Replacement to 1uH - STPI1003-1R0M-HF

In the BOM of the Reference Design, I saw that L507 is STPI1003-1R0M-HF, but couldn’t find this part in market.
Do you have alternate part for this?

Topsun has the part:

I’m sorry, but I can’t get the PDF of this product - I get following error when clicking on the PDF icon:
You don’t have permission to access /Public/ on this server.

Also got error when tried to get into high-current inductors:

Kind reminder - this is a reference design, and should use common items!

Did you contact TopSun to ask about the datasheet?

Anyway: it’s not like 10A+ 1uH SMD inductors are particlarly hard to find. Bourns, Murata, Wurth, and a bunch of other vendors have them. I imagine you’d be fine with anything like this:
Main question is height versus resistance loss; it’s wire wound around ferrite, and all vendors will have approximately the same performance/size trade-off.

Try this?

Thanks, I’ll look for the alternative.

Please understand the situation, which I assume is very common.

We were asked to take NV Carrier-Board design, and perform minimal changes to suit our specific usage.
There are two main reasons for that: Minimize design-time (“cost”), and minimize design issues.

For that, it was great to provide the BOM of the design, with 1-2 alternatives.

There’s no need for revolutions - only evolution, which, in other words I call ‘Continuous Improvement’.
I suggest NV to see how they improve from year to year.

Thanks again

IMHO there will be more difficult find this L507 defined in BOM, than find new one from other supplier.

Please bear in mind, you want item, what is very hard to buy. It´s not very straight way of development and in future can cause lot of problems.

I think snarky is right.

“you want item, what is very hard to buy. It´s not very straight way of development and in future can cause lot of problems.”

No, I don’t want item which is hard to find.
I want that a reference design will have only items which are not hard to find, not NRND (Not Recommended for New Design) etc.

For the record, I think @ishay19bch is correct in that reference designs should use items that are easily commercially available from common distributors.

I think the problem with this reference design is that it has to serve two purposes:

  1. It has to be a reasonable-low-cost implementation to use for the developer kits, which are sold in some volume.
  2. It has to also be a basis on which other OEMs can build solutions.

Making one thing do two things is always going to end up with compromises. So, how did NVIDIA end up selecting a part that’s not available from the usual suspects (at least on European/US mainland?)
Chances are, NVIDIA and their OEM partners for desktop GPUs (the MSIs and Asus-es and EVGAs of the world) already have a component supply chain, established with good volume discount and supply deals, with particular manufacturers. Inside NVIDIA, it’s much easier, and cheaper, to use a known-good part from a previous supplier in a previous product, than to go out and try to qualify new suppliers, who generally are more expensive (because of the lower volume purchase, and perhaps larger sales forcce / distribution network.)
So, the engineer within NVIDIA, sourcing a part, obviously followed the path of least resistance given the situation they were in. The problem is that NVIDIA has set themselves up as a different kind of company than what they are trying to be to support the current Jetson ecosystem.

Can NVIDIA change who they are? Maybe, but that doesn’t seem all that profitable to me.
Can NVIDIA change how they do support a bit, and be more accepting of the need to actually vary the reference designs? Quite likely.
Can NVIDIA support read the actual posts that are posted in more detail, and work a little harder to understand the specific needs of the person asking for the support? Yes.

Meanwhile, try Bourne SRP6050CA-1R0M for the win :-)