Does bq20z65 works?

Hello,

We want to help in the problem solving for this situation. The best way we can move towards a solution is communicate with each other and approach the situation looking for a solution.

Can you please revise post system details to help us understand what you are trying to accomplish such as pack configuration information and system configuration?

Also the bq20z65-R1 and the bq20z65EVM-001 is a gauging part which belongs here in the gauging forum:

http://e2e.ti.com/support/power_management/battery_management/f/180.aspx

The difficulty that you have been experiencing is unfortunate so please communicate with us the technical details of the situation to help resolve what is going on in your design.

This thread is now moved to the Gas Gauge Forum. No re-post of this is needed.

The battery pack I'm using consist of 4 (four) 2.2Ah Samsung 18650 connected in series, connected to terminal block located on the left side of PCB.  I devised a load bank, consists of 5 (five) 18.5 ohms 50 watts, and 1 (one) 3.8 ohms 150 watts power resistors, all connected in parallel, allows me to configure to draw anywhere from <1 Amps to 7 Amps when powered from 4S lithium battery.  I have many battery packs in this configuration, and they all work fine, as I can watch current draw, and discharge behavior, with or without battery protection module I built using Ricoh 5432V chip.  I built the charger using TI chip, and works as expected as outlined in the datasheet on same battery packs, with battery protection module.

When this EVM is connected to Battery and load to draw <1 Amps, it does not turn on, no output current.  Something is latched inside this bqxxEVM, cutting off power to the load.  This is where hardware problem ends.

I also have 2400EVM module that allow this problem EVM to be connected to PC, so that I can monitor/set parameters.  As soon as I connect SMB bus it complains that I need to be connected to I2E bus, either way, there is no communication between bq20z65evm to 2400EVM.

I hope I was clear explaining problems.

Hi,

Have you always had communication problems with EVM, did you ever manage to communicate with the EVM to configure properly before trying to connect your load? 

Did you pull the /PRES pin low?   NR  is set to 0 by default in the Operation Cfg B register so the pack is expecting to see /PRES be pulled low to indicate it is connected to a system before it will turn on FETS. But it sounds like the pack was suppying current at some point if you have previously somehow managed to pass damaging current at some point which is indicated in your first post, so you must have /PRES pulled low?

The next place to look is the permanent fail status ( PFstatus) register. If you have any PF flags set that will disable the FETs. You will need to send the PF clear code.

Mike

It's the first time I'm using TI EVM.

~pres pin (center pin of terminal block located on top of PCB) is connected to the ground plane of the PCB, using jumper wire.

As I've previously stated earlier, there is no communication between this EVM and the PC, therefore, there is no way I can do to reset/set any of the settings.

How is it that units are delivered in non-working condition, two consecutive times, or how is it possible to damage the EVM?  Chances are your exposed solder terminal is shorting to ground somehow, even with plenty of isolation tapes applied.  You really need to pull one unit from inventory, and see if it works,   You may have whole inventory of parts, that is defective.

okay so lets check through to find problem

I assume you loaded the EVM software from here http://www.ti.com/tool/bq20z65-r1_bqevsw-sw

What operating system are you using on your PC and did you get any error messages during that install?

If you are using windows 7 you should right click the  install icon and run as administrator.

Are you using the EV2300 or the EV2400. You have to install appropriate software those as. I use EV2300 if you have one ?

Before you connect to any batteries and while we debug and configure I would suggest you connect resistors across the cell inputs and use a bench pack to apply power across top and bottom of resistor ladder. It is easier to test settings that way as you can turn supply up or down to test thresholds etc.

After you first connect power you must momentarily apply power between Pack+ and the top cell connection to wake up the board. 

Now run the EVM software as administrator . What do you see on PC screen ?

I used software as you noted above, version 0.9.64.  During installation process, there was no error as I recall, and there was no error message, but now when I try load this program, the error message reads "the EV2300 connection could not be established!  Please close this program and try again."  Close this program, and reload, same message appears.  Common sense dictates that program will halt at this point, next program execution will not happen until software validates changing in environment, that user plug in EV2300 module.  So, there is no sense of trying to troubleshoot this issue.

I use Windows 8, and according to your own feedback some weeks ago, EV2300 cannot be worked with Windows 8, consequently forced me to upgraded to EV2400.  My question to you is who architect such a genius initiation procedures, that you limit this tools to be used only with specific tool.  After all, all we're doing is relating (transmitting) message from one test box to a generic PC, using standard protocol.  This one really turns me off, and some people call themselves a programmers....

