- Ask a related questionWhat is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
Tool/software:
From my understanding in the simulation model you can disable burst mode by doing this procedure for UCC256403:
1. Set RBW to 2.7 kΩ.
2. Set BMTL and BMTH to 0.2.
I don't want under any circumstances to have the controller in burst mode due to voltage ripple.
In most scenarios the UCC256403 will not be in burst mode. The UCC256403 shall have an input voltage range from 60 V to 30 V, and Lr, Cr and Lm is tuned in to an nominal input voltage range of 58 V (Pout = 15 W).
The reason for tuning the input voltage range to 58 V is obvious, because in most scenarios the input voltage will be in that region. In some extraordinary settings the input voltage might be 30 V.
However, the load will also sometimes vary. During normal operation the load is approximately 2 Ω, but in other scenarios the load is up to 8 Ω.
When the input voltage is approximately 58-60 V, and the load is 8 Ω the controller is in burst mode. It does make sense that the controller would be in burst mode IF you did not disable burst mode. Because you have a light load, but it should not happen when you disable burst mode.
I've read some forum posts and also read the datasheet, and it seems like the simulation model do in fact sometimes go in burst mode even if it's disabled (BMTH and BMTL = 0.2).
And according to the datasheet, this happens in a scenario where VFB replica voltage is below 0.2 V. In the simulation model the VFB should not be 0.2 V.
VFBreplica = RFB * (IFB - I_optocoupler_current)
IFB current should be approximately 164 uA according to the datasheet, and the optocoupler current is approximately 154 uA (with some ripple) during the burst mode scenario. RFB = 100 kΩ
VFBreplica = 1 V in this scenario.
One thing I have noticed is that if I add a lot of capacitance in the output filter (more than should be needed), the controller is not in burst mode anymore. Obviously, the compensation network is affected by the output voltage ripple,
but I've tried to filter out the ripple in the compensation network, but for some weird reason the ripple is still embedded in the compensation network (and still in burst mode).
This gives me a suspicion that is not the compensation network, VFB_replica or the output voltage ripple that causes the controller to be in burst mode.
Attachment: The excel file with the parameters.
Note:
It doesn't seem like it's possible to upload the simulation file in the forum. So I might send the simulation to someone in the TI Staff support?
Copy of RCT_UCC25640x Design Calculator Rev4 (version 1)_58VDC_nominal_voltage.xlsx
UCC,
I entered RBW value to be 0kohm and 1uA as initial current in the default circuit downloaded from web. I don't see burst mode until the load is applied, when the load is applied, continuous switching is observed. However, once the load is removed, the device stops switching as output overshoot.
Can you share the application specifications and business case?
For actual performance of the controller, I recommend you to evaluate the EVM (UCC25640EVM-020 Evaluation board | TI.com).
Regards
Hemanth
