Ceramic LED Absolute Maximum Ratings and Current Derating

0. If you are seeing a specific symptom, start here

This page is written to be read end to end. If you are here because something is already wrong, start below instead. The table maps current derating symptoms to the three causes that account for most of them in the field, and to the one measurement that separates them. If the measurement contradicts the table, the cause is outside the range this page covers.

Two things to do before you continue production

• Lower the drive current or the duty cycle before anything else. Almost every symptom below is current- or heat-driven, and reducing both is reversible within minutes.
• Measure before you swap parts. A forward-voltage measurement across the assembly, taken with the board powered, separates a driver problem from a LED problem in under five minutes and costs nothing.

When to contact us

Contact us when the measurement contradicts the table, when the same batch shows the same symptom in more than one assembly, or when you need a failure analysis to close a customer case. Bring the bin label, the forward voltage at operating current, and the ambient temperature at the moment the symptom appeared. Send the details to our engineering team.

The chapters below explain the underlying physics and specification in full.

1. What an absolute maximum rating is, and what it is not

An absolute maximum rating is a limit the device must not exceed under any circumstance, whether the cause is a power-supply fault, a wiring error, a mis-set driver or an over-temperature ambient. The ceramic handbook states the position plainly: stress beyond the ratings may cause permanent damage to the device, and exposure to absolute maximum conditions for extended periods may affect device reliability.

Two consequences follow, and they are the reason this page exists. First, the rating is a ceiling, not an operating point. A part whose ceiling is 1,600 mA does not belong at 1,600 mA in production; it belongs well below it, at whatever value keeps the junction inside its limit. Second, the ratings are published per part number and, within several series, per die size, so the number on a family page is a range and not a single figure. Both points are where most datasheet misreadings start.

This page works through the ten consolidated rating parameters in the order a designer meets them: forward current, peak forward current, power dissipation, reverse voltage, junction temperature, thermal resistance, operating and storage temperature, reflow temperature and ESD withstand. The current figures quoted here are the ones published on the individual package pages, which are the figures a quotation and a driver selection have to match.

If you need the underlying principle of why current and temperature trade against each other, the ceramic package technical overview covers it. If you are sizing a heat sink, the thermal management design guide is the relevant page. What follows is specifically the rating set and how to apply it.

2. Drive current: the published ceiling is set by die count, not by package name

Every high-power ceramic LED on this site publishes three drive figures: a standard continuous current, a higher continuous current for variants with adequate thermal management, and a pulsed figure at a 10% duty cycle. These are the numbers to design to, and they are published on the individual package pages rather than in a single consolidated table. That is deliberate: the ceiling moves with the die count and the drive voltage fitted inside a given package, so a single package-level number would be wrong for most of the parts inside that package.

Drive current published on the official package pages

The 7070 package is the one that most often gets overdriven, because it is physically the largest and looks like it should take proportionally more current. The published ceiling is 1,600 mA continuous and 2,000 mA pulsed, and the multi-die versions reach that figure only with the series configuration called for on the page — 6 V, 9 V or 12 V, driven at 350 to 1,600 mA. The 7070 UVA parts are lower again, at 500, 1,000 or 1,400 mA continuous. A 7070 driven at several amps will fail regardless of how good the thermal path looks.

Within one package the applicable figure is set by the die count and by whether the die is wired in series or parallel. A 5050 fitted with a single die and a 5050 fitted with four dies in series are the same package and different ratings. This is the single most common way to overdrive these devices: rating the part by its package name rather than by the die actually fitted.

3. Continuous current, peak current and power dissipation are three separate limits

The handbook lists a continuous forward current, a peak forward current and a power dissipation for the main packages. These are independent limits and a design has to satisfy all three at once. Confusing the continuous and peak figures is the most expensive misreading in this set, because the peak figure is roughly twice the continuous one and looks like usable headroom.

Continuous current, peak current and power dissipation for the two main package families

The handbook’s own note on the peak rating is the operative sentence: IFP is a pulsed rating, and it must not be exceeded by the continuous forward current for steady-state operation. Thermal design has to keep the junction temperature inside the confirmed limit regardless. In practice this means the peak rating only appears in a PWM dimming scheme where the pulse width is short and the average current stays at or below the continuous figure, and where the driver has a defined maximum duty rather than an arbitrary one.

