Other Parts Discussed in Thread: TPS2660
Tool/software:
Hi TI team,
We are considering designing in the TPS1685 into something similar to a diode "or'ing" application- but with strict load-balancing requirements. In this application, two (2) individual 40VDC power supplies are connected to the inputs of two (2) individual TPS1685 devices. The outputs of these two TPS1685 devices are then connected through diodes to a common output net (Vout) from which the total load current is drawn. The power supply output voltages, output impedances, and current sourcing capabilities are all within ±0.01% of each other. (i.e. they are very well matched)
A few considerations follow:
- Each power supply can source ≤2A of current
- The application requires that the equal current sharing between the two supplies must be maintained- even when the current is ≤0.1A.
- The capacitance added to this unified output net Vout is ≈12mF. From a cold-start, we cannot draw more than 1A of current without causing a fault on the power supplies, but we can allow several seconds to charge the capacitor bank if necessary.
- Ideally, the over-current blanking time would be ≈30-100mS
My questions are as follows:
- I have completed the calculation worksheet and have attached it here to this post for your review. Can you please review and let me know if you see any concerns for this application?
- Can you help me understand how the load-balancing is achieved between two TPS1685 devices placed in parallel? It seems that the only way to balance the output current of two (2) TPS1685 devices would be to allow the RDSon of the internal mosfet of one of the two devices (i.e. the one which is sourcing more current than the other) to increase, so that the other e-fuse can "catch up" in terms of output power it is delivering. Is this an accurate summary of how the current is balanced between the two?
- Considering that these devices have very low RDSon and are capable of sourcing very high currents, do you think we will have any issue trying to balance the currents at such a low output current level? With such low output currents is there enough "Dynamic Range" to get the two e-fuses to equally balance?
- When do you anticipate these devices will be released to production? It seems that the datasheet is still showing "Advance Information" and they are listed as pre-production. There also doesn't appear to be any stock on the evaluation modules, but there is some small quantities of the ICs themselves.
Thank you in advance for your help and I look forward to hearing back from you on these questions!
TPS1685_CALC_P0_WLBY.xlsx
