12v in and efficiency

Hi guys,

The project Im working on at the moment is an RFID reader and ignition system for a motorbike and I'm working to get the current draw as low as possible.

Currently Im using a DC step down module to go from 12-14.5v down to 5v to the 5V pin on my Nano but Im wondering what the most efficient way to configure this is?

Also i was trying to read the mA going into it by connecting my multimeter in series to the positive wire into the dc step down module and i got a read by the device didnt power on. No LED's or anything. Am I doing it wrong?

First part:

  • get an efficient converter (in with that input/output)
  • don't draw much current, for example, remove leds and put micro to sleep (second is useless with leds :wink: )

Second part:
We're not psychic. Please post a clear photo of what you do.

Currently Im using a DC step down module to go from 12-14.5v down to 5v

Switching DC-DC converters tend to be very efficient (you can get more than 90% efficiency* and more current-out than goes-in).

With a linear voltage regulator the current-in is the same as the current-out, and with 7V dropped across the voltage regulator and 5V across the load, more power is wasted in the regulator than is getting to the load.

Also i was trying to read the mA going into it by connecting my multimeter in series to the positive wire into the dc step down module and i got a read by the device didnt power on. No LED's or anything. Am I doing it wrong?

A blown current-fuse in the meter, or you're connected wrong. Does the meter have a separate connection for current measurement? Since an ammeter is essentially a short circuit, blown fuses are common. :wink:

*That's at full load. A switching regulator with no load will typically consume more current than a linear regulator with no load.

DVDdoug:
Switching DC-DC converters tend to be very efficient (you can get more than 90% efficiency* and more current-out than goes-in).

With a linear voltage regulator the current-in is the same as the current-out, and with 7V dropped across the voltage regulator and 5V across the load, more power is wasted in the regulator than is getting to the load.
A blown current-fuse in the meter, or you're connected wrong. Does the meter have a separate connection for current measurement? Since an ammeter is essentially a short circuit, blown fuses are common. :wink:

What I'm using is this

Hmm... It's does show power draw but just no body is home on the ardunio.

sonofeevil:
Hi guys,

The project Im working on at the moment is an RFID reader and ignition system for a motorbike and I'm working to get the current draw as low as possible.

Currently Im using a DC step down module to go from 12-14.5v down to 5v to the 5V pin on my Nano but Im wondering what the most efficient way to configure this is?

Also i was trying to read the mA going into it by connecting my multimeter in series to the positive wire into the dc step down module and i got a read by the device didnt power on. No LED's or anything. Am I doing it wrong?

Use the amps rather than mA setting first (much lower series resistance in the meter), also the mA setting
has an internal fuse in the multimeter, the high current setting doesn't.

MarkT:
Use the amps rather than mA setting first (much lower series resistance in the meter), also the mA setting
has an internal fuse in the multimeter, the high current setting doesn't.

Oh perfect. I'll give that a try

septillion:
First part:

  • get an efficient converter (in with that input/output)
  • don't draw much current, for example, remove leds and put micro to sleep (second is useless with leds :wink: )

Second part:
We're not psychic. Please post a clear photo of what you do.

Desoldering the LE'd just as simple as desoldering or do I still need to complete the circuit?

Alright, some research tells me it is that easy.

Follow up question. Am I able to remove the onboard regulator if I'm supplying the 5V power directly to the 5V pin?

Also will lowering my clock speed to 8 and lowing my voltage to 3.3v save on current draw or will a just lose more at the step down module?

sonofeevil:
Follow up question. Am I able to remove the onboard regulator if I'm supplying the 5V power directly to the 5V pin?

Yes

sonofeevil:
Also will lowering my clock speed to 8 and lowing my voltage to 3.3v save on current draw or will a just lose more at the step down module?

Did you try to look that up in the datasheet?

But like I said, you can save power on the micro (I think sleep will save more than a lower clock...) but if you drive anything like a led or a screen that saving is useless. So please show use a Photo or schematic (like suggested before ;))

septillion:
But like I said, you can save power on the micro (I think sleep will save more than a lower clock...) but if you drive anything like a led or a screen that saving is useless. So please show use a Photo or schematic (like suggested before ;))

Ok, So I got the multimeter to work, turns out the fuse had blown I didn't have a replacement but a little bit of solder and a spare blade fuse holder it's now fine. The whole system was drawing 81mAH at idle.

I desoldered all the power LED's, and lowered the output voltage to 3V, still seems stable surprisingly at stock clock which is confusing me... I would have thought it would crash and freeze... or something. I think minimum voltage is 2.7V but I would have expected some doginess around 3. I'll keep it running and see what happens.

I'm currently trying to find some sort of program or online site to draw up a schematic. But I've got it down to 24mAH.

Mm, blade fuses then to be in a complete different order of magnitude (5A+) than DMM fuses (useally 200mA)...

And about 3V and speed. Atmel just guarantees it's stable @ 12MHz (from the top of my head) @ 3,3V. Which doesn't mean it may not work in this situation. There is only no guarantee whatsoever it will stay that way. It might lock up when there is a voltage dip, or when you try to enable all outputs or when the moon is full... You just don't know. So fine for a test, not smart for a permanent project :wink:

About a schematic, KiCad :slight_smile:

sonofeevil:
I desoldered all the power LED's, and lowered the output voltage to 3V, still seems stable surprisingly at stock clock which is confusing me...

Some people run at 16MHz and 3.3V, seemingly OK, but the datasheet doesn't guarantee it, and it probably
doesn't work across the full device temperature range (cooler allows faster operation typically).

septillion:
Mm, blade fuses then to be in a complete different order of magnitude (5A+) than DMM fuses (useally 200mA)...

And about 3V and speed. Atmel just guarantees it's stable @ 12MHz (from the top of my head) @ 3,3V. Which doesn't mean it may not work in this situation. There is only no guarantee whatsoever it will stay that way. It might lock up when there is a voltage dip, or when you try to enable all outputs or when the moon is full... You just don't know. So fine for a test, not smart for a permanent project :wink:

About a schematic, KiCad :slight_smile:

Here is the schematic. Note that the "power supply" in this schematic is a DC-DC Step down module.

I'm also considering adding in 1 or 2 x 18650 batteries to power it when the bike is switched off but I'm not quite sure how to do this yet... Any suggestion? Ideally it would charge when the bike is on and run from it when the bike is off.

I see RFID, do you put that thing into some sort of sleep? Otherwise all you're efforts will be pointless.

And if you want to permanently use the device, don't go out of spec. And I think 16Mhz but in sleep is just fine for a motorcycle.

And why on earth would you add 18650???!!! You have a 12V battery in that thing. Just use that ::slight_smile: