Boost converter startup problem

[Knife]

Hi all

I designed a boost converter for LED driving which is specification listed below. I used TPS40211. It doesn't have simulation tool on TI but TPS40211 is same with TPS40210. You can see simulation circuit below and I built this circuit.

Input Voltage : 6 V - 8,5V
Output Volate : 70 V
Output Current : 0,2 A

The LED's are flashing and boost converter doesn't work at 6 V - 8,5 V input range. If I increase input voltage to 10 V - 12.5 V range, it works properly.

I'm thinking it causes from inductor or sense resistor or input cap. I tried to increase input cap but it didn't work. I will try to increase inductor value or decrease current sense value.

Does everyone have a comment or solution for this issue ?

 

[FvM]

Are you sure that it's only a startup problem? In this case the circuit would run with 6V Vin after start with higher voltage or increased soft start ramp time.

What's the inductor saturation current?

 

[Knife]

Thank you for your answer. The circuit doesn't work 6 V after start up. You are right, I examined that after write this post. I increased soft start timer but it didn't work.

I used 3 pcs in series MSS1210-223ME_ which saturation current is 6.8 A. I'm planning to use 1 pc MSS1210-683ME_ later which saturation current is 3.8 A.

I worked last night on the circuit and I changed compensation network. The circuit works 8 V now. I think compensation network may cause this problem. The TI design tool choosed 1.7 kHz cross-over frequency. The circuit works 260 kHz. How can I select cross-over frequency ? I read cross-over frequency should be one tenth or one fifth of working frequency in Trilogy of Magnetics (Wurth Electronic). I found an application note for TPS40211. If I'm right I calculated 96 kHz cross-over frequency and 500 kHz working frequency for this application note. I calculated all of TPS40211 application notes and I saw cross-over frequency is always one fifth or one fourth of working frequency.

**broken link removed**

 

[FvM]

You have 5A overcurrent limit, so 6.8 A saturation current shouldn't be a problem.

Instable feedback loop can cause triggerring of the overcurrent protection, the loop gain changes with the input voltage, in so far this can be the reason. I believe there are instructions for feedback loop setup in the datasheet.

 

[vaggabond]

Can it be related to the min off time? In the datasheet it's listed as 170ns. If your operating freq is 500kHz, the max duty cycle your boost can achieve is ~90%.
Your boost ratio is ~10:1 (7V->70V) , so it might be borderline.

To see if this is the cause, you can try shorting some LEDs in the load, to decrease the VOUT voltage to ~50V, and see if it can start normally then with VIN=6V.

 

[Knife]

Yes, you may be right, but I tried to decrease output voltage and it didn't work again.

I'm working on it but I didn't solve this problem yet.