• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 218 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ25570: Low input impedance OK?

Kevin Kachikian1
Prodigy 150 points
Part Number: BQ25570

Tool/software:

Howdy,

I am considering designing with this energy harvester and am a little confused about possible input sources. The recommended designs show high input impedance sources such at TEG's or solar cells or vibration elements.

I am wondering if there is some design limitation whereby very low input impedance sources can not be used. In the intended application, the output storage element is a 4.2v LiPo. Specifically I am energy harvesting from disposable batteries which, when fully charged, have an ESR of around 100mΩ and drop to approximately 2Ω when nearly depleted - all very low as compared to the recommended input devices ESR's all in the MΩ's I think.

The question is: can very low input impedance sources be used as power inputs with this part WITHOUT a series resistor? I haven't found anything that would suggest this would not work, but the EVM notes say to use at least a 100Ω in series with the input voltage source when using a bench power supply. A little confused about this...

Any advice is appreciated.

Thanks,

Kevin

bard-designs.com

  • 0
    TI__Mastermind 18005 points

    Hi Kevin,

    The device is designed for use with high impedance sources, for this reason it is recommended that it be used with a 100Ohm series resistor with a bench power supply. This is to imitate a high-impedance source for evaluating device behavior with sources such as the ones you listed. The input voltage should be at least 400mV lower than VBAT_OV. Low impedance sources will not be able to take advantage of the MPPT feature since input voltage is not likely to drop. A low impedance source will result in the device pulling the full cycle-by-cycle current limit of the charger from the input source. 

    Best Regards,

    Juan Ospina

  • 0 in reply to Juan Ospina
    Prodigy 150 points

    Thanks for the information Juan. I have 2 followups for which I am a little confused:

    I am using the bq25570evm-206 configured from the factory to output 4.2v. I have the VOC_SAMP jumper set to 80%. I have the /EN jumper set to GND to enable the chip. I have VOUT_EN set to GND to disable the onboard buck converter. No other jumpers are in place on the EVM board. Nothing attached to the VSTOR terminal block.

    1) When I apply direct power to VIN of around 1.15V via a mostly depleted AA battery, I see the VBAT output outputting 4.15V when unloaded - nothing attached to VBAT - great. The instant I attach a LiPo (charged only to 3.566V), the VBAT output drops from 4.15V to 3.587V (with 7mA charging the LiPo and a 34mA draw on the battery) - not what I expected. Why doesn't the bq25570 maintain the 4.15V to the LiPo? Even when I apply 3.5v from a beefy bench supply to VIN, the output unloaded is 4.15V (good), but when I attach the LiPo to VBAT, the voltage drops to 3.822V (with 92mA charging the LiPo and a 115mA draw on VIN). So it seems even under ideal conditions, the VBAT output does not maintain its regulation under load of a low voltage LiPo. What am I not understanding?

    2) Again when I apply a mostly depleted battery to the input when the LiPo is not connected - unloaded output, I see the VIN voltage at 1.15V. The instant I attach the LiPo load, the VIN drops to 0.87V (with 7mA charging the LiPo and 34mA draw on the battery)? Again, I must not be understanding something because I expect VIN to remain at 1.15V given the battery is a low ESR device and the bq25570 does "magic" to not load down the expected high impedance input power source. Even when I use a brand new fully charged AA as VIN, the unloaded VIN is 1.63V but when I connect the LiPo load to VBAT, VIN drops to 1.51V (with 41mA charging the LiPo and 121mA draw on the battery). I would expect this to stay at 1.63V due to the (assumed) high input impedance of the bq25570.

    Can you please clarify mostly #1 above?

    Thank you,

    Kevin

  • 0 in reply to Kevin Kachikian1
    TI__Mastermind 18005 points

    Hi Kevin,

    Kevin Kachikian1 said:
    1) When I apply direct power to VIN of around 1.15V via a mostly depleted AA battery, I see the VBAT output outputting 4.15V when unloaded - nothing attached to VBAT - great. The instant I attach a LiPo (charged only to 3.566V), the VBAT output drops from 4.15V to 3.587V (with 7mA charging the LiPo and a 34mA draw on the battery) - not what I expected. Why doesn't the bq25570 maintain the 4.15V to the LiPo? Even when I apply 3.5v from a beefy bench supply to VIN, the output unloaded is 4.15V (good), but when I attach the LiPo to VBAT, the voltage drops to 3.822V (with 92mA charging the LiPo and a 115mA draw on VIN). So it seems even under ideal conditions, the VBAT output does not maintain its regulation under load of a low voltage LiPo. What am I not understanding?

