Battery Management Solutions

Communication between PIC microcontroller - BQ78pl114 and BQ78pl114 Socket for PCB

Brayts — Sat, 21 Nov 2009 14:58:27 GMT

Hello,

In the past few months i've being working with bq78pl114 for monitoring a li-ion battery pack, but I have some doubts before I end my application, those are:

- Can I use a microcontroller like PIC18F2680 to acess the memory of bq78pl114 while is doing polling data? The ideia is replace the computer with de bqwizard software, by a microcontroler properly programmed to read the memory of bq78pl114 (through SBS commands) while is "working"(i.e read parameters from de battery pack)??

- I've done some PCB's to support bq78pl114 chip and the other electronic parts but some times the chip burns (I don't now if is very normal this hapen, but with my expirience I guess this chip is to much sensitive), so I have the need of replace the bq78 in the board some times, but is very difficult due to is package.
So my questions are, if exist any socket for the package of bq78pl114 (QFN RGZ 48 pin) and if is normal this chip getting easily damage??

Best regards,
Brayts

Require softeware to evaluate BQ2010 (EV2010.exe, AP10A.exe)

Deckerg — Fri, 20 Nov 2009 16:41:10 GMT

I'm trying to find software to read and write to the BQ2010 fuel gauge. The datasheet for the EV2010 evaluation board indicates that there are two programs that can be used to interact with the fuel gauge: EV2010.exe and AP10A.exe. These programs don't appear to be available for download from the TI website.

Where might I locate copies of these executables?

How to Configure EEPROM of BQ27000

Mahesh Hooli — Tue, 10 Nov 2009 07:25:31 GMT

We are using 3.7V /950mAh Li Ion Battery. We are using BQ27000 to monitor the Battery status. This is interfaced to OMAP3 through HDQ interface.

What is the default configuration on EEPROM?

How to configure it to our battery?

UCC3911 EVM Availability

Dmitry Vasilenko - Arrow — Tue, 10 Nov 2009 04:41:50 GMT

Colleagues, there is a document ( http://focus.ti.com/lit/ug/sluu088/sluu088.pdf ) that tells that there is an EVM for UCC3911. There is no possibility to order it at the product page, please advise if the EVM is available.

USing BQ24070 charger

moome Liu — Wed, 19 Aug 2009 09:56:17 GMT

I use BQ24070 for 18650 Li-on battery charge, I set charge current 600mA, Iset2=high, Mode=high,

as datasheet mentioned, the Stat1 will go high and Stat2 will go low, but when I test BQ24070 with charge current <60mA,

the Stat2 remain high, not go low . so the silicon result is different from datasheet.

which part is correct? does charger still on after charge complete?

thanks

Battery Packs

Bob Sachsalber — Tue, 17 Nov 2009 21:07:20 GMT

Most of the battery vendors do not supply what electronic design is available inside a battery pack, especially a smart battery pack?

Do you have any insight as to how to obtain such information to determine what is required for Battery Charger design for these OEM battery packs?

Are there any common practices available to interface to these battery packs?

Is it expected that if you are a Smart Battery Pack that you support SMBus or i2C?

Thanks

bq27510 without external thermistor... and other basics

Jakub Pilat — Thu, 12 Nov 2009 17:19:53 GMT

It is my first post here, and i have few pretty basic questions. I am working on project that is going to use bq27510 as fuel gauge. Datasheet gives quite solid info about this IC but it does not clearly explain the configuration without external thermistor (useing internal one instead). I know how to reconfigure it to use internal temperature measurment, but what am i supposed to do with remaining TS pin. I guess that i can leave it unconnected, not sure though. Additionally in recommended design the BI/TOUT pin is pulled up through 1.8M resistor to 2.5V Vcc, and through battery pack thermistor to GND (cell connection pulls BI/TOUT low = battery connected). Because my battery won't have built in thermistor, can I just connect BI/TOUT pin to battery GND (in my opinion it will work the same as with thermistor)?

Ps. Beacause it is a single device project, i am not going to buy EV2300 just to calibrate one bq27510 with bqeasy soft, so could anybody approximately describe its accuracy without additional calibration? Is it possible not to do it?

