Repeat Orders — How to Keep Batch Two Like Batch One
A repeat order is where flash memory supply usually goes wrong. The first order is built to a specification; the second is often rebuilt from whatever material is available, and nothing in the paperwork says so. Four things move the odds in your favour: the silicon source and grade written into the order, the controller solution stated rather than implied, a written acceptance test, and a retained reference sample from the batch you approved. Hold all four and the second order is comparable to the first. Skip them and the price you negotiated is close to meaningless, because you are not buying the same product.
What changes between order one and order two
| Variable | What can change | How to control it |
|---|---|---|
| Silicon grade and source | Whatever is on the market that week | Stated in the order, not implied by a description |
| Die package | TSOP, BGA, LGA, COB and eMMC availability all move | Fixed in writing against the approved configuration |
| Controller solution | A different controller changes host behaviour | Named in the order, with the package type |
| Firmware and device ID | Re-programming changes enumeration behaviour | Locked from the approved sample |
| Print method and colour | Colour drift between presses is normal | A reference sample plus a stated colour reference |
| Packaging | Suppliers change when stock runs low | Stated as part of the unit, not the outer carton |
What we hold against a repeat order
A repeat order should start from evidence rather than memory. We keep the approved configuration, the test record for the batch you accepted, and a retained reference sample, so a later run can be checked against the first rather than against a description of it.
- The approved configuration — capacity, interface, silicon source and grade, controller solution
- The test record for the batch you signed off, so a future result is comparable rather than anecdotal
- A retained reference sample from that batch, held so a later order can be verified against it
- Your artwork file and print method, so the marking is reproduced rather than re-created
The acceptance test worth writing down
The most useful part of a repeat order is not a clause, it is a test both sides agree will be run on the next delivery. Write down what is tested, how it is tested, and what result counts as acceptance.
- Sample size — a random sample from the batch, not a unit picked afterwards
- Method — capacity write-and-read-back on the published method, plus a sustained write on the target device
- Criteria — what result is a pass, and what happens when a sample fails
- Timing — tested before payment or on delivery, whichever you intend to enforce
When a repeat order is genuinely a new order
Some changes make the old specification irrelevant. In those cases, argue from the new requirement rather than the old order — quoting the previous configuration back at us will not get you a useful answer.
- The capacity or interface version changed — the silicon and controller selection may both change with it
- New tooling was added, or the artwork was revised after approval
- The product itself changed — a new housing, a new host board, a new firmware
- A different supplier held the order previously: treat the new one as a first order
Send your requirement
We reply to business enquiries within 1 business day. If you are comparing suppliers, send the specification you were quoted — it makes the comparison meaningful.
Request a quoteFrequently asked questions
Is a repeat order always identical?
No. Flash is bought against a market, not against a permanent contract. What can be held steady is the configuration, the test and the documentation — and where a change is unavoidable you should hear it before production rather than on arrival.
How do I know the second batch uses the same silicon?
Ask for the source and grade in writing against the repeat order and keep the reference sample from the first batch. Grade and source cannot be inspected on a delivered card; they are established through disclosure and then verified on capacity.
Do you retain samples of past batches?
Yes. We keep the approved configuration, the test record and a retained sample from the batch you accepted. Give us a repeat quantity and we can check the new run against the earlier one before you commit to it.
What if the material is unavailable for the repeat?
Tell us at quotation stage what would happen. A substitution that goes back to you first is a supply decision; a substitution you discover on arrival is a failure. We would rather ask early than explain late.