    If I understand correctly, you have the charger configured to charge VBAT up to 4.15V when VIN is applied which is the voltage you see on VBAT, but this voltage drops to 3.587 when the battery is connected rather than maintaining a 4.15V output. This is expected since the voltage at VBAT will be equal to the battery cell voltage. The VBAT voltage will continue to increase as the battery is charged, all the way up to 4.15V. VBAT is not meant to maintain a constant regulated voltage since the battery capacity would always be too large to just immediately charge it up to 4.15V. The voltage regulation comes into play by stopping additional charge once the battery voltage has risen to 4.15V.

    You can crudely model the battery as a circuit where RBAT is very small and CBAT is very large. CBAT slowly increases as the battery is charged:

    Kevin Kachikian1 said:
    2) Again when I apply a mostly depleted battery to the input when the LiPo is not connected - unloaded output, I see the VIN voltage at 1.15V. The instant I attach the LiPo load, the VIN drops to 0.87V (with 7mA charging the LiPo and 34mA draw on the battery)? Again, I must not be understanding something because I expect VIN to remain at 1.15V given the battery is a low ESR device and the bq25570 does "magic" to not load down the expected high impedance input power source. Even when I use a brand new fully charged AA as VIN, the unloaded VIN is 1.63V but when I connect the LiPo load to VBAT, VIN drops to 1.51V (with 41mA charging the LiPo and 121mA draw on the battery). I would expect this to stay at 1.63V due to the (assumed) high input impedance of the bq25570.

    This is going to depend primarily on your input source. The difference between a LiPo battery attached and not attached is the magnitude of the current load that is pulled from the input source. When no battery is attached, the current being pulled from the input is very small quiescent current assuming no VOUT load. The small load means AA's voltage will not be pulled down significantly. Once a target battery is connected that is below the configured 4.15V, the charger will begin to pull larger amounts of current from the input source in order to charge target battery. Based on a 121mA draw from the source battery, and a 1.63-1.51 voltage drop, this seems to indicate about 1 Ohm of resistance. A 1.15- 0.87V drop at 34mA of load current indicates an 8Ohm resistance. This input impedance is a property of the source battery and the resistive path between the battery to the charger, and not really something that the charger has control over.

    Best Regards,

    Juan Ospina

  • 0 in reply to Juan Ospina
    Prodigy 150 points

    Thanks for the thorough explanations Juan! Truly appreciated.

    In your 2nd paragraph you mention "the charger will begin to pull larger amounts of current from the input source in order to charge target battery." when the LiPo load is connected. That makes sense. The thing that confuses me is I thought the BQ25570 modulates the VIN current draw so that it will not load down such high impedance power sources as solar cells or TEG's. In these use cases, I thought the BQ25570 "goes easy" on the power input so as not to  try to draw hundreds of mA that are not available on VIN to begin with. IOW, I thought the MPPT samples the input to figure out how much of a load it is allowed to put on the VIN power source without it sagging or drooping VIN thereby causing inefficient energy transfer. In essence, I thought it modulates the current draw to find the sweet spot of energy transfer (maximum current at maximum voltage) for highest power delivery. Do I understand that right?

    If so, then referring back to my #2 above, why would the VIN drop from 1.15V down to 0.87V when VBAT is loaded with the LiPo? I thought the whole idea of MPPT was to regulate the current draw to not allow sagging or drooping of the input voltage (especially required for TEG's and solar cells). No?

    Thanks,

    Kevin

  • 0 in reply to Kevin Kachikian1
    TI__Mastermind 18005 points

    Hi Kevin,

    Kevin Kachikian1 said:
    If so, then referring back to my #2 above, why would the VIN drop from 1.15V down to 0.87V when VBAT is loaded with the LiPo? I thought the whole idea of MPPT was to regulate the current draw to not allow sagging or drooping of the input voltage (especially required for TEG's and solar cells). No?

    The MPPT is meant to prevent droop below a certain voltage which is configurable using the VOC_SAMP pin, but some droop will be required as current will be pulled from the input source. For a solar cell this MPPT point would be configured to the cells Maximum Power Point voltage.

    Best Regards,

    Juan Ospina

  • 0 in reply to Juan Ospina
    Prodigy 150 points

    OK Juan. Thanks for all your help!

  • 0 in reply to Kevin Kachikian1
    TI__Mastermind 18005 points

    Happy to help, Kevin!