Anyway, thanks in advance for any answear.

Kuba

TI bq27200 Gas Gauge IC - Fails over time in some environments

Jonathan Murphy — Mon, 16 Nov 2009 04:02:23 GMT

The product (unit) using the bq27200 has been shipping since 2007. Recently we have had a few units fail after weeks or months of use when the gas gauge started reading the battery voltage as zero. Our TI DSP is programmed to interpret this as a low battery and shuts the unit down, effectively disabling the whole instrument.

I have investigated these failures and found that the bq27200 is still working when connected to the TI bq27200 programming and test tool. The bq27200 communicates well, constants are all still stored in memory, and it passes various tests with the exception that the battery voltage is reported as 0.0vdc. Using a DVM, I measured a battery voltage of 4.2Vdc at the pad connecting to pin 6 on the bq27200 (U7 on our power board). I hit the bq27200 with a heat gun and the bq27200 still read zero. However when I pressed my finger on the chip and applied pressure the bq27200 reported 4.5Vdc. When I removed my finger the voltage slowly floated down to 3.2 and then ultimately 0.0Vdc again. This made me think our problem was a cold solder joint (although 4.5Vdc is not really possible on pin 6).

I put the board under a scope and was not able to see any evidence of a cold solder joint. I did note that the thermal pad plating on the PCB (where visible) was not totally covered in solder. I hit pin 6 with a soldering iron and melted the solder. No change, the bq27200 still reported 0.0Vdc. I took the boards to our assembly house and had them use directed heat to reflow all the solder under and around the IC. No change, the bq27200 still reported 0.0Vdc. Currently I am having them remove the bq27200 from the board and replace it with an new IC. They will save the bad one.

I am hoping you can help resolve this issue. I am looking for a failure mechanism that can develop over time given some environmental stress (temperature -20 to +60C, vibration above 10 g's, occasional drops and bumps associated with hand held use, ESD). Perhaps you have seen something like this before

. I will send the schematic if needed.

Jon Murphy

Can bq24005 charge 2 Li-ion cells with an input voltage Vin=5V?

Vamsi Krishna — Sat, 20 Jun 2009 14:18:10 GMT

My circuit uses 2 Li-ion cells and i am designing a charger circuit for them using bq24005. Though the input voltage Vin range given in datasheet is from 4.5V-10V, will i be really able to charge 2 Li-ion cells with 5V input Vin???

problems with i2c access to the bq27501

Ben Wu — Tue, 10 Nov 2009 03:36:05 GMT

I'm trying to get the software written for accessing the bq27501 chip through i2c. The datasheet information regarding the i2c interface isn't very good, however, slua467a.pdf that seems to help.

The problem I'm having is that I'm not able to read the chip firmware version (or anything in the control register for that matter) despite following the app note documentation. All I'm reading is 0xFF00 from anything in the control register. I can read other registers just fine, so I don't know what I'm doing wrong.

Additionally Figure 2 in slua467 describing the command sequence has a discrepancy with the instructions in the paragraph above it. There a line stating "The gauge issues a NO ACKNOWLEDGE" - but the figure shows that it does. Any insight into the matter would be appreciated. Thanks.

Golden Image Firmware (DFI) update on bq2750x

Dani — Tue, 03 Nov 2009 09:57:26 GMT

Hi all,

Please, can you help us with an issue related to upgrading the golden image firmware (DFI) of bq250x?

The status is the following:

1) We have made successful calibration of the homebox PCB with our LiION battery. For this:
        * We have used the "bq Evaluation Software " ver.0.9.58.
        * We have made the necessary configuration, calibration , Impedance track initialization with the appropriate battery charges and discharges.
        * We have generated the golden firmware image file (bq27501ROM.dfi)

We have flashed one box with the upper image and the battery gauge with calibrated impedance track activated was working excellent.

2) We developed a firmware upgrade application (the source code for that application we can provide on request) for our "armv7a" platform that upgrades the bq27501 firmware with the attached golden DFI firmware. The firmware update procedure is according to TI's "Going to Production with the bq2750x (Rev. C)" (http://www.ti.com/litv/pdf/slua449c) document. The log output from this application is also attached to the post.

