LTC4006-6 Charger woes Any help appreciated!!

[bentomo]

Hello Guys, been a lurker and a long time fan of EEVBLOG.

I've been designing a circuit board that is a power supply and battery charger combo board. It's supposed to charge a 2 cell 7.4v lithium ion pack, and supply 1.8v, 3.3v. and 5v. The regulators are working fine and I can power whatever I want with them.

The problem is that whenever I connect a battery the output voltage drops to slightly higher than the battery and trickle charges at 20-30mA. It should be conditioning dead batteries at 300mA, and charging at 3A. I've tried different packs and power supplies. I've even ordered a stencil and reflowed the board to make sure my soldering was not to blame. I've checked all my connections with a meter and everything looks good. I've gotten it to work once but it was still only charging at about 1.5amps and I'm not even sure how I did it. That's when I ordered a stencil and assumed it was the soldering, turns out it wasn't.

There was actually a thread here on this website that was posted a few years ago about this. https://www.eevblog.com/forum/projects/ltc4006-stange-mosfet-drive-signal-behavior-am-i-missing-something-basic/

However he seems to have fixed the issue by adding a totem driver, which is not recommended anywhere in the data sheet as the mosfet gate pins are already suitable gate drives, so I'm not looking to add those. I also have a demo board that was sent to me after I contacted Linear Tech, and they have no kind of extra components for any special filtering or gate driving, it's supposed to be a stand alone charger.



Here's the data sheet:
http://cds.linear.com/docs/en/datasheet/4006fa.pdf

Here's my schematic:


Here's my BOM:
http://www.digikey.com/short/t700b1

Here's the top and bottom gate outputs of the demo board:
[imghttp://i.imgur.com/MfVjiWL.jpg]http://[/img]

Here's the top and bottom gate outputs of my board:


The major difference is that when I connect a battery to my charger the square waves become very dim and the normal 8.4v output drops to the battery voltage.

Here's the front of the board:


Here's the bottom:


You can essentially ignore the right hand side of the board beond the sw1 and sw2 through holes. The small circuit with the potentiometer is a poor mans battery sensor for the charge capacity.

I had the board fabbed at pcbway:

(Big Pic)

Assembled:

(Big Pic)

The air wire and the short across RCL were just a few of the many many things I've tried to get his to work. It doesn't look like it's an issue with the sense resistors as my traces are shorter than the demo board and adding another in parallel doesn't seem to increase the charge current. As I said before I've tried multiple IC's, packs, power sources, and more and they all seem to exhibit the same behavior so it must something wrong with my circuit, which is less likely as it's nearly identical to the demo board, or there's something wrong the the parts that I picked or the design of the circuit board.

I'm so close to just throwing this design out the windows and cutting my losses, and switching over to a max1873R. But I'd really like to avoid that as I've got so much time and money invested into this board.

You guys would be just amazing if you can help me fix my issue.

Thanks!

[DutchGert]

U swapped Tgate and Bgate?

Also a more general comment: your schematic is not very readable, try and use more space next time 

[AlfBaz]

The problem is that whenever I connect a battery the output voltage drops to slightly higher than the battery and trickle charges at 20-30mA. It should be conditioning dead batteries at 300mA, and charging at 3A.

From the "operation" section of the data sheet
...
At the beginning of the charge cycle, if the cell
voltage is below 2.5V, the charger will trickle charge the
battery with 10% of the maximum programmed current.
If the cell voltage stays below 2.5V for 25% of the total
charge time, the charge sequence will be terminated immediately
and the CHG pin will be set to a high impedance
...

and just prior to the above quoted section, are the conditions required on the various other pins for it to go into charging mode. Have all those pin states been met?

[bentomo]

Well leave it to the day that I post this, after weeks of debugging, that I figure out the problem.

I disabled the current limiting by removing the feedback r8 and c7 and shorted the cln and clp pins. The problem before when I did this is that I left the resistor and cap on which caused problems.

If you look at the trace going from RCL to the ic, the total trace length is about 1.5in and it's wrapping around and under the inductor, this was obviously a very poor design.

After shorting the pins the batteries now charge at a solid  amps, no more bouncing around and a very clean voltage signal.

Bad news is I'll have to do a new board run and I've got an almost useless stencil. Good news is that I can use all of the same components.

And thanks for the tips about the schematic, I'll be sure to be more organized next time.