UCC39002EVM: UCC39002-EVM does not share current

Hi Melek 

Thanks for reaching out. 

Could you please share the schematic of your load share design. I'll review the design, compensator design values and other parameter also. if you have followed our EXCEL calculator tool for the design, please share that too, for better understanding of the design.

For initial debugging, since LS pin is high and EAO pin is also high, it means controller is trying to share the load by adjusting the feedback to max, but by some reason it's not able to adjust the feedback. It means, either feedback to one module is not connected properly or connection at adj pin is not proper. Please check for these points and also, check weather output voltage of module, one which is not sharing load is as expected. Just check once all the connection are proper as mentioned in user guide of EVM. UCC39002 Advanced Load Share Controller Reference Design (Rev. A)

Regards,

Anil

I have just started designing my schematic, and before sharing it, I would like to ask something. I will be paralleling two modules, and each of my modules is rated at 25 A, 12 V output. However, in the user guide you shared, it is mentioned in the caution section that a maximum of 10 A per module and up to 15 V output is recommended.

In this case, since my output voltage is 12 V, does that mean I can load each module with 12.5 A?

If you could clarify this first, I will adjust my schematic accordingly.

Thank you in advance.

Hi Melek,

Evaluation board is designed for maximum of 10amp from each module with maximum output voltage of 15voltages. Load current is directly flowing through the eval board only and its trace width for the load current to flow was designed for 10amp. So, recommended loading from each module is 10amp.Reducing the output voltage will not change its current carrying capabilities of the board. Still, I think you can push it for 12.5amp, generally this much of margin is considered while designing a PCB, but cannot guarantee the proper operation, you can test it and check for the any abnormality, mostly heating, if that's in acceptable range for your design requirements, you can go ahead with design. Provided all, I would strongly suggest, keep the temperature in check at all times during initial testing, otherwise board can get permanently damaged. other than load current path, sensing parameters, filter design, feedback adjustment can be modified as per requirement.  

Sorry for delayed response as there was public holiday in TI India followed by weekend.

Regards,

Anil

Hi Anıl, 

I understand that on the evaluation board I can draw a maximum of 10 A per module. However, since I will be using the UCC39002 in my own design, I should be able to handle currents higher than 10 A by designing appropriate trace widths. Am I understanding this correctly?

Also, is the UCC29002 design tool Excel file you shared still applicable and up to date for use with the UCC39002 as well?

Hi Melek,

Melek Toker said:

I understand that on the evaluation board I can draw a maximum of 10 A per module. However, since I will be using the UCC39002 in my own design, I should be able to handle currents higher than 10 A by designing appropriate trace widths. Am I understanding this correctly?

My answer was with respect to using of evaluation board, if you are doing your own design, definitely yes, you can design it for higher current also. 

Melek Toker said:

Also, is the UCC29002 design tool Excel file you shared still applicable and up to date for use with the UCC39002 as well?

Yes, EXCEL calculator tool is same for both the device. I would highly recommend following EXCEL calculator. Other than different operating temp range, all the operation are mostly same for both the device. 

Let me know if you need any other help in understanding of working of the device.

Regards,

Anil

Hello Anıl , I have successfully managed to parallel two power supplies. I noticed the following situation: when the voltage difference between the two power supplies (for example, at 12V output) exceeds 0.3V, one module starts taking all the current. However, when I equalize the output voltages of the two power supplies, this issue disappears. What is the reason for this, and is this situation mentioned in the user guide or datasheet?

Secondly, I have determined my design values based on your Excel table. I am using the low-out, high-side current sense application. My values are: Rshunt = 2 mΩ, Radjust = 49.9 Ω, ACSA = 100, using 100kΩ and 1kΩ resistors. With these values, the current sharing from the power supply directly is 7.81A and 6.24A respectively. Do you have any suggestions for improving this?

Hi Melek,

Sorry for delayed response as I was out of office yesterday.

Voltage difference is based on module's specification decided by user as in how much of maximum voltage adjustment is possible on each module, generally it has been considered as 2% to 5% of output voltage. Based on maximum voltage adjustment allowed per module Radj resistor is selected so that output voltage adjustment should not exceed the maximum adjustable voltage limit, if Radj is chosen based on 0.3V max allowable voltage change and you are giving the difference more than it than controller will not been able to adjust the voltage in order to equalize the load. In your case I think your Radj value is low, even though it depends on RSENSE also, still I feel it's low. Please cross check the values with EXCEL calculator also. If you could share EXCEL file with me, I'll review and let you know if any change is required.

Regards,

Anil

No problem :)

i can't share excel file but i added the images and actually in our design Rsense=20 ohm but i can't choose 20 ohm bcs Radj resistor is - value :(