• TI Thinks Resolved

# SN6501: isolated power supply from 12V to 12V

Part Number: SN6501

Dear,

Current I'm doing a design which has to be an isolated power supply.

The source voltage I could use is 12V (minimum could be 10.4V) at the primary side, I used a 3V3-LDO to get 3V3 first, and then increased 3V3 to ~13.86V (2*2.1*3.3V) by SN6501 high output voltage application, and then decreased 13.86V to 8V2, another 3V3 LDO is used to get another 3V3. The 3V3 at the primary and secondary side is used for the optical coupler, and 8V2 is used for the power of input detection circuit.

I know this is not good for the efficiency because the voltage from 12V to 3V3 and to 13.86V and then to 3V3 again. But the input voltage of SN6501 only can reach 5.5V max.

Do you have some better idea which could realize this function? Maybe switch power supply could, but I don't have the concept currently, do you have some low cost and low current consumption concept?

Thanks a lot and have a nice day!

• Hi Jian Jin,

Thank you for posting to E2E! Does the optocoupler require 3.3V power supplies on both of its sides?

If you'd like, digital isolators like ISO7710 or ISO7021 can be used in place of the optocoupler here to isolate signals. Devices with different quantities of forward and reverse channels are available, and they can operate with independent 3.3V or 5V power supplies on each side.

Regarding the SN6501,it is possible to use a transformer with a smaller turns ratio, like N = 1.5, to isolate the input 3.3V, reducing the voltage drop across the second LDO.

Which transformer is currently being used?

Respectfully,
Manuel Chavez

• In reply to Manuel Chavez:

Dear Manuel,

I found PMP30440 in TI, which introduce a kind of isolated power supply which could from 12V to 12V.

But I cannot find 750317829 which is for 12Vin, 12Vout@100mA (pls see below).

My question is if I use 12VDC in below circuit, and selected the transformer as WE 760390012 (1:1.1), but the related Vin of 760390012 is 5VDC, what's the problem of the transformer after that?

And it's always my headache which is I cannot find the Pspice file of the transformer in WE website. Do you have some better idea for the simulation?

Thanks a lot! Have a nice vacation!

• In reply to Manuel Chavez:

Hello Manuel,

For this 12V-12V isolated power supply design, which kind of transformer do you prefer in your opinion? I cannot find the suitable Vin=12V transformer in WE.

Have a nice day!

• In reply to JIAN JIN:

Hi Jian Jin,

Thank you for your patience! 750317829 is not a catalog part, but it is available from Würth Elektronik if you contact them via their website.

As you said, transformer 760390012 will likely not work for 12V:12V operation due to its low V-t product. If 750317829 is not used, Würth's 750315240 or Coilcraft's DA2303-AL should be suitable replacement transformers. For systems with >5V input and output voltages, we do recommend using the SN6505B instead of SN6501 for greater power capabilities and protection features.

I have reached out to Würth to check if they have spice simulation files for their parts; please allow me to follow up with you regarding this final point.

Respectfully,
Manuel Chavez

• In reply to Manuel Chavez:

Hi Jian Jin,

Additionally, below is the datasheet for transformer 750317829. This device may be sampled or purchased by contacting Würth.

750317829.pdf

Manuel Chavez

• In reply to Manuel Chavez:

Dear Manuel,

Thank you very much for your information and sorry for my late response because there's no e-mail to remind me. :-)

1. For the simulation model, yes, I found the related LTSpice model, but do you have the LTSpice model for SN6501/SN6505? (Because I didn't find it in the TI website, only see its Pspice model)

2. The transformer you recommended, WE, 750315240, Vin is rated 5V, is it some problem if I used in Vin=12V application?

3. I did a schematic as attached for the concept of 12V-12V isolated power supply, could you pls help to see if there's something wrong with it? I'm very care about the efficiency, when I_Load=~2mA, what's the efficiency will be like? Or can you tell me how to calculate?

Thanks a lot again and have a nice weekend in advance!

• In reply to Manuel Chavez:

Dear Manuel,

Thank you for sharing the datasheet for 750317829, but I'd like to know why I cannot search it in Wurth website? Does it already End of Life?

Thanks a lot! Have a nice weekend!

• In reply to JIAN JIN:

Hi Jian Jin,

You're welcome!

1. Unencrypted models for SN650x devices are being made available publicly, so I will link them to the bottom of this response.
2. Transformer 750315240 is rated for 5V operation with SN6505A, however, its V-t product is high enough to support 12V operation with SN6501 or SN6505B.
3. In the schematic you've shared, the topology is good, but the FETs and Schottky diodes are 20V-rated parts, and both will need to be 30V-rated parts or greater due to the push-pull operation of SN650x devices. This is explained in Section 8.3.1 of the SN6501 datasheet.

At low loads, transformers will be inefficient, although the exact amount varies from part to part due to their parameters and manufacturing tolerances. At loads around 2mA, efficiency may be anywhere between 3% - 30%.

Transformer 750317829 is a custom part available from Wurth by request, that is why it is not seen in their online catalog. It is active and samples are available for order.

Please let me know if you have further questions.

Thank you!
Manuel Chavez

SN6501_TRANS.lib

SN6505A_TRANS.lib

SN6505B_TRANS.lib

• In reply to Manuel Chavez:

Dear Manuel,

Thanks a lot for your so detail information!

Wow, it seems the efficiency is too low when the load is light. Would you pls give us some recommend about the isolated power supply (12V-12V) concept by low power loss and low current consumption?

Thanks a lot!

• In reply to JIAN JIN:

Hi Jian Jin,

You're welcome! Yes, at low-current loads transformers are inefficient due to parasitics like leakage and magnetizing currents, which.remain constant in magnitude as load currents increase.

Exact efficiency in these systems varies from transformer to transformer; below is a measured efficiency curve from PMP30440 using 750317829:

I realize the efficiency at 2mA loads is not shown, but is output current in your system expected to be ~2mA continuously?

Higher efficiencies can be achieved using different transformers, so please let me know if you are willing to test alternate transformers in your design, and I can recommend 2 - 3 part numbers that might yield improved efficiency values if this is what you are looking for.

Thank you,
Manuel Chavez