The 5050 continuous current is given as two values, 350 mA and 700 mA, because the package is published in both current classes. The 5050 power dissipation is given as a range, 3 to 12 W, and the handbook ties that range to chip count: 1 to 4 chips. A design that reads 12 W as a generic figure for any 5050 part will be wrong on a single-chip 5050, whose dissipation rating is at the bottom of that range.

4. Reverse voltage and why it matters in a multi-channel design

The reverse voltage rating for all ceramic series is 5 V. That is a low number relative to the currents these devices carry, and it has one practical consequence: a ceramic LED does not tolerate being reverse-biased at all, and the protection network has to be sized for the 5 V limit rather than for the forward voltage.

In a multi-channel colour design, where several dies sit in one package and channels are driven independently, the risk is a channel being driven while another is off, or a transistor failure in the driver applying a reverse bias. Both paths exceed 5 V easily. The rating is also the reason the ceramic series are handled with the same ESD discipline as the rest of the range, discussed in section 9 below.

5. Junction temperature: two limits, and the one that is lower is the one to design to

The handbook sets junction temperature limits in two groups. The 3535 packages and the multi-colour packages are rated to Tj 150 °C, described as a series design limit. The 5050, 7070 and 1860 packages are rated to Tj 125 °C, and this one is marked as both the official rating and the recommended limit.

The distinction matters more than the numbers do. A 150 °C rating on the 3535 is a design limit of that series, so it is available if the thermal design genuinely holds the junction there. The 125 °C figure is stated as a recommended limit as well as an official rating, which means a 5050 design that runs at 128 °C is outside the recommendation even though an absolute maximum reading might appear to permit it. For the large multi-die packages, the 125 °C figure is the number to design against.

The same caution applies to the ambient. The operating temperature range, Topr, is specified as −30 to +60 °C and is an absolute maximum rating in its own right, which means +60 °C is the highest ambient at which the rating statements hold. It is not a recommended operating ambient, and it leaves no margin for a junction rise. Storage is wider at −40 to +80 °C, but storage temperature is a passive rating: it says what the device survives on a shelf, not what it can be driven at.

6. Thermal resistance: the one published figure, and why it is not enough on its own

A junction-to-board thermal resistance is published for the 5050 package at 8 °C/W, sourced from the official 5050 rating. This is the only thermal resistance value given in the consolidated rating table, and it is worth being precise about what it does and does not tell you.

Consolidated rating parameters that are not forward current, with their scope

Rth j-b of 8 °C/W describes the path from the junction to the board, measured on the package. It deliberately excludes everything above the board: the PCB copper, the thermal interface material, the heat sink and the ambient air. A design that divides its allowable temperature rise by 8 and stops has designed the package, not the luminaire. The full budget has to be closed across the whole path, which is the method set out in the thermal management design guide.

That is also why the current ceilings in section 2 are not simply a function of package size. The 7070 carries 1,600 mA continuous while the 5050 tops out at 700 mA, and both reach their figures only with the die arrangement and drive voltage their page specifies; the package outline on its own carries no thermal information.

7. Reflow and ESD: the two handling ratings

Manual soldering temperature is specified as 260 ± 10 °C for 3 to 5 seconds. This is a process limit rather than an electrical one, and it is the number that has to match the reflow profile actually used in production. The tolerance band and the dwell time are both part of the rating: a 260 °C peak held longer than 5 s is outside it, and a peak above 270 °C is outside it regardless of duration.

ESD withstand is 2,000 V Human Body Model across the ceramic series. The handbook is explicit that the device is ESD-sensitive and calls for standard precautions during assembly: a grounded wrist strap, a dissipative work surface and ionised air. Combined with the 5 V reverse rating from section 4, this sets the handling regime for the whole ceramic range, and it is the same regime the wider QUEENDOM range is assembled under.

