Oscilloscope to measure DC current drop over time

Thread Starter

Lesliev

Joined Jan 13, 2021
14
Hello,
Very new to using my digital oscilloscope and may be asking for something that is not possible on an Oscope.

I am testing a 6v LED lamp light (Orbitec LE-G4 3w)using two different battery types - Nimh and Lipo. With the Nimh I am using 4 AAA battery (rechargeable Eneloop) to get the required 6v and for the 4.7v Lipo I am using a Polulu Booster to up the 4.7v to 6v. While I can perform the calculations to determine the amount of time the 3w lamp should remain on I am also interested in measuring the drop in brightness the lamp shows when running on the Nimh batteries. I have set up a test using a light sensor breakout and an Arduino and from the time stamped results appearing in the IDE monitor I will create a graph of the brightness variation/decrease - I expect it might be a straight line but not sure what I will actually get. I would also like to measure the voltage drop as a component of that brightness decrease (again probably a straight line graph) and while it is easy to take a voltage measurement with the oscope or Multimeter and see it decrease over time I cant see how to record the decrease over time so I can graph it or if that is even possible? The Polulu and Lipo arrangement in theory should not show any decrease in lamp brightness until input voltage from the Lipo powerpack drops below a threshold for the booster at which time the lamp will simply go out.
Long winded explanation but hope an outline of what I am trying to achieve will allow some comments and advice about my process as well as any advice on recording the voltage drop over time. OR is all of the measurement unnecessary and two straight lines on a graph is all I am going to get anyway:)
Thanks in advance for any assistance.
 

MrChips

Joined Oct 2, 2009
23,541
You did not say which digital oscilloscope you have. Nevertheless, most oscilloscopes go down to sweep times in the order of seconds, not hours which is what I thing you will need. Perhaps there might be a way to set up an oscilloscope with advanced data logging functions that can be triggered to record voltages at long intervals.

You would be better to use the Arduino with an analog front end to record time-stamped data.

Note that the brightness of an LED as measured with any light sensor is very unlikely to be linearly related to the voltage of the LED.
 

DickCappels

Joined Aug 21, 2008
7,442
There is something to be said for using a pencil and paper to record the readings taken with the instrument of your choice. You can then plot the data with a spreadsheet program, Excel for example. Kind of old school but that's what we did when there were no personal computers.
 

nsaspook

Joined Aug 27, 2009
8,390
If your scope has a 'Roll' mode then the slowest sweep can be used as a graph of input level changes. For my SDS-1202X-E 50 Seconds per screen division is the most you can chart on a single sweep. With SCPI you could command slow roll sweep, query the completion, download the data and restart another sweep continuously. I've not tried this configuration but it should be possible along with a single data point query from the scope using LXI Tools.

https://siglentna.com/USA_website_2014/Documents/Program_Material/SIGLENT_Digital_Oscilloscopes_Remote Control Manual.pdf

 
Last edited:

Reloadron

Joined Jan 15, 2015
5,934
A Google of Arduino Data Logger will get you plenty of hits. Additionally a Google of Arduino Data Logger Shield will get plenty of hits on shield units which will piggy back right on an Arduino Uno board allowing you to collect and save data. Anyway, what you really want is a data logger function. Now if your scope has that capability you should be set, if not then try the Googles I suggested.

Ron
 

sparky 1

Joined Nov 3, 2018
544
The equivalent light output of a bulb run on DC would be a close approximation to the output amperage of a bulb using complex
AC source. The role of the oscilloscope would be to match out put levels. Because a complex AC source may change over time,
The ability to sense current and port data would allow testing under different system conditions.

Dick Cappels point is well taken that hand calculation is a common ground to establish within an peer group that trying to evaluate such systems .
Some technologies even after being evaluated by competent scholars has for a long record of non acceptance.

The first diesel truck was made in 1923 now 98 years later drivers are saying don't try to change my Peterbuilt truck.
We can see how parts built for continuous work need to undergo many miles of service, whereas acceptance of battery devices
are more planned obsolencent shortlived cost effective novelties when compared to yesterdays dynamos.
The challenge to advance from shortlived portable battery dilemma is seen in our quest to build better test instruments
and more importantly in having better understanding power electronics systems.
 
Last edited:

djsfantasi

Joined Apr 11, 2010
7,693
A Google search of “Arduino to Excel” returns several hits that allow you to send data collected by an Arduino sketch directly to Excel running on a PC.

...With no special hardware.

I thought you may find this of interest.


EDIT: added underlined text
 
Last edited:

nsaspook

Joined Aug 27, 2009
8,390
My next suggestion would be to sample with a high-precision ADC with at least 12-bit resolution. The small changes in voltage with small current changes will be lost with most full-scale measurements using 10-bit ADC hardware. Sure, you can oversample to maybe gain a few bits (in theory) if there is noise but it's been my experience you won't gain accuracy when monitoring lower impedance battery operated circuits.
 

Thread Starter

Lesliev

Joined Jan 13, 2021
14
You did not say which digital oscilloscope you have. Nevertheless, most oscilloscopes go down to sweep times in the order of seconds, not hours which is what I thing you will need. Perhaps there might be a way to set up an oscilloscope with advanced data logging functions that can be triggered to record voltages at long intervals.

You would be better to use the Arduino with an analog front end to record time-stamped data.

Note that the brightness of an LED as measured with any light sensor is very unlikely to be linearly related to the voltage of the LED.
Thank you Mr Chips for your responses, very helpful and much appreciated. I am particularly interested in your note regarding the the LED brightness and voltage. In my naivety i imagined that as the battery drains and i can see the dimming of the LED i am using the voltage and brightness would in fact relate linearly but from your comment i understand you doubt that and am hoping you can help me better understand the reasoning and what sort of relationship i might expect to actually find. Hope to hear from you again. Regards
 

MrChips

Joined Oct 2, 2009
23,541
You have to look at the data sheet of the specific LED you are using.

The first thing to note is that the relationship between LED Vf and If is non-linear. In fact it follows an exponential function.
Hence you want to measure LED current, not voltage.

Having measured the LED current, the LED brightness is also non-linear. Then added to this, the response of the light sensor to intensity would need to be determined. And on top of all of the above, we have yet to consider the effect of temperature on both the LED and the light sensor.

LED Characteristics.jpg
Reference:
http://www.farnell.com/datasheets/1498852.pdf
 

Audioguru again

Joined Oct 21, 2019
3,201
You do not get 6V from four Ni-MH Eneloop battery cells. It will be about 5.4V that will drop to 4.4V. If the cells are AAA size with a 800mAh rating then 4.4V will be reached in about 2 hours if they were properly fully charged.
Of course the LEDs will become dimmer as the battery voltage runs down. Since you gave no LED details then we do not know the lowest voltage they need before they stop producing any light.
 
Top