This checklist is for anyone who's ever stared at a BOM and thought, 'A capacitor is a capacitor, right?' — then regretted it. I've been handling component procurement for 12 years, and for the first 5, I was the guy making those $890 redo mistakes.
(Should mention: I now maintain our team's 18-point pre-order checklist. It's caught 47 potential errors in the past 18 months, totaling roughly $14,000 in avoided waste.)
When to Use This Checklist
This is for engineers, prototype buyers, and small-team procurement folks who order medium-volume runs (100 to 10,000 units) of passive components like ceramic capacitors, tantalum capacitors, relays, and connectors. It's specifically tuned for scenarios where you're sourcing from a distributor like Kemet or evaluating substitutes against established lines like Murata or the Kemet T520 series of tantalum capacitors.
If you're ordering single-unit prototypes or multi-million-unit contract manufacturing, parts of this still apply, but your process is (hopefully) more automated.
Here are the 4 steps I follow now, every single time, after learning the hard way.
Step 1: Verify the Specs, Not Just the Part Number
It's tempting to think you can just match part numbers. 'Kemet T520'? Same thing, right? The 'you just need the part number' advice ignores subtle but critical version changes.
In March 2022, I ordered 2,000 Kemet T520 series tantalum capacitors. The part number matched our current BOM exactly. But I'd failed to check the revision letter — the new revision had a different footprint and a slightly lower ripple current rating. They physically fit the board (barely), but the thermal performance was marginal. That $1,200 order ended up in a re-spin because of intermittent failures in the test chamber. Cost to fix: $3,800 plus a 2-week delay.
Your checklist item:
- Don't trust the part number alone — check the specific datasheet for the revision code (e.g., Kemet T520B476M006ATE045 vs T520B476M006ATE045R).
- Keep a copy of the exact datasheet you spec'd against (as of the date you approved it). Vendors occasionally update datasheets without changing the core PN.
- For ceramic capacitors, verify voltage rating vs. the DC bias derating. A 50V-rated MLCC might only give you 15% capacitance at 40V. Ask me how I know about the 200-ceramic-capacitor mishap in Q3 2023. (Ugh.)
Step 2: Check the Source's Authenticity — This is the Big One
This is where 80% of my mistakes happened earlier in my career. I said 'authorized distributor parts.' They heard 'any parts on the open market that say Kemet on them.' Result: mismatched date codes, suspect markings, and one batch of capacitors that failed after 100 hours.
I once saved $220 by buying 'Kemet ceramic capacitors' from a non-franchised broker through a marketplace. They arrived looking pristine. We tested 50 out of 1,500 units — all good. We populated the rest. 47% failed within 6 months. The cost of rework? $4,500. The $220 savings became a $4,720 net loss. (Thankfully, we caught it during warranty before the product shipped to the end customer.)
Your checklist item:
- Purchase from authorized distributors for critical components (especially Kemet tantalum or high-reliability ceramic capacitors). As of January 2025, Kemet's authorized distributor list is on their official site — verify from there.
- For non-critical inventory or bridge buys, request original manufacturer documentation. A PDF is not proof. Request shipment directly from an authorized warehouse, not a third party.
- Beware of 'Kemet T520' listings that are significantly below the market average price (circa Q4 2024 data). If it's $0.35 and everyone else is at $0.68, there's a reason.
Step 3: Evaluate Substitutes with Brutal Honesty — Not Just 'Looks Good'
We all face times when the primary vendor has a 20-week lead time. You need a substitute. The 'just find a cross-reference' advice ignores the real-world performance differences.
For example, a client wanted to substitute a proprietary medical device component (something like a Platinum BP5450 connector, though not exactly that) with a more standard switch from another vendor, say a 'switches vs cisco switches' analogy in component terms. They just cross-referenced pinout and voltage. What they didn't verify was the shock and vibration tolerance, which was critical for their wearable monitor (similar to spec'ing parts for a blood pressure cuff). The 'compatible' switch failed lab testing. $2,300 wasted on a short run.
Your checklist item:
- I recommend an alternative Kemet capacitor if you're dealing with standard commercial applications where the form factor matches. But if you're designing for high-vibration environments or medical devices (like the components inside a blood pressure cuff), stick with the original design spec unless you physically test the alternative.
- This solution (the cross-reference method) works for 80% of cases. Here's how to know if you're in the other 20%: if the component is in a feedback loop, a timing circuit, or a safety-critical path, do NOT simply cross-reference.
- For connectors and relays, check the mating cycle rating and insertion force. A cheap alternative might save $0.50 per unit but cause a 3-day production jam because the assembly robot couldn't insert it reliably.
Step 4: Confirm the Full Lead Time & Total Cost
The cheapest price and shortest quoted lead time is almost never the truth. Had 2 hours to decide on a rush order for a client's prototype. Normally I'd get three formal quotes and check stock levels. But with the CEO breathing down my neck, I went with a vendor based on an online 'in stock' flag and a low price. The shipment arrived late (the 'in stock' was a lie), and 15% were visibly damaged. Net loss on that single order: $890 for the parts plus $450 in expedited freight for the real order. Total: $1,340 down the drain.
In hindsight, I should have pushed back on the timeline. But with the constraint, I did the best I could with available information. Now I have a rule.
Your checklist item:
- Always request a written lead time quote for Kemet products specifically. As of January 2025, standard MLCCs are ~8-12 weeks, but tantalums can stretch to 16-20. (Oh, and I should add that these fluctuate wildly, so verify at the time of your quote.)
- Total cost includes: base price + shipping + handling + rush fees (if any) + the cost of inventory that sits idle for 4 extra weeks (that's carrying cost, which is real).
- Don't be afraid to ask about alternates. If the Kemet T520 is on a long lead, ask your distributor if a Kemet T521 series (or a comparable Murata part) is available. That swap saved my team $2,100 in 2023 when we were desperate for a specific MLCC (circa October 2023).
Common Errors & Final Advice
If you remember nothing else, remember these three:
- The Footprint Trap: It looks the same but it's not. Measure the actual tape-and-reel dimensions. I caught a 1-mm pitch difference on a reel of 5,000 relays last year just by checking the physical drawing. That saved a production halt. (Note to self: always keep a caliper on the desk.)
- The Data Code Fallacy: I said 'mixed date codes are fine.' The test engineer heard 'do not use for new designs.' Mixed date codes can mean different dielectric batches on ceramic capacitors, leading to performance variance. Establish a maximum variance policy (e.g., no more than 6 months difference).
- The 'It Came From Kemet So It's Good' Assumption. This isn't about Kemet's quality — it's about the supply chain. A genuine Kemet capacitor is excellent. A component that claims to be Kemet but was mishandled, stored outside its humidity specs, or is counterfeit is not. Verification is your responsibility.
I've made these mistakes so you don't have to. Print this checklist (I really should do that with all our team's docs) and keep it next to your BOM review station.