Ignore this error message, and connected to SMB bus port provided on 2400, then guess what the error message shows?  It tells me to connected to I2C bus.  I had no ideas why it's calling for I2C bus when this box utilizes SMB bus.  So, I guess whoever wrote this s/w knows things that is beyond common sense, so re-routed provided cable  to I2C bus, and do you think it will work?  As I stated in original post, this s/w is a joke by itself.

Here are some additional resources you may not have seen and want to try

EV2300 64 bit drivers http://e2e.ti.com/support/power_management/battery_management/f/180/t/128220.aspx

EV2400 support components http://www.ti.com/lit/zip/sluc291

The bq20z65 EVM has been available for many years and we shipped many. I  recently had to order one oneself which was fine on my Windows 7 install with EV2300 so from my perspective it seems unlikely the board itself is faulty or that you received two fualty ones. Do you have access to a windows 7 or older machine you could try and check the boards operation with ?

TI already stated EV2300 is not workable on Windows 8.  Not only it's not compatible with OS, it caused a fire.  As a result of accident, I was forced to upgrade to EV2400.  By upgrading to EV2400, somehow, I was lead to believe that all my problems are no more, but I'm still nowhere after one month of wait.

It appears that bq20z65EVM is not compatible with EV2400, and you can't neither confirm/deny after wasting another day.  You can't perform a simple test, but you resort to your own sales history, which is completely irrelevant to issues I'm struggling since I received both unit last Friday.

Going back to old OS is out of the question, since that is not the issues I'm having to deal with this time.  Neither, my writing isn't clear enough, or you do not read previous post carefully,  I'm not interested in bashing TI, but desperately seeking solution to remedy, and you're no help.  It's like whole month is wasted such way; you simply do NOT listen to your customers, and insist on something completely irrelevant.

More than two decades ago, I worked on a system using ARINC 629 bus, and as an hardware engineer, no one in our side were familiar with newly developed spec.  So, we asked our vendor, to come up with a simulation software, such as your EVM to see what's in it, and to confirm/validate FAA requirement using this newly developed bus architecture.  I really like to see your touted Battery Gauge with Impedance Track technology, and see what I can do to with it to enhance my own unit.  It seems that it's now becoming failed project because of many flaws we discussed.  I wasted one month of time, along with other damages incurred by an accident, and haven't accomplish anything at all.  We only have so much patience, and I'm really disappointed at how TI support its customers. 

I am sorry you are having these difficulties and am prepared to try and help you work through them.

To operate the gauge has to first be correctly configured for number of cells, cell capacity, paramters that affect gauging , paramters that set IVT protection thresholds etc etc. Then the pack and gauge have to run learning cycles so pack capacity and resistance can be learned. To do any of this you have to communicate with the device and it sounds like this has never been achieved.

I am therefore struggling to understand how an unconfigured gauge got placed into a situation where it could be blamed for a fire? 

I cannot establish communication between bq20z65emv because I don't have the pc peripherals; bq20z65evm refuses to make connection to 2400EV as noted above.

How can I do all that wonderful thing, i.e. setting parameters, if I do not have the communication between this EVM and the PC?

I am trying to see if the software team has anything new that might help. In the meantime please at least check the two software links I posted above which might help,

the windows 64 bit drivers might help get the EV2300 to work with Win 8

the EV2400 install pack is needed for EV2400 you should certainly load these as a minimum if you are using EV2400

at least telling me you tried these things allows me to communicate what has/hasn't been tried to my sw team.

Window 64 driver had its own warning during installation time.  It showed something like, "use it at your own risk..."

Now, I learned to heed these warnings, because you never want to see lithium battery caught in a fire, 5Ah 4S battery can do a l lot of damage.  Its temperature is so high, it melted the metal desk, and there is nothing to extinguish them right away, but let it burn all the way through.  From my past experience, I know that fire extinguisher will make it worse to pissed lithium battery.

I no longer have the EV2300 in my possession, and I don't intent to use them ever again.

Please install the EV2400 support components and run the EVSW. Send me a screen shot of any error message you get if this does not work.

Mike

This fix gets rid of startup error message, but I still have no load current and cell voltage reading is erroneous.

What do I do now?

Edit; Just like last time, it does something for a while, not giving me correct information, and now the Visual Basic Screen is going crazy again..  No communication can be established, same old same old garbage.

Have you configured the gauge yet for number of cells, CHEM ID , thresholds etc using the bqEasy screen.

Can you post a screen shot of the main SBS screen capturing status of all the registers. I want to see the status of the PD Status register

There is a 'keep logging'  box to tick below the menu bars that will make the readings all update automatically. This is where the resistor ladder comes in usefull as you can turn up the external supply biasing the ladder to make all the cells go up and down and make sure things seem to be working as expected before you go anywhere near a battery. Small steps.

