Measuring current ???

[NewBie]

[ QUOTE ]
haveblue said:
For simplicity if your light was running at 100% efficiency then you would get 4.5 (4 hour, 32 minutes) of light before it dropped below the 50% light output. The heat generated by the light is inefficiency. Say the light is 75% efficient because of heat loss then the run time would be about 4 hours.

[/ QUOTE ]

More like 3.4 hours

 

[greenLED]

I need help figuring this out:

I'm getting different current-to-the-Lux readings depending on the setting I use on my multimeter. My DMM has ranges of 200m and 10A. When I set the range to 200m, the current to the Lux reads 114mA; whereas it is 0.13A when the DMM is set to 10A (not a huge difference). This was using an alkie.

The difference is greater when using a CR123 battery (ie. reading=190 when DMM on 200ma range and 0.38A when on the 10A scale).

I did change the leads to the appropriate connection on the DMM, and I'm measuring current by breaking the circuit and placing the DMM in series between battery and Lux. The voltage on the alkie is 1.49, and the CR123 is 2.87V. The circuit I'm playing with is a Micropuck (high current, I fried the regular version tweaking the current output /ubbthreads/images/graemlins/tongue.gif).

 

[PhotonWrangler]

I think what you're seeing is the % of full scale accuracy issue. I would trust the readings on the 200ma scale more than those obtained on the 10A scale because 114ma is such a teensy percentage of the 10A maximum that you're falling into the percentage-of-full-scale error zone.

 

[drizzle]

That's probably true pw but there may be something else involved here too. DMM's measure current by inserting a resistor into the circuit (inside the DMM) and measuring voltage across the resistor. I've been doing some reading and some looking up of DMM specs and I still am confused but my own basic electronics knowledge tells me that to measure a smaller current you would have to insert a higher valued resistor in order to get enough voltage to take an accurate reading. This could give a lower current reading as a result.

Of course if there are active regulating/boosting circuits involved I'm not sure what would happen.

Your best bet is, if you already have a resistor in series in the circuit:
<ul type="square">[*]take the resistor out of the circuit
[*]get an accurate resistance measurement of the resistor out of circuit
[*]put the resistor back into the circuit
[*]start the current flow
[*]measure the voltage across the resistor
[/list]

One thing DMM's are very good at is NOT affecting circuits when measuring voltage.

 

[Doug Owen]

First of all, it's easy to sort out which of the readings is 'real' (that is reflects what typically happens) WRT the 123 cell case.....use your eyes. You should easily be able to see that the higher current is 'real', that is the light level is closer to what it would be without the meter. Right?

Secondly, in this case you're right, the internal resistor (one Ohm, BTW) is a major factor. However, the current is like half of the current using the 10 Amp range....I suspect there's some more resistance there somewhere.

Percentage of range arguments aside (which really have more to do with analog meters), you probably want ot use the highest current range that will give you useful resolution in cases like this. Most often, it'll be the ten Amp one (which avoids extra resistances by switching the leads rather than using the range switch.

Doug Owen

 

[SilverFox]

Hello Doug,

Do I understand you correctly?

With the typical Fluke meter, the 10A range only has the resistance of the test leads to deal with and does not use an internal resistor?

Tom

 

[Doug Owen]

[ QUOTE ]
SilverFox said:
Hello Doug,

Do I understand you correctly?

With the typical Fluke meter, the 10A range only has the resistance of the test leads to deal with and does not use an internal resistor?

Tom

[/ QUOTE ]

No, the DMM measures the voltage drop acorss a (small vlaue) fixed resistor in the meter box. It uses the lowest DC voltage range (typically 200 mV), meaning ten Amps is displayed as .100 Volts so there's a (.1 Volts divided by 10 Amps) .01 Ohm resistor (basically a bit of bare wire) in there for a shunt. The meter leads are extra, typically more than this. Using heavier leads (and/or shorter) makes the overall loss even less preserving the basic accuracy (depends on the ohm value of the shunt only).

Doug Owen

 

[SilverFox]

Hello Doug,

Thanks.

Tom

 

[reefphilic]

[ QUOTE ]
greenLED said:
I need help figuring this out:

I'm getting different current-to-the-Lux readings depending on the setting I use on my multimeter. My DMM has ranges of 200m and 10A. When I set the range to 200m, the current to the Lux reads 114mA; whereas it is 0.13A when the DMM is set to 10A (not a huge difference). This was using an alkie.

The difference is greater when using a CR123 battery (ie. reading=190 when DMM on 200ma range and 0.38A when on the 10A scale).

I did change the leads to the appropriate connection on the DMM, and I'm measuring current by breaking the circuit and placing the DMM in series between battery and Lux. The voltage on the alkie is 1.49, and the CR123 is 2.87V. The circuit I'm playing with is a Micropuck (high current, I fried the regular version tweaking the current output /ubbthreads/images/graemlins/tongue.gif).

[/ QUOTE ]


I've noticed this same effect and did some experiment some time ago. I measured the current of a simple circuit consisting of only a resistor, a LED and a battery. Different brands and models of DMM were used. Each time, the 10A scale will give a higher reading than the milliamp scale. The differences in reading varies from brands to brands and can be up to +-300% even when highest end models from reputable manufacturers are used.

I tried the same experiments but replaced the LED with a filament bulb. The differences in readings between 10A scale and milliamp scale is much smaller(about 10%).

