Project insight

Samsung SDI BESS Sourcing: An Office Admin's Honest FAQ

A first-time buyer's FAQ on energy storage system sourcing—covering Samsung SDI, US BESS questions, spec guide shortcuts, and when a lithium battery backup simply isn't worth it.

After our third unplanned outage in 2024—each one landing right when the building's morning shift was logging in—our COO asked me to "look into" backup power. I manage facility vendor relationships for a 260-person company. I do not have an engineering degree. Suddenly I was reading datasheets about lithium iron phosphate cells and something called UL 9540, and pretending I understood.

Here's what I wish someone had told me instead of the sales deck. Not an engineer's guide—just a FAQ for the rest of us.

Q1: When is an energy storage system the wrong choice?

If your outages are brief voltage dips (under 10 seconds, a couple of times a month), a surge protector and a properly sized UPS usually handle it—for a fraction of the cost. A lithium battery energy storage system earns its place when outages run for hours and downtime costs real money. We built a simple model: last year's total outage hours × hourly labor cost + emergency electrician call-outs. If that number comes in under $50K, honestly, I'd wait. The math has to earn the capital request, not the other way around.

Q2: What actually matters when evaluating an energy storage system manufacturer?

From the outside, it looks like you're comparing battery specs. The reality is you're comparing warranty structure, integration experience, and who picks up the phone when a site goes down at 2 a.m.

I made the mistake of optimizing for cycles and energy density first. What I should have led with:

  • Cell-level vs. system-level warranty—and what triggers each
  • Who commissions the system on-site (the manufacturer, or a sub?)
  • How RMA works domestically (this one bit us—turns out the quote didn't include decommissioning labor)

If a manufacturer can't answer those three in writing, that tells you something.

Q3: What's the difference between OEM and private-label BESS?

Short version: OEM means the manufacturer builds hardware to your spec and you put your name on it. Private label is closer to white-labeling—you pick from their standard catalog and only the branding changes. For a 260-person company with no engineering team, true OEM was more than we could actually manage. We ended up evaluating OEM-adjacent arrangements where the manufacturer handled integration but allowed some spec flexibility.

If you ask me, if you don't have engineering bandwidth to write a real spec, don't pretend you do. Stick with a manufacturer's catalog product and negotiate on service, not on custom cells.

Q4: Who is Samsung SDI, and do they actually do BESS in the US?

Samsung SDI is a Tier-1 lithium battery manufacturer—one of the global cell producers you'll see on shortlists when an energy storage system manufacturer is being selected for a project. In most BESS deployments, they operate as the cell supplier underneath an integrator or branded system. That layered structure matters more than people realize.

On the US BESS question: there's activity here, but I'd push on what "US" means in a specific proposal. Is it US-based assembly? US-based service? Or just a US sales entity pointing at an overseas factory? Those are three very different answers for warranty and dispatch. I started asking "which legal entity signs the service contract" and it changed how I read every bid.

Q5: What's the shortcut version of an energy storage system specification guide?

Here's what I'd put on page one if I were writing an energy storage system specification guide for non-engineers:

  1. Use case first—grid-tied backup, peak shaving, or off-grid? Ours was all three, which is why our first scope dragged.
  2. Capacity (kWh) vs. power (kW)—vocabulary matters. Vendors will use these interchangeably if you let them.
  3. Cycle life and how it's counted
  4. Certifications—UL 9540, UL 9540A, NFPA 855. If a bid doesn't list these, it goes in the reject pile.
  5. Thermal management and fire suppression approach

Number four is the real filter. Everything downstream depends on it.

Q6: Why does searching "Samsung SDI Apple battery supplier" dominate the results?

There were reports around a decade ago that Samsung SDI supplied battery cells for some Apple devices. That's it. And I want to be careful here—public reporting on any specific customer relationship is not something I can verify, and definitely not something I'd hang a procurement decision on.

The broader lesson for me: don't lean on borrowed brand names to justify a vendor pick. If a sales rep leads with "we supply [famous company]," that's a deflection from the actual question—which is whether they've done work like yours, in your region, with traceable references. Ask for three comparable installs and call them. It's uncomfortable and it takes an hour and it is worth every minute.

Q7: When would you tell someone to walk away from BESS entirely?

Direct answer, because I wish someone had given me one:

  • Peak demand under 50 kW with no reliability-critical load
  • Site can't meet setback or ventilation requirements
  • No one can sign a 10-year O&M commitment
  • Load data is thin or estimated

If two of those apply, BESS probably isn't the right move. Not because the product is flawed—because the fit isn't there. I told my COO exactly that about a smaller site we were considering, and it made the recommendation for our main facility land harder, not softer.

Q8: What would you do differently on the next sourcing round?

Looking back, I should have insisted on two things in the first RFP instead of "following up later": (1) a named scope owner for commissioning and decommissioning, and (2) a 10-year total cost of ownership model, not just sticker price. We spent six weeks chasing those answers mid-process (ugh, again)—it slowed the whole timeline.

If I could redo it, I'd put both questions in the very first screening email. Whoever answers clearly moves forward. Whoever hedges doesn't. That one filter would have saved us most of the drama.