LM5122: Regarding DMAX limit of LM5122 in asynchronous mode

Hi Waqar,

Thanks for using the e2e forum.
Unfortunately, the LM5122 has no such feature as duty cycle limitation up to 50%.
The overcurrent protection feature will limit the duty cycle in case of strong current flows, but here the inductor current is monitored and the level of how strong the duty cycle is limited is only a result of this protection.

TI should have some PWM controllers for forward converter applications, which have this 50% duty cycle limitation.
Can you give more details about your application? Based on this it should be easier to find a suitable device.

Best regards,
Niklas

Hi Niklas, 

Thank you for the reply. My application is a 40 kW 2 Phase Interleaved boost converter. The input and output voltage will ofcourse require scaling in case I use LM5122. I found the controller to be a good option since it offers interleaving. The minimum duty cycle is 6%, while the maximum is 52%. 
The current during low input voltage conditions can be high, therefore I wanted to limit the duty cycle for a relatively slower precharge of the bus. So the limitation of duty cycle is only required in case of soft-start. Is there anyway to accomplish this with this controller?

Hi Waqar,

Thanks for the additional details.
20kW per one phase is quite a lot.
If the main goal of the duty cycle limitation is to slow down the charge time and reduce current peaks, the overcurrent protection feature and softstart feature of the device might be sufficient here.
The cycle by cycle current protection turns off the LO side gate if the CSP-CSN voltage exceeds 75mV. This happens every cycle, so duty cycle can be limited for a high inrush current scenario by setting the sense resistor accordingly.
The device has an additional hiccup feature which fully turns off the device for a short time if the overcurrent state persists, but this feature can be turned off via the RES pin.

If the high inrush current only occurs at initial start-up, the softstart feature can be used to limit the duty cycle during this time as well.

Please let me know if these features can be helpful for your application.

Best regards,
Niklas

Niklas,

Thank you for the detailed reply. You mentioned that 20kW is alot for per phase. Is there an alternative switching controller that you could suggest? Im curious as to what made you say that. Is LM5122 not a good fit based on my specs? 

Hi Waqar,

Can you give the the input voltage range and output voltage?
If the input of the power stage is the same as for the controller, you need to stay within the abs max ratings of LM5122.
The max input voltage would be 65V.
There is no limit for maximum current, but the higher this value, the more difficult it gets to control temperature and stay within the external component ratings.
(Just for example, even at 98% efficiency, 20kW would mean 400W loss dissipation on each phase)

Even reference designs of 6 phase interleaved LM5122 are normally not rated higher than 2kW.

Best regards,
Niklas

The input voltage is between 450-890V. The DC bus being regulated is at 940V. This amounts to a duty cycle of 52% and 6% respectively. The switching frequency per phase is 35 kHz.

My plan was to use a resistor divider network to scale the input and output voltages with 0.1% tolerance resistors, and I wanted to perform inductor current sensing with a LEM current sensor. 
You mentioned that it gets more difficult to control the temperature. Are you referring the temperature of the switching regulator itself?

Hi Waqar,

I am afraid the LM5122 device is not capable of supporting these voltages.
You can scale down the input voltage or use an external supply to power the controller, but for the high side switch, the SW voltage is required to provide the necessary Vgs voltage.
The rating of the SW pin is 105V abs max.
I also don't see any other parts within our low power DC-DC boost controller portfolio which could support these specs.

Best regards,
Niklas

Niklas,

I am using a non-synchronous boost converter topology so I wouldnt require a HS switch in this case. I should be able to operate LM5122 in asynchronous mode based on the example of a SEPIC converter given in the datasheet (Fig. 37). 
The other options that I am currently looking into in UCC28220 and LM5032. Whats your opinion on these parts?