BQ24075 Discharge Current

Hi Robert,

The IC has a short circuit protection circuit built in. If during battery supplement mode, the voltage at OUT drops by more than VO(SC2) (250mV) below the BAT voltage, the output is turned off if the overload exists after for more  than tDGL(SC2)=250us.

The short circuit recovery timer then starts counting. After tREC(SC2)=60ms, OUT turns ON and attempts to restart. If the short circuit remains, OUT is turned OFF and the counter restarts.

Can you measure your OUT and BAT pin, what is the voltage difference between them?

Thanks

Tahar

Hi Tahar,

Thanks for the quick response.

As I said, I know that the short-circuit protection is being triggered.  I'm getting 250mV between OUT and BAT; after 250us of this, OUT turns off.  Then, about 60ms later, it comes back on.

What I don't understand is why I get such a large voltage drop between OUT and BAT when the current that I'm drawing is less than 4A (actually only about 3.5A when short circuit is triggered).  Page 3 of the datasheet states that the maximum recommended Ibat (discharging) and Iout is 4.5A.  Why would this be specified at 4.5A, when in practice the part won't allow anything greater than 3.5A before it thinks there's a short circuit and turns itself off?

More importantly, is there a way around this?  Is there any way to disable short-circuit protection, or prevent it from kicking in so early?

Thanks again.

Hi Robert,

Actually the part looks at the difference in VBAT and VOUT. If it is above 250mV, it shuts down without considering the Iout. And that is how the part normaly operates.  

What are you using in your system rail, is it another converter?

One of the solutions is putting an external cap at the output (System Rail) so the surges of the current will come from the cap instead of the battery. This may help preventing the drops on Vout and VBAT.

Also, can you try to use different battery type?    

Thanks!

Tahar

Hi Tahar,

Yes, I understand that the part looks at the voltage between VBAT and VOUT.  However, while in discharge mode, this voltage is simply equal to system current x impedance of bq24075.  On page 8 of the datasheet we see that the path between VBAT and VOUT is approximately just a MOSFET, whose impedance is primarily resistive.  Therefore, the voltage between VBAT and VOUT is directly proportional to system current.  At my current of 3.5A, I am getting 250mV drop so that tells me that Zinternal = 0.25V / 3.5A = 66mohms.  To be able to reach the rated 4.5A of output current, this impedance would need to be 0.25V / 4.5A = 56mohms or less.

At VOUT, I have two DC/DC converters to provide my system voltages (5V and 3.3V).  I have tried increasing the capacitance on VOUT from 100uF all the way up to 4700uF, with no improvement.

I cannot use a different battery type; I am restricted to this particular lithium-polymer cell.  However, the characteristics of the battery should have nothing to do with the voltage drop from VBAT to VOUT.  As I said above, the only two parameters which affect this voltage are the system current and the impedance through the bq24075.

Thanks again.

Hi Robert,

I agree that on page 8 on the datasheet, between VOUT and BAT is just a MOSFET. But, if you look at the electrical characteristic table, Vo (sc2)  has tolerance associate with it. The figure on page 8 is shows just a simplified block diagram, and we do not publish a detailed block diagram.

If the external cap solution does not solve the issue, then the solution you proposed in the previous post will work too.

Thanks!

Tahar

Hi Tahar,

What I'd really like to know is how TI arrived at the specification of Iout and Ibat(discharging), which on the datasheet are specified as 4.5A max.  I've tested probably a dozen parts at this point, with several different power sources (batteries), and have never been able to even come close to 4.5A through the OUT pin.  As an "easiest case" scenario, I connected a 3.7V bench power supply with maximum current output of 6A to the BAT input on the bq24075EVM evaluation board, and connected a power rheostat from the OUT pin to GND.  The maximum current I was able to sustain through the OUT pin was 3.07A before short circuit protection was activated.

Unless I am doing something wrong, this is in direct contradiction to the Iout max and Ibat(discharging) max values listed on the datasheet.  At best, this is extremely misleading.  In the future, I will think twice about trusting the specifications on TI's datasheets.  If I had known that the maximum output current was only 3.0A max (30% lower than specified!), I would have saved my company a whole lot of time and money by specifying a part that actually met the requirements for our product.

Thanks again for your consideration.

-Bob

Thank you Charles for the clarification…

Robert, here is another post that addresses your question

http://e2e.ti.com/support/power_management/battery_management/f/179/t/132274.aspx

I will close this issue and please let me know if you have more questions

Thanks

Tahar

Hi Charles,

Thank you for your response, this finally answers my question. 

I still maintain, however, that the datasheet is misleading and does not provide enough information.  The RDSon parameter for the battery FET does not actually appear to be listed anywhere, so there would be no simple way to calculate the minimum Ibat(discharge)(max) current using the information provided on the datasheet.  At the very least, I would suggest that TI add the MOSFET RDSon to a revision of the datasheet; even better, add the 2A minimum Ibat(discharge)(max) as well.

I will look into the bq24030, however I'm not sure this will fit my application because it lacks a SYSOFF pin (I'm also using the bq27500-v130 fuel gauge), so I would have to add a series MOSFET switch between the battery and charger to add that feature.  I will probably just pursue my original idea, to add a pair of series MOSFETs in parallel with the bq24075 battery MOSFET during discharge.

Thanks again for your help.

-Bob

The RDS0n is the Battery drop out voltage, second line under the Power Path specs.

Do you have a good thermal design?  Six to eight vias in the power pad and 2 oz copper to spread the heat out.

Every power IC on the market has a wide range of operation based on how good the cooling is and all manufactures set peak proformance with optimal conditions and the designer has to derate the part based on how it will be used.  Do you have the EVM?

I did a test on one EVM I had and it put out 4.2A for 1 second, see plot.  My 1W resistor started smoking so I had to discontinue the load, but at that time the short circuit protection had not kicked in.  The drop from the BAT connector to the OUT connector was 225mV and it was probably closer to 210mV at the chip so I believe that the current could go 20% higher.

Hi Charles,

I suppose one could calculate the RDSon from the Vdo(BAT-OUT) spec.  Thanks.

My thermal design should be adequate - what's more, I was getting identical behavior on my EVM, whose thermal design I think should be pretty ideal.  By connecting a power rheostat to the output of the bq24075, I was not able to source more than about 3.1A for any length of time.  Which is fine - it's within the spec you derived for me in your previous post - but it seems as though, even with a good thermal design, the RDSon of that MOSFET can be on the high side of its range on any given part.

Anyway, thanks again for your help.

-Bob