Jump to content
43oh

uCurrent power controller


Recommended Posts

I guess most of you are the proud owners of Dave's uCurrent [1].

 

I also have one of them and I happily use it. but most of the times I forget to turn it off. needless to say that next day the enclosed CR2032 is flat.

a new revision of that device is soon to be released, however due to physical constraints it will still lack an auto-off feature [2].

 

early this morning I decided to make a small circuit [3] that will be housed in the uCurrent box and that will provide power to it for a given interval. I put together a schematic [4] and a pcb [5] for this.

 

the idea is that the user can leave the uCurrent in the ON position and it will automatically power off either if the output voltage is below a given level for an interval or simply timeout after an interval - this can be reset/tweaked by a button press.

 

my biggest worry is that the analog inputs (P6.0, P6.1 tied to uCurrent's VGND, Vout+) of the msp430f5510 will somehow skew the otherwire precise instrument. Dave's schematic is here [6].

what are your thoughts on this?

 

I am very tempted to simply send the gerbers to manufacture and find out the hard way one month later, but I thought I ask first :rolleyes:

 

[1] http://www.eevblog.com/projects/ucurrent/

[2] http://www.eevblog.com/forum/crowd-funded-projects/current-gold-on-kickstarter/msg354450/#msg354450

[3] https://github.com/rodan/ucurrent_ctrl

[4] https://raw.github.com/rodan/ucurrent_ctrl/master/hardware/schematic.png

[5] https://raw.github.com/rodan/ucurrent_ctrl/master/hardware/board.png

[6] http://www.eevblog.com/files/uCurrentRev5schematic.pdf

Link to post
Share on other sites

yeah, that's an overkill :) I just put 1.3Ah Li-Poly cell in the box and I'm done with it. I drilled a hole in the box and added JST connector for recharging.  For more advanced discharge control, you can just use onboard battery status chip (which is already in uCurrent) and connect it to a MOSFET, thus if battery status is not good it will cut off the power. With this li-po cell you will have up to 8 days to notice that you left it on :)

 

As for skewing the results - it will skew them some, definitely. But how much I can check when I'll get back to my lab (IIRC input impedance for ADC is about 100k?)

 

As for sending the gerbers to manufacture - I think the version on the eevblog site is not the final one yet ;) I backed Dave on the Kickstarter, but with the current delivery time slip with one of the parts @ DigiKey means I'll have my unit probably April/May next year, so I'm thinking of building one myself ;)

Link to post
Share on other sites

yeah, that's an overkill :) I just put 1.3Ah Li-Poly cell in the box and I'm done with it. I drilled a hole in the box and added JST connector for recharging.  For more advanced discharge control, you can just use onboard battery status chip (which is already in uCurrent) and connect it to a MOSFET, thus if battery status is not good it will cut off the power. With this li-po cell you will have up to 8 days to notice that you left it on :)

 

As for skewing the results - it will skew them some, definitely. But how much I can check when I'll get back to my lab (IIRC input impedance for ADC is about 100k?)

 

thanks for your pointers. I did some quick tests by sending the differential output signal given by the uCurrent to some analogic pins on the msp and it hit me that that small ~10mV readouts are pretty much covered by noise. so I added a tiny 16bit ADC (ADS1110) to the schematic to hopefully better detect if the uCurrent is unused. overkill++

oh and the input impedance will be somewhere between 700k-1.2M.

 

I also decided to use a Li-ion 18650 cell (3.7V, 2600mAh) since I have a bunch of them lying arround and they fit great inside the uCurrent enclosure. these cells usually come in laptop battery packs.

 

As for sending the gerbers to manufacture - I think the version on the eevblog site is not the final one yet ;) I backed Dave on the Kickstarter, but with the current delivery time slip with one of the parts @ DigiKey means I'll have my unit probably April/May next year, so I'm thinking of building one myself ;)

 

I thought Dave said he bought the world's stock of precision resistors.

I see a pillow fight on the horizon :biggrin:

Link to post
Share on other sites

O.M.G. 16bit ADC just to see if the output is active? I approve. TBH precision op-amp I had in mind could be more expensive :D Also - what about negative polarity signals?

 

And regarding the precision resistors - After Dave had to order 1000 of them (as a non-stock part), they got back in stock ;) Murphy sends his regards :D

Link to post
Share on other sites

O.M.G. 16bit ADC just to see if the output is active? I approve. TBH precision op-amp I had in mind could be more expensive :D Also - what about negative polarity signals?

 

I am sooo happy I don't do this for a living. otherwise I would have to obey random stuff like cost constraints :rolleyes:

 

uCurrent has a Vout- (VREF in my schematic) that has a constant Vbat/2 level (1.5V in respect to GND) and a Vout+ that varies between 0 and Vbat (max 3V in respect to GND). so a multimeter that measures these two can get negative signals since it's reference is Vout-, however at no point my circuit receives a negative polarity signal in reference to GND.

 

as far as I can tell, the differential input ADS1110 is perfectly capable to handle the voltage swings with a theoretical 0.5V of headroom in both directions.

 

currently my only worry is those bulky switches Dave uses in the new revision, I hope their pins don't interfere with the enclosure's emptiness. I plan to place my board on one side of the enclosure between the wall and the 2 screw posts.

 

later edit:

 

I sent the gerbers to the pcb manufacturer :)

now let's cross those fingers.

 

I will keep this thread updated with latest developments, but in the meantime

 

     Have a Happy New Year !!!1111eleven

post-34829-0-50930000-1389242897_thumb.jpg

Edited by petertux
Link to post
Share on other sites
  • 4 weeks later...

great news everyone!

 

the pcbs arrived a couple of days ago so I went ahead populating one of the boards and writing the firmware [1]. pics attached.

 

unfortunately the mod itself is quite not-user-friendly. I had to use small drills for creating a slit on the side for a tact button and a hole for installing a light pipe to my led. well, actual light pipes would be cool, but I had none. so some polimorph plastic was used instead with poor results. a dremel was used to chop the input posts shorter. but generally speaking all went well and the uCurrent is now powered by a huge accu and it's being shut down after 30min of the last button press.

 

one can connect via UART to tweak the settings and see the ads1110 registers. so the external ADC is functional, but I guess it needs some calibration at the 0V input mark (a relative delta equivalent), I will provide something in the menu for this in the near future (Later edit: rel delta has been implemented and now very small signals (<1mV) are accurately measured)

 

I am not sure how to detect if the uCurrent is not being actively used. 0-ish voltage for an interval? "non-changing" voltage for a given interval?

 

[1] https://github.com/rodan/ucurrent_ctrl

post-34829-0-25267500-1390753319_thumb.png

post-34829-0-99830100-1390753330_thumb.jpg

post-34829-0-85306200-1390753342_thumb.jpg

post-34829-0-51533500-1390753350_thumb.jpg

post-34829-0-65355600-1390753356_thumb.jpg

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...