I don't trust the milliamp scale when measuring LED current anymore. I only use the 10A scale since then. When measuring current of a few hundred mA, there shouldn't be much problem. But for measuring lower current readings. The accuracy may be very low. /ubbthreads/images/graemlins/frown.gif /ubbthreads/images/graemlins/help.gif

A

 

[greenLED]

Reefphilic, let me see if I am interpreting your post correctly. You knew the theoretical current of your circuit in your experiments, and concluded that the 10A scale was the most accurate?

I'm a bit confused about your last paragraph. Do you mean accuracy may be very low using the 10A scale?

I found it interesting that the difference in readings was smaller using a filament bulb. Any ideas why this may be hapenning?

 

[evan9162]

The current through a filament bulb is not linear with the voltage across it.

So a small voltage drop introduced by the prescense of the meter will cause the current through a bulb to change less than it would other devices.

In this thread, I plotted I/V curves for various PR-based bulbs:

Current vs. Voltage: Luxeons and krypton bulbs

Bulbs and LEDs are quite different in their response to changes in voltage. In a bulb, a small change in voltage results in an even smaller change in current, whereas with an LED, a small change in voltage results in a huge change in current.

 

[greenLED]

I see, thanks. So, bottomline, use the 10A scale always for LED stuff?

 

[reefphilic]

GreenLED, I noticed that the LED became much dimmer when the milliamp scale was used compared to when the 10A scale was used. The LED is also brighter when the ammeter was removed from the circuit. Since the ammeter cannot be giving extra power to the LED. My conclusion was that the milliamp scale is impeding the current flow. I will do some tests with known theoretical current this weekend.

Since the accuracy of DMMs are stated in the format : 0-10A (± 1.5% ± 20 d). The accuracy will be much lower if you're using it to measure current of 50mA compared to 5A.

Since an ammeter work by introducing a resistor to the circuit and measuring the voltage drop across the internal resistor of the DMM. My guess is that the resistance of the internal resistor is much lower then the resistance of the filament bulb so the "percentage error" is much lower.

Someone please correct me if I'm wrong.

 

[reefphilic]

[ QUOTE ]
evan9162 said:

a small voltage drop introduced by the prescense of the meter will cause the current through a bulb to change less than it would other devices.

Bulbs and LEDs are quite different in their response to changes in voltage. In a bulb, a small change in voltage results in an even smaller change in current, whereas with an LED, a small change in voltage results in a huge change in current.

[/ QUOTE ]


So does it mean that the milliamp scale is virtually useless for circuits that are greatly affected by small drop in potential difference?

Many people suggested adding a small value resistor and calculate the current by measuring the voltage drop across the resistor. Dosen't the ammeter already does that "internally"? /ubbthreads/images/graemlins/confused.gif

 

[evan9162]

[ QUOTE ]
reefphilic said:
[ QUOTE ]
evan9162 said:

a small voltage drop introduced by the prescense of the meter will cause the current through a bulb to change less than it would other devices.

Bulbs and LEDs are quite different in their response to changes in voltage. In a bulb, a small change in voltage results in an even smaller change in current, whereas with an LED, a small change in voltage results in a huge change in current.

[/ QUOTE ]


So does it mean that the milliamp scale is virtually useless for circuits that are greatly affected by small drop in potential difference?

Many people suggested adding a small value resistor and calculate the current by measuring the voltage drop across the resistor. Dosen't the ammeter already does that "internally"? /ubbthreads/images/graemlins/confused.gif

[/ QUOTE ]

The milliamp scale is useful for certian circumstances. Since it usually involves a 1 or so ohm resistor, using it to measure current where high value resistors (100 ohms or larger) are in the circuit means it won't affect the circuit much at all.

The amp meter does use a small value resistor, but using an external sense resistor is better in many ways. First, you eliminate the resistance of the test leads, which aren't used to measure the current, but still introduce resistance into the circuit. Second, you can choose a higher precision/lower value resistor than what might be in the meter already. Finally, putting the sense resistor in the circuit at the beginning lets you measure current without needing to break the circuit in that location to introduce the DMM - you can measure the current at any time with the voltage scale - it saves you from having to change ranges on the meter, and from needing multiple meters to concurrently measure voltage and current.

 

[greenLED]

Could someone fix that scrolling link? Thx.

 

[evan9162]

Sorry bout that - I fixed the link.

 

[greenLED]

Thanks!

 

[NewBie]

Example of the sense "resistor" in the meter:

 

[andrewwynn]

some circuits (like fatman) have a feedback resistor built in.. basically any constant-current circuit will have a built in ammeter... you can measure the voltage off that sense resistor and get a very good reading of the current... example with the fatman.. it has a 0.1ohm resistor between the LED- and the ground... measuring the mV and multiplying by 10 will give you mA.. just put the test leads on right on the pcb not at the led.. because the wire resistance will throw off the measurement.

with many LED drivers.. you can put the meter in series even on the mA scale and the ckt will boost the voltage automatically to compensate for the meter loading... and i've built some 12ga 1' long meter leads for doing current reading with my fluke 87s... very little meter loading.. 2' of copper 12ga wire is only 4 thousandths of an ohm.. if the meter's shunt is 1/100th... than my meter 'loading' is only 0.014 ohms including the meter leads (plus any connection losses).... if you have 6' of typical 24ga meter leads.. that's at least .18 ohm resistance (a typical DMM will measure it's own leads to be between 0.2 and 0.5ohm).... so a typical set of meter leads will have 13 times the loading of the special leads i made which work very well for direct reading of current in a typical LED circuit.. and often it's more convenient so i recommend making some heavy-duty meter leads like i did.. i used some 12ga monster cable.. very flexible... going to extremes i made them 1' long... doing it again.. i'd maybe make them 18".. 1' is mighty mighty short.