So we recently did some surgery on some of these very cheap night lights and I kind of thought I’d improved them a lot
but I did also think how I actually made it worse.
So what I wanted to do was check the actual power usage of the original board and compare it against the radar device
and see if I’ve actually made these night lights much worse.
So I’ve got my Nordic power profileer kit hooked up, so let’s power things on.
So if we turn this on, enable the power output and then we start logging.
Obviously, initially the lights came on and now they’ve gone off because…
Where have they gone off? What’s happening?
The power is drawing 0.8.
There we go, the light is on.
OK, so we’re drawing 80 milliamps at the moment because the LEDs are on.
So if we give it some time, hopefully the LEDs will turn themselves off and we’ll go back down to a lower current.
Now I have blocked off the light dependent resistor.
It doesn’t seem to make very much difference on the current draw when it’s in low power mode.
So let’s give it some time and we’ll see what numbers we get out for this original device.
OK, so the LEDs are out and it’s gone into “waiting for movement” mode.
Now what’s really interesting with this device is you can see that every so often
it goes into this slightly weird mode.
This is like an ultra low power mode and it seems like it goes down to a really tiny amount of current.
So that’s 30 microamps and then occasionally jumps back up to about 0.8 amps.
So I think what it’s doing is it’s going to sleep a lot and only waking up when it actually needs to.
And so long as I don’t move it stays in this ultra low power mode.
So if I have to put this back to live you can see it scrolling.
If I do some movement it does seem to wake up into the more normal mode
but what’s really annoying it doesn’t actually activate
and it is really hard to actually get this to actually light up.
Now I don’t know if it’s because my little bit of sticky stuff over the LDR is not good enough
but the problem I had with these before is they just wouldn’t light up properly
which is really annoying if it’s supposed to be a night light.
Let’s make sure this is really on so that it’s covered up.
And let’s make sure we’re actually connected to power just to become disconnected.
So oh there we go that triggered.
Okay the light’s off.
So we go down to super low current 0.7 milliamps
and then if I stay still it’s switching to this funny mode
where it’s only waking up occasionally.
Let’s see if I can get it to trigger again.
So here we go I managed to get the trigger that’s good performing much better than usual.
But if we scroll back to that ultra low power part then if we just select some of this
then the average current is just 72 microamps.
That’s pretty amazing.
And then in the period where it’s kind of monitoring for movement
the average current is 0.8 milliamps around 0.8 milliamps on average 700 microamps.
So that’s a pretty impressive low current draw.
Obviously when the LEDs are lit it is drawing around 80 milliamps.
That’s what we measured last time.
So let’s compare that to the radar version.
So I’ll just disconnect this.
And let’s plug in the radar.
But before we measure that a quick shout out to PCBWay who’ve been making a bunch of PCBs for me.
Check out our link to them in the description.
They do a great job and I’ve got some great PCBs to play with.
Okay so we’ll switch back to live view.
This is a very… I’ve got a big resistor for this little LED here.
So it doesn’t draw much current when the LED is on.
And we do need to wait about… I think it was about 40 seconds for this to time out.
And then we can see the current draw when this isn’t driving an LED.
Okay the LED is out and we’re now measuring around 2.85 milliamps.
So 2.85 - 2.9 milliamps.
Let’s see if I can move very slowly and see if I can select some without triggering the radar sensor.
So yeah, so on average 2.86 milliamps.
So that’s quite a bit higher than the 0.7 milliamps that the original gets
but it’s not as high as I thought it would be.
It’s actually pretty reasonable.
So around kind of… let’s call it three milliamps.
We’ve got an 850 milliamp hour battery which should actually last quite a long time.
If we run this again it will trigger.
But what I was interested in is does making the LDR visible actually change anything on this?
Or does it stay the same around three milliamps?
So let’s wait for this to turn off the LED again.
Okay so the LED is out and it’s pretty much exactly the same.
So it doesn’t do anything clever with the LDR.
It’s purely gating the output rather than disabling the entire IC.
That’s pretty interesting.
What I might do… let’s remove the LED.
If we wave our hand around does that do anything?
There’s a small amount of change based on is our hand moving around?
Nothing obvious though.
Yeah, seems pretty much the same.
But I’m going to say it’s drawing a lot more power but it’s not that much more power.
And with our 850 milliamp hour battery we should get quite a lot more.
Let’s actually crunch the numbers and see if my improved version is actually improved.
So let’s look at the original.
So it runs around 72 microamps on average.
It comes with a 350 milliamp power battery.
In theory that should last for 200 days.
Now it definitely doesn’t but that could be the batteries to start up to spec.
And also it doesn’t trigger very well.
If I could get these to trigger properly they’d be really good.
Now what about our new improved version?
2.8 milliamps.
I’ve put an 820 milliamp power battery.
That should only give us 12 days.
That’s pretty rubbish.
But all is not lost.
I do have these 2000 milliamp power batteries so I might try putting one of these in.
That should give us around 30 days between charges which is actually not too bad.
But if I could only get the original versions to trigger properly they could be absolutely amazing.
So I might investigate that.