• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 62 subscribers
  • Views 707 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.

LM74610 has excessively long off time.

Anthony Rhodes
Prodigy 30 points
Other Parts Discussed in Thread: LM74610-Q1, SM74611

I have a simple PCB using a LM74610-q1 with a 0.22uF charge pump capacitor and a 27 volt bidirectional TVS across drain-source.

I find the off time to be about 17% and the frequency to be about 5Hz.   This is nowhere near the data sheet specs.

I am using a IPB019N06 FET with about 166 uC gate charge.

Do you have any suggestions as to what the do to get the off time down to the rated 2%?

My current rciit layout has short and written de traces

  • 0
    Prodigy 30 points
    Sorry, the message seems to have been garbled. I will re-phrase.

    I am using a LM74610 with a 0.22uF external charge pump capacitor. The FET is a IPB019N06 which has an ON resistance of 1.9mR. The gate charge is 166uC at 4.5V. The Vgs-th is 2.2V.

    I am finding that the switching frequency is about 5Hz, and the off time is actually about 6%.

    I tested the smart diode using a bench power supply set to 14.2V and the smart diode is on the low side. I have a power resistance box controlling the current. When I use more than about 23 ohms of resistance, The smart diode stops charging the capacitor.

    Why does the smart diode malfunction at low current? Why is the off time so much more than the rated value? Does the gate capacitance have a negative effect, causing the prolonged off time?

    The SM74611 works perfectly in this test fixture with an off time of 0.5% to 1%. I would use the SM74611 in my application, but I need about 30A continuous capability. Suggestions?

    -Tony
  • 0 in reply to Anthony Rhodes
    TI__Expert 5910 points
    Was there a reason why you picked 0.22uF. Also are you measuring the duty cycle by probing the gate pin.The probe would introduce leakage and affect the duty cycle. Its best to measure the output voltage. For some reason I cannot find the fet datasheet online. Please send me the datasheet of the fet to mathew.jacob@ti.com
    If the body diode voltage across the fet is <0.5V at the load current of >23ohms then yes that would affect the duty cycle.
  • 0 in reply to Mathew Jacob1
    Prodigy 30 points
    Here is a link to the PDF of the datasheet
    www.ti.com/.../snoscz1b.pdf

    I picked the 0.22uF charge pump cap (rather than 2.2uF) to increase on-off frequency in order to reduce the off time. I am using the smart diode as a blocking diode in an automotive voltage regulator for antique cars with generators instead of alternators. The original regulators were mechanical relay devices. I could use a big schottky diode, but a ideal diode is a better solution.

    I am finding that the prolonged off time is causing the voltage regulator to see a brief voltage INCREASE because it senses voltage from the generator side of the smart diode. When this happens it has time to react and lessen the dynamo output voltage. The regulator has about a 40Hz operating frequency (25mS period), so I need to get the off time below 25mS. preferably below 10mS. I am not sure exactly how short the off time must be. 3mS works fine.

    The SM74611 has an operating frequency of about 3Hz, with an off duty factor of 0.5 to 1%. So, the off time is about 3mS. This smart diode works perfectly with my regulator, but it can only handle 15A. So, I can't use it. But the 3mS off time is fine with my regulator

    If I need to have an off time of 10mS. Then for a properly operating LM74610 with a 2% off duty factor, then I need the operating frequency of 2Hz, but with the much longer off time, even a 5Hz operating frequency is too long. If I were to need the 3mS off time that the SM74611 provides, then the LM74610 would need to be working at 6.6Hz. This is why I used a 0.22uF capacitor. It should give a pretty short off-time. But this does not occur.

    I am measuring the operating frequency and duty cycle my measuring the voltage across the Drain-Source of the FET. I do find that when I measure from the gate of the FET, I cause artifactual slowing of the switching.

    I will double check the body diode voltage but I recall it being about 0.6 volts when the gate was off, when running with 23 ohms (14.2V, about 0.6 amps)