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.

LM3550: Flash does not occur with EOC low

Part Number: LM3550

I use optimal charge mode, enshure that the EOC pin (with pull-up resistor) is LOW, set bit 2 in register at address 0x10 to enable Flash, but LED Flashing does not occur. When I repeat this cycle (enable Flash, wait 200ms (Flashing time was set to 192ms), then Start optimal charging, EOC pin goes low at once, (because supercapacitor already charged)), wait 1s, after 10-15 such cycles Flash occur as it should be. Communication with LM3550 is correct, because each sent byte is acknowledged.

Register configuration is as follows:

Address Value

0xB0      0x0B

0xA0      0x00

0x10       0x04 when in Flash mode

0x10       0x02 when in Optimal Charge mode

0xC0      Power on value

0xD0      Power on value

  • Hello,

    Have you tried using the FGATE bit disabled to see if things behave?  If you using the external EOC flag to trigger your flash, disabling the gating should ensure the part is at EOC and should allow a flash.

    Greg

  • Hello,
    Sorry for delay, when FGATE bit disabled, and using charging mode to 4.5V in my case, flash occur as should be, each time when I send command.
  • I was try to use gating(FGATE bit = 0), because incorrectly think that Optimal charging mode not work without gating(FGATE bit = 1),
    now I try Optimal charging mode without gating and Flash occur. In my case gating not needed at all, I use EOC pin state to determine, when I can start Flash. But gating function still not work for me, interesting to figure out this issue.
  • Hello,

    I am glad that at least for your current implementation that removing the gating is working for you.  I personally have not run into the gating problem in the past and will need to dig into this further to try and figure out what is going on.  Can you share the super cap part number you are using as well as the ceramic output cap part number?  I am not sure this matters but if the ESR of the supercap is high and the caps leakage current is high, it could be causing the part to think the output is not fully charged all the way even though it is close.  Again, as I have never seen this issue this is just a place to start.  Can you share a scope photo of the output voltage, output current and EOC pin state during a good flash followed by the non-flash condition?  I might be able to see something in the image that gives a clue.  Thanks.

    Greg

  • Hello,

    Supercapacitor SCCR16B205SRB, for output cap I have only description: 2,2 uF 1206 10% 16V X7R

    Here CH2 is EOC pin, CH1 is supercap voltage. 

  • It is interesting that there appears to be something loading the cap.  I am assuming the larger dips are flash pulses?  I am curious as to what the secondary drop is from.  Do you have other loads on the cap?  Additionally, it is odd that the EOC voltage is moving.  Is there some sort of loading?  The change in EOC voltage is synchronized with the recovery of the supercap.

    What is the pulse width of your flashing?  Freq?

    Thanks

    Greg

  • Also, I am assuming you have two supercaps in series as the caps look to only be rated to 2.7V each. Please confirm. If this is the case that is a good amount of ESR which could be causing the EOC to transition.
  • In the first time on the photo EOC is Low and I send command Flash, Supercap start discharging and at once, EOC going High, flash time is 192ms (but I use fixed delay of 200ms) after that I start Optimal Charging as had been in previous cycle, after some time EOC is Low, but I send new commant to Flash after fixed delay of 500ms, thus some amount of time we see EOC Low,
    then I again send command Flash, but with Gating, as usually EOC go High at once, Flash does not occur, and after fixed delay of 200ms I start Optimal Charging again, EOC goes Low at once, because Supercap not been discharged, after fixed delay of 500ms that cycle continues.
    Supercap do not have any other loads. I use 2 supercap in series.