Dangerous Prototypes

One of the things I have noticed missing, for a pretty long time, is a good *safe* lithium cell charger that also supplies a load with a good efficient regulator.  My answer, I hope is the BLiVIT.  This is a moderate-sized board that incorporates charging from either USB or a wall wart, efficiently manages a single Li+ or LiFePO4 cell, has a power-on/off soft switch (Just like my Powah-1), a fully isolated load supply (no need to blow up my scope when the wall wart is connected), and an efficient boost/buck regulator that supplies either 3.3V or 5V.  I am going to try to see if I can get a photo of a prototype board and/or Sketchup photo in this post.  I also wanted to offer the schematic, as all my designs are open-source, including the "big" ones, but it would be a full page drawing and I don't think that would work here.

I am currently awaiting the rev. L pcb to arrive sometime today, but when that happens I will get parts on it soonest and post a photo.  One of the things I have been needing to do is decide whether to custom fab the enclosure for this.  For the M rev I am going to mount it in a Sick-of-Beige box.  The board will get larger, but more parts can relocate back to the topside, and a larger Li+ cell can be housed.  The placement of the next project daughter board will be decided then.

Here is a top view of the rev K board:
(http://http://www.flickr.com/photos/61021965@N06/8575632198/)
P01-109K_BLiVIT_Top (http://http://www.flickr.com/photos/61021965@N06/8575632198/) by MisterWidget (http://http://www.flickr.com/people/61021965@N06/), on Flickr

Here is a bottom view of the rev K board:
(http://http://www.flickr.com/photos/61021965@N06/8575632298/)
P01-109K_BLiVIT_Bottom (http://http://www.flickr.com/photos/61021965@N06/8575632298/) by MisterWidget (http://http://www.flickr.com/people/61021965@N06/), on Flickr

you are posting link to the flickr photopage, not to the actual photo.

I'd like to know more.  I agree that there is a need for a really elegant, reusable design for a Lipo charger and buck/boost with low voltage lockout that will deliver 3.3v from a battery from 4.2v to 3.0v

O.K. then.  Here is the start of a running dialogue on the whys and wherefores of a charger/regulator (The BLiVIT).

Elsewhere, in the OSHW blogosphere and marketplace, I have been seeing not a few LIFePO4 chargers.  Most of the designs out there have been intended for standalone charging function only.  Mostly these were for use on cell form-factors identical to AA and C type alkaline batteries.

The deficiencies of this design are numerous, and I will enumerate some of them here.

The AA cell chemistry does not allow for a great deal of charge density.  There are also a lot of thermal problems, mostly having to do with the shape of the cells.  Another problem is that the typical charger device has about 4 cell slots, and the charging circuitry simply is not designed properly to handle multiple-cell geometries.

That is why I have elected to support a single cell in the BLiVIT project.  Many characteristics and compositions of cell chemistry are now available.  I have chosen a cell with 950mAH charge capacity, while taking up only about 1.5 x 2 x .25 inch volume.

The next major deficiency of typical charger systems is that they rely on physically isolating the cell from the power load during the charging cycle.  Aside from design life issues with mechanical connectors there _are_ safety issues with handling a fully charged lithium cell.  That is why for the BLiVIT I required the capability of supporting the power load during the charging cycle while also striving for preserving cell working life.  There are numerous articles and books published which warn very strongly that ineffective charging cycle management will have severe effect on the working life of the cell.

That's all for now.  I will continue with more later.
J.R.

Continuing the story of the BLiVIT project.

The necessity of overcoming some of these limitations of previous charger designs required much research.  Not being a power management engineer, I was not likely to use jellybean parts alone.  So I opted to just start at the "front" and work my way back.
I exhaustively worked over many datasheets for power management devices and found many that "almost" fit my needs.  Finally I came across the Maxim MAX1874.  The first criterion for the device was to have the outside circuitry be as simple as possible.  The second being that at least one of a DC adapter or USB power be permissible for sourcing the charger.  Happily, the MAX1874 can be used with both sources, the DC power taking precedence when connected.

My next requirement was to be able to drive the load or charge the cell with automatic switchover.  It was nice to find out that the MAX1874 could do that as well, as well as drive the load circuit, also automatically.

Thermal issues with the cell are important.  Lithium ion cells generate a lot of heat, which can be slow to dissipate depending on the geometry of the cell.  The MAX1874 has a thermistor sensor feature that I have chosen to employ with an NTC thermistor that has resistance 10K at 25 C.  The combination of constant current sensing and temperature sensing means the charging cycles can be constantly monitored for best efficiency.

That accounts for most of the front-end of the BLiVIT.  Next we must look at downstream requirements.

Even if the power load is miniscule it is not zero, so a soft-toggle switch using two schmitt inverters and a PFET is put downstream of the cell.  I wanted the charger to be functioning and keeping the cell trickle-charged even if the load is disconnected or turned off.
The switch is activated with the TACT switch on-board, or any momentary NO switch connected to a header.

One of my particular issues is that the power be fully isolated from any connected source (the USB or DC input).  It would be a very bad day if I was powering a test circuit and a ground loop happened and blew out my oscilloscope.  I am using an isolation transformer and driver chip to protect the load from the upstream.

The voltage range of a lithium cell can range from 3.1 to 4.2V, which are not altogether useful for digital logic, so I use a switching boost/buck regulator to drive any of the input voltage to a locked output voltage.  I am using an adjustable regulator so that a jumper can be used to select 3.3V or 5V output, which is available at a header or the terminal block.

