LM2621: Set up sample application for testing. Current with enable pin set to GND causing ~50uA instead of expected 2.5uA

Hi Jimmy,

Let me check and get feedback to you tomorrow.

Regards

Lei

Hi Jimmy,

In your setup, there will be additional leakages from the path of inductor --> diode --> RF1 --> RF2 --> GND (the major part) and from the input and output capacitors. These leakage are not counted in the max 2.5μA leakage in datasheet.

What's the sawtooth signal? Vout? And what's the voltage for each div in the picture?

You may type in Chinese

Regards

Lei

Hi Lei, 

Thank you very much for your quick response. Would you have any insight on how to decrease the leakage current from those components? 

For the signal output the max and minimum Vout on the sawtooth signal is 9.4V and 7V with a 2V per div. For the time, it is a 200ms per division. 

Thank you! 

Jimmy 

Hi Jimmy,

Jimmy Cao said:

how to decrease the leakage current from those components

The leakage from the path of inductor --> diode --> RF1 --> RF2 --> GND is decided by the topology. The only way of reducing this leakage is to increase the RF1 and RF2 simultaneously, for example, RF2 = 100kΩ and RF1 = 300kΩ. Please note that if RF1 and RF2 are too large, the Vout DC accuracy may decrease slightly. And also, as the RF1 increases, please decrease the CF1 proportionally to make the RF1*CF1 as constant.

Jimmy Cao said:

For the signal output the max and minimum Vout on the sawtooth signal is 9.4V and 7V with a 2V per div

This is abnormal. The Vout should be very stable without such a big ripple. 

Please check whether the 68μF output cap(C2) is correct. And what's the Iout during your measurement?

Regards

Lei

Hi Lei, 

Thank you for your swift response. I will look into modifying the resistor and capacitor values to see if I can decrease the leakage current. 

Regarding the Vout, I have the enable pin connected to a 2pos switch. One position to GND and the other to Vin. I noticed that when the EN is connected to GND the Vout is equal to Vin. When EN is connected to 5V Vin, the Vout is the sawtooth signal shown in the image previously sent. However, when the EN pin is left floating the output is stable with minimum ripple. 

When the EN pin is connected to Pin 6 node like in the datasheet, I still receive a sawtooth signal but with a shorter tooth amplitude. Would you have an explanation for this? 

For C2 I am using the T495C686M016ATE250 (CAP TANT 68UF 20% 16V 2313. 

When measuring the signal, Iout is connected to a device that pulls just a few uA. When a higher load is presented the signal is more stable. 

Thank you for your help, 

Jimmy 

Hi Jimmy,

Jimmy Cao said:

when the EN is connected to GND the Vout is equal to Vin

It is normal. Because when the LM2621 is disabled, the Vout equals to the Vin if ignored the voltage drops on inductor and diode.

Jimmy Cao said:

When EN is connected to 5V Vin, the Vout is the sawtooth signal shown in the image previously sent. However, when the EN pin is left floating the output is stable with minimum ripple

It's quite abnormal. We can find out the reason here first then check your second question.

Could you help double confirm that when the 2pos switch switches to 5V, is the voltage on EN pin a stable 5V?

Regards

Lei 

Hi Lei, 

Thank you for the response. Sorry some things got side tracked for a bit. 

I am at a loss for the issue that I am seeing as well. I'm currently working on a board to test this circuit on as it looks like I am having a lot of issues with my hand made circuit. 

Regarding your last question, that is correct. The EN pin is connected to the 5V input of a lab bench power supply. I have not been able to get a stable on voltage yet. 

Thank you, 

Jimmy