Mike 

Same old question again, how am I suppose to do that capture thing, when there is no communication between this junk and the EV2400?

I am confused. You said the error message was gone and seemed to suggest you could see a screen but that the voltage was not updating. If that is the case you have communication ?? 

I think I can explain more clearly.

As soon as I download and install your program, it went fine.  I was just happy to see the error massage disappeared.  It seem to communicate for a while, but measured voltage was completely off.  I didn't pay much attention.  I immediate move to set up screen, set parameter as best as I know, and logged out, and restarted again, hoping this will fix the reading error.  Now, it was back to where it was before, error message cannot be established...  VB acknowledged is flagged. and nothing works.  If I hit a refresh button, same old message, read something like, "do this at your own risk..."

I hope I am clear, because I'm really irritated and upset now.  Can someone at your lab perform this exact same, and then give me the clear direction, instead of coming with quick fixes?

Won't you rather get a rat, like the one they use in the laboratory.  You inject different drugs and watch its behavior...

More the rat suffers, more you seem to enjoy.

You people have very strange taste.

All I asked for month, since accident that nearly burnt my place, is to have someone at your lab verify that this breakout box, whatever the garbage you call it is working. Is that too hard?

each time you run/restart you must remember to run as administrator. Also make sure you only have the one instance of the programming running, so no closed windows at bottom of screen. Also if you power down and restart you may need to wake the evm again by connecting top cell to pack + momentarily.

I asked and I ask again,

Did anyone in your own lab verified this box works with EV2400, running on Windows 8?

I have asked same question over and over, through telecon, and numerous times in this thread.

Why don't you simply answer "Yes", or "No," instead of inventing this garbage?

Kwang,

It is not permitted to post anything that is obscene like that on E2E. I can understand frustration but please lets keep it professional. Any content like your last line will be immediately removed.

Blake

Has anyone in TI attempted to use bq20z65evm in conjunction with EV2400, running on Windows 8 yet?  I've already tried bq20z65EVM with EV2300, and TI already confirmed that EV2300 should not be used on Windows 8 OS, and I can't make it run on EV2400 neither.

It shouldn't take anyone more than few minutes to setup to verify this setup, and be able to tell if they're useable. 

Your QA should have done this years ago, not your customer.

I had this setup more than 1 month, yet I haven't yet to apply any power, due to hardware/software flaws in either one of two units.  Never ever waited this long to get simple request replied.

I was afraid to hear this, forces me to do another upgrade. 

Few month ago, after I completed detailed testing of my own Battery Protection module implementing hardware, I made a difficult decision to move to software implementation, because I suspected it can be made to substantially higher efficiency, while offering some fantastic user experience, I've never heard of.  This is month ago.

More than one month past, nothing happened, mostly wasted waiting for your response, either through telephone support or this board.  I don't blame TI, because I should have posted question, asking TI if this unit is workable, because there is no signs anywhere in the universe, TI's own Evaluation Module have so much flaws.  It began with EV2300, no one ever suggested that this is engineered to be so finicky to computer's operating system. 

Why.. I have a GPS unit, made by some unknown company probably in a sweatshop overseas, purchased during Windows 2000 era, more than 10 years ago, and still works flawlessly.  All my hardware are ancient, never have this kind of problem, unless some printers where some manufacturer, corporate money counters, believes in this lunacy of "planned obsolescence," where they intentionally obsolete perfectly useful hardwareto boost its figures, therefore boost their bonus check.  That is why I refuse to deal with some company, such as Apple, infamous for lure in young naïve kids, and deliberately making unit useless after a short time of use.  If few more people behaves like some corporate pigs, we as a human race will be perished, because our natural resources are limited, very limited, and we run of them, we have no alternate plan, but the end of human civilization.

After so much of agony of replacing EV2300, now you're suggesting me, I need another upgrade to look at your Impedance Track technology.  This is outrageous.  Nowhere in you document, has it ever suggest some breakout boxes are obsolete, no where in the TI site hint people not to waste their time due to hardware obsolescence.  There is no excuse for this kind of craps.

It's not the operating system, but whoever wrote this software made sure that this bq20z65EVM is not workable outside EV2300, and when you read EV2400 document, it said it's backward compatible with older units, presumably including this junk too.  Looks like TI is contradicting themselves, by saying one in one place and flip it on the other side, and my last question is, you keep selling this flawed test box, that cannot worked, and cannot provide sufficient support causing millions of grief for its believers.  Is this TI way of doing business?  All these flaws should have been filtered out during your own development time, including your EV2400, and all your document should have stated clearly.  Lot of people will get hurt, because of your document discrepancy, and when you won't issue a refund/replacement for part, you really have a serious problem.