3) The issue we have is that after flashing the battery gauge (bq27501), we cannot communicate with that device anymore - neither on the gauge I2C address (0xAA), nor on the ROM mode I2C address (0x16). Actually the bq27501 device does not respond on any I2C address on the bus. The "bq Evaluation Software" also cannot communicate with the device.

4) The questions that we have are the following:
    *) How we can recover the original bq27501 firmware? Actually how we can make the device to work again?
    *) What's wrong with our firmware update procedure when all the checksum verifications pass?
    *) Do TI has an sample firmware update code or DFI file, which we could use for testing our firmware update procedure on our hardware?

Any help will be deeply appreciated!

Daniel

Maximum battery capacity of BQ2060A

Tony Wong — Fri, 13 Nov 2009 03:13:27 GMT

We are going to develop a 20Ah Li-Ion battery pack and intend to use this IC for this project but we are confusing on the ability of maximum battery capacity.

From the BQ2060A datasheet , we found that the max. battery capacity is 65Ah as below:

From the TI Selection Guide(page 49), we see the following table which stated that the max. battery capacity is 10Ah:

Could anyone can tell us if this chip can fit to handle 20AH battery?

Thanks!

bq77pl900 cell-V measurement calibration procedure

Brian Myers — Thu, 12 Nov 2009 21:03:18 GMT

I am using the bq77pl900 chip to manage a high-capacity, 8-cell lithium-polymer battery pack in host-control mode. I've managed to get everything working pretty much as I want it to, with the exception of the cell-V measurement calibration. I'm starting to think I must be doing something fundamentally wrong somewhere, because by far my greatest source of variability in my cell voltage measurements, is coming from the chip calibration process itself.

The process of cell-V calibration involves presenting 6 calibration signals on the VOUT pin, and performing calculations using the voltage measurements obtained, to generate two calibration constants which are use to scale and offset the cell-voltage measurements.

On my hardware, VOUT is connected to a 14-bit ADC chip. With a 1.25 reference voltage and 14 bits of resolution, each ADC count represents about 76 microVolts (1250000 mV reference, divided by 16384 counts in 14 bits).

The following are typical ADC counts I see at VOUT, for the 6 calibration measurements :

        ADC counts
Step 1:    0
Step 2:    15708
Step 3:    32
Step 4:    69
Step 5:    12611
Step 6:    9161

The values calculated from this data are as follows :

1. Vdout_0V = 0.00000
2. VREF_m = 1.19965
3. Vdout_VREF_m = 0.00243
   Kdact = -0.00203
4. Vdout_2V5 = 0.00527
   Vref_2V5 = 2.59603
5. Vout_1V2 = 0.96312
6. Vout_2V5 = 0.69962
   Kact = 0.18871
   Vout_0V = 1.18951

The first number at each step is the direct data measurement, in Volts; other numbers (steps 3, 4, and 6) are derived values, as decribed in the datasheet procedure. The final two numbers are the results, used to scale and offset cell-V measurements. Names reflect the names as used in the datasheet.

My problem is that I am finding that a difference of only 1 ADC count (on the 3rd of the 6 calibration values measured) can result in a cell-V measurement difference of over 60 milliVolts ! That's less than 76 microVolts difference on that signal, which certainly seems to me to be "in the noise".

This means that I'll be doing cell-V readings (on an unchanging test signal), and getting numbers within 2 or 3 milliVolts of each other, which is about what I'm aiming for - all good. Then, a calibration occurs, and after this, my cell-V results all jump up or down, by up to 60 milliVolts. The signal is the same, but my calibration constants have been changed enough to completely skew the resulting voltage.

There's a lot more detail I can supply if it will help - code that's doing these calculations; the settling/sampling/sorting/averaging that I'm doing on the ADC results; per-device uV measurement of the 1.25V reference's exact value; etc.

I was previously doing a calibration every 10 minutes (when actively charging or discharging), but have now switched to limiting these calibrations, so that one does not occur during any single charge/discharge cycle. I suspect I should do a re-calibration whenever I detect any significant temperature changes, but quite frankly I have not yet seen any drift which is nearly as large as the calibration uncertainty itself introduces!

