expensive specialty equipment.
I hope this helps you make a decision.

 

[HKJ]

would a $200+ fluke multimeter like 87-5 with default probes/etc measure it more accurately than equus 3320 or same issues would arise>?

It would also be an issue with the expensive meters, only way around the problem is a DC clamp meter and you might need to search a bit for one really useful at the "small" currents used with flashlights.

 

[MarNav1]

I recommend these leads highly. You'll have to buy the appropriate ends too. It will increase the flexibility of your meter as well. Notice these leads are 10 amp rated, many are not. Are they cheap? Nope. Also, no comparison of a 87-5 and a 3320 but I'm sure you know that. And at 10X the price there shouldn't be. Anyway, these leads are uber nice, you'll like them.
http://www.tequipment.net/FlukeTL224.html

 

[Curious_character]

It's not the low voltage that's the issue for the ammeter here. It's the high current that you're asking it to measure. DMMs measure currents at low voltages and low currents quite accurately. (ie. integrated circuits). They are likewise as inaccurate at high voltages and high currents. . .

Sorry I didn't explain more clearly.

The drop caused by the meter is independent of the circuit voltage and depends only on the meter characteristics and the current being measured. If the circuit under test is fed from a relatively high voltage, then the drop caused by the ammeter reduces the circuit voltage by only a small fraction. But when the circuit voltage is low, as in a flashlight supplied by a single cell, the same meter drop causes a significant fractional drop in the voltage supplied to the circuit. Therefore, the effect of the meter drop on the circuit being measured is greater.

The accuracy of the measured current is, as you say, the same in either case. My point is that the meter disturbs the circuit being measured, and for a given meter and current this disturbance is greater when the circuit voltage is low. So you end up accurately measuring the current, but in a circuit that's operating differently than it would if the meter wasn't connected.

c_c

 

[ptolemy]

thanks everyone:)

 

[march.brown]

If you have two meters , you can measure the resistance of the Ammeter on one meter with the other one on Ohms.

After deducting the lead resistance , one of my meters is 0.1 ohms on the 10 Amp range , 1.0 ohms on the 200mA range and 10.0 ohms on the 20mA range ... It is just not possible to check current accurately this way on a low resistance device , as the meter and leads resistance is greater than the LED circuitry plus the battery internal resistance.

It has already been mentioned that the best way is to use a very low value resistor and measure the voltage across it ... You can then work out the current easily ... You must know exactly what value the resistor is.

When you measure current , the resistance of the leads is also to be considered ... This in itself can be half an ohm or so ... When measuring DC voltage , the impedance of the meter is so high that the lead resistance can be ignored.
.

 

[MarNav1]

Fluke used to make an external shunt for measuring current, I don't know if they still do or not. Good for high currents for extended periods.

 

[ptolemy]

Friend of mine has a spare fluke 322 clamp meter, so we traded.

it should be here in a few weeks...hopefully it can help :)

 

[Meterman]

The Fluke 322 clamp meter can't measure DC current so it won't be of any value to you. :mecry:

Wulf

 

[VidPro]

put better leads on the meter just for doing amps
then test with it connected, then disconnect and test Straight without meter
Then GUESS about where it is really at :-)

other methods use Bench power supply to Simulate battery:
read battery voltage under the load (ampmeter out)
set up bench supply to same voltage battery would be at
raise current on bench supply carefully and read the current (at simulated battery voltage).

by knowing the battery voltage, then simulating it with a external bench supply you can get a much better guess as to what will occur.
Remembering that it is the voltage of the battery UNDER that actual load, not some voltage you get from the battery without the same load.

 

[MarNav1]

One option but not really economical unless you do alot of testing.............
http://www.tequipment.net/Flukei410.html

 

[Lynx_Arc]

one method to find the current is to measure the voltage across the battery when running the light, get two meters and a power supply and hook one meter up across the power supply (the one you measured the voltage with) and the other to measure the current. You can then replace the battery with the power supply and crank the voltage up to the battery voltage measured and then get the current reading at that voltage.