LNK605: how will the frequency and on time change with CC/CV operation

Posted by: tuotuo1225 on 04/06/2011

Hi,
I'm using LNK605 and the DER-261 for my flyback converter. I feel confused about the frequency change in normal operation:

(1) In CV mode, as the datasheet said, the switch is turned on/off with a state machine to keep the output voltage constant. However,it seems that the on time does not vary significantly when the load current decrease. But the switch seem to be off for the whole switching period sometimes.With a smaller load current, the complete off period occur more frequently. Does this mean the state machine can only turn off the switch for the whole period but cannot change the on time?

Then what control the on time of the switch? will it simply turn off when the current limit is reached?
So such "complete-off" period is the normal operation of LNK605, but not a sign of oscillation in some conventional duty-cycle controlled regulator?

(2) In CC mode, when the load is larger, the output is supposed to drop, will this increase or decrease the switching frequency?

(3) In CV mode, will the frequency change with the output voltage in the same way as in CV mode?

Thanks.

Comments

Hi,

In response to number 1. The control method used by LNK-II is "On/Off" control. This means that for the CV portion of the output characteristic the voltage is regulated by the controller skipping entire switching cycles at the fixed nominal frequency as listed in the datasheet. The primary current will go to the current limit of the device as also listed in the datasheet. The state machine will also reduce the drain current limit at light load to reduce transformer flux density.

So the short answer is yes, the state machine can only turn off the switch for the whole period and it cannot change the on time. To expand, the on time is actually primarily a function of the inductance value and the DC bus voltage. Using VDC = Lp*di/dt you can see that the on-time, dt = Lp*di/VDC or doing the "calculus": ton = Lp*Ilimit/VDC

2) The frequency decreases as the voltage on the output decreases. This is dictated by the frequency ratio (FRATIO(CC)) in the datasheet.

3) No the frequency is fixed in CV mode and the cycle skipping on/off control method is used.

Thanks for your explanation. I have seen the CC operation with small load resistors.

(1) However, I don't know why there is still frequency change in CV mode:
The flyback was tuned to around 12V for CV, 330mA for CC(measured with 32ohm and 28 ohm resistance). When I used a 50ohm resistance, the current was 239mA and voltage was 11.95V, I think it should operate in CV mode, but the frequency is still fluctuating, from around 77kHz to 86kHz. How can I fix it?

(2) The datasheet said the frequency is programmed by Ifb * Ton. But based on my observation, when input changes, Ton changes, frequency does not have a clear relation with Ton. The maximum frequecy do not occur with maximum or minimum input. Also, the frequency does not change continuously. I saw it reached 86.8kHz, skipped a cycle, and then jumped to 80.4kHz,75.1 kHz and 76.7kHz in the next 3 switching periods. Why did this happen?

(3) The period skipping does not have a regular pattern. I thought it should occur more frequently with higher input. But it also does not change continuously. Sometimes it occurs after 8 cycles and the next skipping occurs after around 20 cycles. With light load, the pattern is more messy. 3 consecutive periods may be skipped and then another 3 are not skipped, and then maybe another one skip after 2 or 4 periods. I cannot see any periodicity in the skipping pattern. Does this mean my circuit is not stable especially with light load?

Thanks.