Is anyone using this calibration procedure successfully? It's not helping me much (if at all), and I'm tempted to just throw it all away.

Many thanks to everyone who has read this far.

regards,
Brian

bq27510 v1.11 not compatible with bqEVSW 00.09.59 bq27510v1.12

FordFairlane — Thu, 12 Nov 2009 13:22:12 GMT

Hello

I've just downloaded bqEV-EASY-HHSWSetup00.09.59, bq27510v1.12, bqEasy v1.36.00 (Rev. C) to calibrate a bq27510 (device:510, ver:1.11).

When I start the bqEVSW, the only supported target is "bq27510 v1.12 (Dev:510,Ver:1.12) -510_1_12", then a message box says "The detected device is not compatible with this application! Proceed at your own risk."

I think I need an older bqEVSW or a newer target. ¿Where can I find a bqEVSW compatible with my target?, where can I find a bq27510 v1.12 senc file, I've found bq27510 v1.11 Senc File.

bq2710 is not recommended for new desings, I'd like to get some bq27510-G1 samples (they will be available from TI in 10th January), but the only samples I've been able to buy are bq2710 (without G1), any help?

I've found bq27510 v1.12 Senc (Rev. A) for bq27510-G1 (http://focus.ti.com/docs/prod/folders/print/bq27510-g1.html), can v1.12 senc file be programmed in bq27510?, what's the difference between bq27510-G1 and bq27510?, only software?, does a bq27510 become a bq27510-G1 when programmed with v.12 senc file?.

Thank you in advance

About VFC offset in the BQ2060A

bajybukn — Wed, 21 Oct 2009 13:18:51 GMT

Hello, everyone. i have encoutered some problems in bq2060a and i don't know how to deal with it..

1. If the average current is nonzero, Can i send a command to instruct BQ2060A to start VFC Offset. while the current's values is 0. Is it necessary to start VFC Offset while the current and average 's value are all 0???

After starting VFC Offset and have not come to end..IF applying current to the pack., What problems may occure??

2. In mass production. After learning (charge to full .discharge to empty .charge to suitable capaticy), Some pack's eeprom (eg: Pack Capacity and Pack configuration are charged into 65535) is changed..

In sbs interface, the charging voltage and charging current.also can be turned in ohter vlaues, IF I try write correct vlaue to charging voltage, the charging current can be restored to correct one...

How could i to do to void these problems.........

bq77pl900 EEPROM register question

Brian Myers — Thu, 05 Nov 2009 00:54:16 GMT

I have noticed that after writing (or attempting to write) the configuration registers, register 0x0b (EEPROM) retains the value 0xE2 - fair enough. Accordingly, I have put in code after my programming routine that sets this register back to zero.

I have not yet actually used my code to re-program the configuration registers (by not yet toggling the EEPROM pin), so I don't know if this register is changed when the write succeeds (because I'm still testing my code, and only 3 writes are suggested for the life of the chip).

So this post is part question, and part suggestion :

Does the EEPROM register get set back to zero on a successful update? (I hope so, but I'll find out, soon).

I strongly urge anyone writing code for this part to set the EEPROM register back to zero, after success or failure of configuration register updates - otherwise, a glitch on the EEPROM pin could attempt an update, after you think you've failed (I suspect).

It would be a good idea to suggest this in the user manual, if I am correct in how I think this works. If I'm not correct, I'd like to know what the story is.

thanks,

Brian

bqWizard questions in conjunction with bq78pl114

Ada — Fri, 02 Oct 2009 12:35:29 GMT

Hi, i´ve got an issue with reading cellvoltage. In bqWizard cells window are the correct values shown, but with the SBD Read/Write Utility "wrong" values for same cell are shown, see picture.

Is there any conversion factor needed?

(Please visit the site to view this media)

Is there a SMB command for reading safety flags / register?

thanks for any suggestions!

best regards

how to get a fresh .senc in BQ 20z70

bajybukn — Thu, 29 Oct 2009 13:01:22 GMT

In creating a pre-learned pack, there is fresh.senc which is needed... How can i get a fresh senc ???

