Switch power of GPS module

Go To Last Post
23 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello,

 

I am working on a GPS test board. It is running on a battery and in order to save energy I want to switch off the GPS module controlled by a 328.

Following scenario:

 

- Single power source on board

- Using a GPS module which required 2 power supplies

   - One for Vcc (normal operation)

   - One for backup power (V-backup)

 

Now I was thinking to use a transistor to switch the Vcc ON/OFF. Normally in such scenario the load (GPS module) is on the collector side.

The challenge here is that the V_backup uses the same power source (common round) and for that the ground needs to be connected :-( (meaning the Emmitter ground is useless).

See attached sketch

 

Any suggestions on how to get this up and running?

 

Thanks

 

 

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

What is the backup VCC for?
How if the main VCC get cut off while the backup stays on? Would it reduce the power cunsumption?
.
MG

I don't know why I'm still doing this hobby

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The backup is used to store GPS related information (consuming 7 uA) / that is all it does.

The normal Vcc is used to receive satellite information and determines location (20mA and peaking 100mA)

 

Without backup connected the GPS modules will have to do a cold start each time (takes up to a couple of minutes)

When we have the backup information stored a warm start only takes a couple of seconds (<10 secs)

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Provide a link or part number for the GPS unit please. Your description does not make sense to me.

 

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sure - the GPS module I am talking about is the Quectel L80

 

http://www.quectel.com/UploadIma...

 

Last Edited: Fri. Aug 4, 2017 - 09:13 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You've not told us what voltage the avr is.
Use a pnp transistor or p channel mosfet to switch the 3V to the gps module. Do you want to switch the backup voltage?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I feel really dumb now .. yes of course I should you the pnp (agh). The backup battery is fine / see no reason why I would like to switch that one.

Using 328p at 3.6(V)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Use a P-channel MOS-fet

Bipolar transistors are so '90's.

 

A SOT-23 fet like the IRLML6402 can handle multiple amps continuous ( and a monstrous 22A peak).

With an Rds(on) of 65mOhm they also only drop 65mV @1A, which is only 10% of the voltage loss of a NPN transistor.

This means you can drop the overall voltage with 500mV which results in instant power saving in your circuit.

 

Also make sure your GSM module has big caps on it. They have a high peak current demand.

 

As an alternative you can use a separate smps chip to generate Vcc for your GSM module.

A lot of these chips have a logic level input to turn the regulator on or off.

Paul van der Hoeven.
Bunch of old projects with AVR's:
http://www.hoevendesign.com

Last Edited: Fri. Aug 4, 2017 - 03:20 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Paul mentioned one device.

I've used the DMC2038LVT which is a small, 6-Pin, combination N-Fet and P-Fet device, rated at 2 Amps.

Diagram below:

 

JC

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You do need to think about how much off time you'll have & whether it will save energy.  Part of this is determined by whatever caps are in the unit begin switched (here GPS)...if there is much capacitance, when you turn it on, there will be a huge surge while the cap "tanks" fill up.  Then during the off time, the module is not really off, it is running from the filled "tanks".  Then you turn it back on & the tanks refill in big gulps--so the energy savings could be practically nil.  Typically the modules are made as cheaply as possible so they tend to not have much storage capacity--so as long as you don't add too much capacitance on the switched side, you should be ok (or ensure you use long enough off times). 

When in the dark remember-the future looks brighter than ever.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If the GPS module can keep on running while the MOSfet is switched of a short time it would be futile to shut it off anyway.

Are we talking us, ms or s here?

But when you go into minutes, hours, days it is definitely worth turning the GPS off.

 

Important:

If the GPS is off, then make sure there is no current running through any connected data pins.

(Switch them to ouput low, or input (high impedance).

 

There are additional dangers, which need to be adressed.

It is very possible that the "tank" caps on the GPS board draw so much current during initial charging that your Vcc collapses when you turn the module on. ( uC Brownout etc).

Even though the MOSfet's mentioned can handle peak currents of -22A and -17A, the Rds(on) is so low that even this might not be enough:  3 [V] / 0.065 [Ohm] =  46.154 A

This is in the same realm as when you see sparks ( = molten metal) fly  when you dead-short a charged capacitor. This damages the leads and may even weld them together.

One of the ways to handle this is to put a relatively big RC combination on the Gate of the MOSfet, so it switches very slowly.

I'm thinking about 10ms to 100ms switch time (This is NOT the RC time constant !)

The actual time needed depends on the total capacitance on the GPS board and the current your power supply can deliver safely.

Paul van der Hoeven.
Bunch of old projects with AVR's:
http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I don't think I'd be expecting large capacitance on the input to the GNSS module. Note that it is not a phone module.