Maxim has developed a reputation for announcing many interesting chips but backing them up with very limited (or nonexistent) production quantities.  So I, and many of my acquaintances, are wary of using Maxim parts even in hobby projects.  My own rule of thumb is that if Digikey and/or Mouser don't have substantial quantities of a Maxim part *in stock*, I look for an alternative.  FWIW, both Mouser and Digikey only have a handful of MAX1874.

Would you mind to share your schematic?

I'm especially interested how you are doing the isolation. What parts are you using for that? On your board rendering I can't see any big transformer or self-wound coils.

Great idea, I too think there's a lack of solutions that manage charge AND supply at the same time. It's great to see someone thinking about this - I haven't the expertise to do so, so I like that you're articulating decisions too, can help me think about my scenario.  I also want to use a solar panel (which I have been looking at LT3652 power tracking / charge chip for).

In response to a previous post I have enclosed the partial schematic of the BLiVIT design.  This is the soft-power switch and power isolation sections of the BLiVIT.  The isolation transformer is the larger black SMD component (looks a bit like an SOIC-6 on steroids :)
(http://http://www.flickr.com/photos/61021965@N06/8587526655/)

For funsies I include here the next revision BLiVIT board, which is now a 37x60mm form factor.  This will fit nicely in the Sick Of Beige box I have already ordered.  Much magic occurs at the little 2-wire connector labeled Qi.  Much about that later. (http://http://www.flickr.com/photos/61021965@N06/8589415957/)

While noodling a bit on the schematic posted previously I had a bit of an epiphany.  Why in blazes was I doing that with the power on/off stage?  It is an exact duplicate of something I have already designed:  The Powah-1.  That is a little .5x.5" module that was orignally intended to turn on and off breadboard circuits (it can be used as a motion control limit sensor or anti-tamper switch, too).

Here are a couple of photos:

(http://http://www.flickr.com/photos/61021965@N06/8572449515/)

(http://http://www.flickr.com/photos/61021965@N06/8572449627/)

Here is the schematic:

(http://http://www.flickr.com/photos/61021965@N06/8573542056/)

My thinking here is that if I put four pads on .1" centers on the BLiVIT board for the Powah-1, the pins on the Powah-1 will extend sufficiently to place the tact switch flush with the top surface of the S.O.B. enclosure that was already going to be used.  Neat, eh?

That adds a line-item for the BOM cost of a Powah-1, but also reduces the BOM cost for the BLiVIT, so it's a bit of a wash.
Here is what the BLiVIT schematic is looking like right now:

(http://http://www.flickr.com/photos/61021965@N06/8594392413/)

Well, yippee!  The BLiVIT, POQiTX, and POQiRX combo has been accepted for inclusion in the July/Aug issue of Elektor.
What that means in practice, is unknown.  Maybe I better get my hamster wheels spinning just a little bit harder.

[quote author="asgard"]Well, yippee!  The BLiVIT, POQiTX, and POQiRX combo has been accepted for inclusion in the July/Aug issue of Elektor.
What that means in practice, is unknown.  Maybe I better get my hamster wheels spinning just a little bit harder.[/quote]

Just wondering what did you do to get included? just mailed them?

No.  I had to prepare photos, schematic, parts list, and a fairly detailed description of the project.  I also included the CAD files.
I still am a little unclear on the legalities and liabilities.  What rights to commercialization of the design does the magazine reader possess, and what rights do I retain as the original designer?  That's a conundrum.

I guess if they're "open" projects then you don't really lose anything by publishing, since that would be a goal anyway? But yeah I'd probably say it's worth checking out what ramifications (the bad stuff, for you) there are associated with publishing something in a well-respected location, in fact I'd be surprised if they didn't have a guidance pack for contributors.

More different:

I have been playing panellizing dominoes with the components.  With a bit of fiddling around I can get 5 complete kits of boards on a single 7.8 x 10 panel.  That is by only using Eagle and Gerbmerge.  All well and good, but Gerbmerge doesn't automagicallty do anything with breakoff tabs, so I probably have to do it with v-scoring.  I do not have the capability to graphically edit Gerber files, so that process of eliminating unnecessary score liknes is probably going to be funky and slow.  In addition, there is a nifty ULP for Eagle that prepares a centroid placement file in .csv form.  One problem is that there needs to be some way of expanding the centroids to accommodate multiple placements on a panel, in varying rotations.  I have developed a manual process using the placement list from Gerbmerge, combined with the centroid ULP output, within a spreadsheet, to do that, but it takes multiple days just to do the 20-board panel of the BLiVIT Project.  Just imagine the brain pain that would result from doing a 273-board panel for just the .5x.5 Powah's  That sounds like a job for a computer, eh?  That results in just a list of component ID's, X and Y offsets, and rotation angles for the pick-and-place machine.  I have started to look around at various assembly house web sites for technical requirements, but they do not seem to be consistent whether they can take individual board centroids and BOM's and do the placement themselves, or if I have to do the placement myself, with an expanded BOM to take in the entire panel.

On another note, regarding my previous post:  The reason I asked in that way is that I have adapted a project from last year's Elektor project issue into a much better packaging and layout, while keeping the previous circuit.  I wanted to build a few of these to give away as Christmas and birthday gifts this year, but it seems like a pretty nifty thing that other people can use.  My boggle is that would it be legal for me to commercialize the design, even if I got the original circuit from a magazine article, but everything else was designed by me?  Probably can't get a definitive answer except from the publisher (which is dicey), but I was curious if anybody else (DP, Seeed, Sparkfun, that crazy bloke from "not Austria"?) have been through that particular mill.

I wonder if this blog should relocate to the "project log" area, since it seems more appropriate for that, now.