1.Currently i just write bq20z70_1_50_default_ID0100.SENC to the pack and write the selected chemistry ID in 4B(bqEASY) . Then i export the senc file. Can i think the senc is fresh??

2. If i export .ROM from golden pack., can i write the .ROM to other packs in 1B(see picture ) instead of writing binary files?

Li-ion battery management question (bq20z95)

Neil Dominico — Thu, 04 Dec 2008 13:26:40 GMT

Hi everyone,

I'm a newbie to battery management and would like a sanity check to see if I'm going along the right lines. Having read a few TI app notes, I'm about to purchase the bq20z95 EVM for my project. This is my general plan:

1) Purchase the bq20z95 EVM and the 2-cell battery pack I'm planning to use.

2) Connect the EVM to the terminals of the battery and computer.

3) Using the software, set a few flash settings for number of cells, pack capacity etc.

4) Then do a learning cycle (discharge and charge) which will allow the bq20 module to learn performance data specific to the type of battery I'll be using. I can then copy that flash data across to the other bq20 modules I plan to use, without doing another learning cycle.

Is this generally correct? What discharge and charge currents should I use for the learning cycle? Should I be using a dedicated Li-ion charger at this stage? If so, can you recommend one from TI?

Thanks!

Regards

Neil

Questions on Using the Bq20z45 with Bq247xx for Battery Pack Design

MikePienovi — Mon, 26 Oct 2009 21:11:00 GMT

My customer is planning to use the BQ20z45 and a BQ24747 in a battery pack design and they had the following questions. Can someone help to provide answers?

1.) Can a host controller, through the SMbus, have the AFE disconnect the load from battery and on reconnecting return to normal operation?

2.) When using Bq20z45 with Bq247xx, what is the best approach to isolate the positive charge input be protected & isolated from reverse current (if it is not easy, then I will add an isolation diode)?

3.) What external components can be optimized / eliminated when using Bq20z45 with Bq247xx? Could a single temperature sensing element be used? Could some of the pass switch elements be eliminated? Can they share current sense information (to use one Isense instead of two)? Is there any reason to place a switch (AFE controlled, as in the eval board) between the charger and battery?

4.) Since our intended power supply (for charging) will always be significantly higher voltage than the battery, could we safely replace a mosfet in the charger with a catch diode?

5.) Is there a a.) method, b.) advantage, to sharing information about the charge state that the AFE/Fuel gauge is aware of, to the charger? Could we get into a situation where the fuel gauge sees that more charge could be accepted at the battery, but the charger has “given up” i.e. completed it’s cycle, or gone into float/trickle mode?

senc file

Mike_chute — Mon, 02 Nov 2009 22:18:41 GMT

Reprogram the pack with a fresh firmware (*.senc) file with selected chemistry ID, to clear all hidden

constants and set the correct chemistry specific table.

How do I verify the correct chemistry specific table is been loaded into the 20z95? I can't find it in the data flash.

mike

Programming code fround on SLUA355A, page 5.

Mike_chute — Mon, 02 Nov 2009 17:48:45 GMT

This is code to be complied in VB. Where do I find the SMB library function calls? like

WriteSMBusCommand(&H8)?

Mike

Current selection for PowerPump balancing circuits

PK — Sat, 31 Oct 2009 19:39:08 GMT

I've read through the PowerPump Balancing app note and the bq78PL114 datasheet, but I haven't seen any information regarding the selection of an appropriate balancing current for the charge pump circuits. Is there a rule of thumb to follow, like a certain % of the total battery capacity?

parallel cell balancing using bq76pl102

Srivatsa Raghunath — Fri, 30 Oct 2009 11:37:41 GMT

In page 7 of bq76pl102 datasheet it says it can be used for balancing one or two Li-Ion cells in series, and any number in parallel (limited by other design considerations)

Pls let me know how to do cell balacing in a 2S2P system using bq76pl102?

Extended SBdata commands in bq78pl114

Castano — Wed, 28 Oct 2009 18:16:55 GMT

Where are they available?