According to the specs, it draws around 1mA in sleep but an active antenna may keep on drawing 11mA. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Too much text...

OP provided link to the datasheet (#5).

Module seems to small for big caps and  the power supply section advises to add 2 100nF and a 10uF capacitor.

 

OP has also said that the modules needs several minutes to initialise after complete power loss ( #3).

Actually I don't understand why OP made his own schematic instead of copy/paste from the datasheet:

 

Paul van der Hoeven.
Bunch of old projects with AVR's:
http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi,  Paulvdh

Is it possible to move the capacitors to the input switch side?

 Like this pic.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I wanted to keep my posts short (I tend to elaborate too much) and I almost added that I would very likely move the 10uF in front of the switch and leave the 100nF cap at the L80's side.

The 10uF cap is probably already somewhere on the powersupply anyway and adding a 2nd 10uF cap after the switch would introduce the problem from #11.

Whether also moving the 100nF cap and adding a 10nF cap keeps the needed decoupling perfomance depends on the quality of the switch and pcb layout.

The P-MOSfets mentioned are much better switches than Bipolar transistors.

Why do you want to know Goorman, are you also using this module?

 

Paul van der Hoeven.
Bunch of old projects with AVR's:
http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The bottom line is--if the module's or other caps have enough storage to keep the module running during the entire off time, then the switch makes no difference.  This scenario is based on cap size, off time, power consumed, and minimum operating voltage.

 

In fact, in such a scenario the switching might cause MORE power to be used, since the recharge must occur at a higher current.  For example, say it was on 50% & off 50%, but the module was always running normally due to caps.  This means during the on time, the current must be at least double what it would be without any 50%/50% switching.  Doubling the current causes 4x any resistive losses, compared to just simply supplying the power steadily.  In fact, if the module turns off for a only very small fraction of the off time, the increase in surge losses at turn on could negate any savings.   The sure way to save energy is to ensure that the off time is long enough so the module is off during most of it.

 

Of course, this becomes somewhat more important when talking about a 10 amp load rather than a 10mA load.   However, for maximizing battery life, any excess loss should be investigated.

When in the dark remember-the future looks brighter than ever.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Paulvdh: "Why do you want to know Goorman, are you also using this module?"

 

I didn’t use this module, and I look at it in the form of a black box. I just want to know if the capacitors can be used before the power switch.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 

- For me it makes sense to switch off the GPS module // as it will stay off for 30 minutes or even a couple of hours.

- I am going to experiment using the P-MOSfet  to switch the power (with capacitors)

 

When I find time this weekend and get things working I will post the results here ..... 

 

Thanks

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

In the end it boils down to impedance. If you're adding the switch to the power loop then it has it's own resistance, inductance etc so the power coming from that 10uF (or any other capacitor) has quite a long path with voltage drops. Looking what the circuit after the power switch consumes in current (in this case a GPS) then all of the capacitors run flat in pretty much no time at all (nobody is running the GPS off of a PWM signal I hope :D). In those cases where power is switched on-off the OFF time is usually substantially longer than the ON time and current consumption big enough to matter. 

 

TLDR: putting the decoupling capacitor after a switch is like putting it on the other end of the PCB. Pointless (10uF is not considered "bulk capacitance" in most cases)

Last Edited: Tue. Aug 8, 2017 - 09:03 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

What are your proposed on & off times--that's really the crux of the consideration (off for 20 ms, 2 seconds, 2 minutes, 2 hours???)

If your caps can keep the unit running for 10 seconds & you switch it off for only 11,  your savings is nil.

When in the dark remember-the future looks brighter than ever.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

bloody-orc wrote:
TLDR: putting the decoupling capacitor after a switch is like putting it on the other end of the PCB. Pointless (10uF is not considered "bulk capacitance" in most cases)

Those SOT23 fet's are pretty small & with an Rds(on) of 46mOhm ...

How about drawing the schematic with 2 10uF caps (one on both sides of the switch) in the schematic & pcb and doing some performance measurements / SPICE simulation on voltage drops & current peaks before settling on a final solution.

Paul van der Hoeven.
Bunch of old projects with AVR's:
http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The GPS module definitely needs its own 0.1 uF by-pass cap mounted after the PFet switch, as close to the GPS module as possible.

Also, read the GPS module's data sheet to see what is says about decoupling the module.

 

As mentioned above, where to but the 10 uF cap is a design decision, with trade offs regardless of where you put it, before or after the PFet switch.

As mentioned, the duration of the OFF time is a factor in making this decision.

 

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

as it will stay off for 30 minutes or even a couple of hours

 

That is a nice long time, so  most caps (1uf, 22uf, etc) will likely discharge long before then & you will save energy.

When in the dark remember-the future looks brighter than ever.