8. Applying the ratings: a selection sequence

The ratings are most useful as a sequence rather than as a table to memorise. The following order follows the constraints from tightest to loosest, and each step can disqualify a candidate before the next one is evaluated. The table below states each step as the specific rating it is checked against, so a candidate package can be worked through without hunting back through the section.

Selection sequence with the rating each step is checked against

• Fix the ambient. Topr is the ceiling and it is an absolute maximum, so an ambient above +60 °C is outside the published ratings regardless of package.
• Set the junction target. Design to Tj 125 °C for the 5050, 7070 and 1860 packages, since that figure is stated as both official rating and recommended limit. For 3535 and multi-colour, Tj 150 °C is the series design limit.
• Close the thermal budget. Allocate the available temperature rise across junction-to-board, board-to-ambient, and the interface, then work out what current that supports. Package Rth j-b is only one term.
• Read the rating for the exact part. Within a package the forward-current rating is set by the die and the drive voltage, so the published class is the boundary, not the answer.
• Check dissipation and reverse voltage. For the 5050, confirm the chip count against the 3 – 12 W dissipation range, and confirm the driver cannot reverse-bias a channel against the 5 V limit.
• Only then use IFP. Peak current is a pulsed allowance for a defined duty and an average current at or below the continuous figure. It is not an operating point.

One practical consequence deserves emphasis. Because the ratings are bounded per current class and per die arrangement, a substitution that looks harmless on the datasheet can be a rating violation. Moving a 7070 from the 350 mA class to the 1,600 mA class is not a drop-in replacement in the electrical sense even though the footprint matches, and it is only available in the drive voltage configuration that page lists. Check the rating, not the footprint.

9. Where the rating data comes from, and what it does not include

Every current figure on this page is taken from the package pages on this site, not from a handbook. That distinction matters in practice: the drive current published for a package is the figure a quotation, a driver selection and a customer approval have to match, and it is stated per current class, with the die arrangement and drive voltage spelled out. The consolidated rating table in the ceramic handbook is useful for the temperature, voltage, dissipation and ESD parameters, all of which are stated there as consolidated from the official pages.

One warning about cross-checking. Where a handbook series table and a package page give different forward-current figures, the package page is the one to design to and the one to quote from. Current ceilings are the figures most often revised as die arrangements change, and a figure carried in a secondary document can outlive the product page it came from. Two limits of this data are worth stating plainly. The first is that the rating table does not include a measured thermal-resistance value for every package; only the 5050 junction-to-board figure is published, so other packages have to be characterised at the board level rather than read off a number. The second is that the ratings are absolute maxima. They are not a quality grade, not a life expectancy, and not a guarantee of performance at any particular point inside them.

For lumen maintenance, reliability testing and failure mechanisms, the relevant pages are the LM-80 testing and lifetime estimation guide and the industrial reliability white paper. Neither substitutes for the rating set above: a device operated well inside its ratings still needs a lifetime model, and a lifetime model does not license operating outside the ratings.

10. Frequently asked questions

Is the absolute maximum forward current a safe operating current?

No. The handbook is careful about the wording: stress beyond the ratings may cause permanent damage, and exposure to absolute maximum conditions for extended periods may affect device reliability. A part whose ceiling is 1,600 mA can be damaged by 1,600 mA sustained. The rating tells you where the failure boundary is; the thermal budget tells you where to actually run.

Why does the same package have a range of forward-current ratings?

Because the rating is set by the die and the drive voltage fitted, not by the package. The 7070 is published as 350 mA standard and up to 1,600 mA for the high-power variants, with the 6 V, 9 V and 12 V series configurations driven at 350 to 1,600 mA. When you order a specific part number, the die is fixed and the rating narrows to a single figure. Before that, work from the range.

The 5050 continuous current is quoted as 350 / 700 mA. Which one applies to my part?

Both are valid; they are different current classes of the same package. The 5050 is published in both, and the 3 to 12 W dissipation range is tied to chip count from 1 to 4. The reliable route is to take the value from the specific part number’s series table rather than choose between the two.

Can I use IFP as my normal drive current?

No. IFP is a pulsed rating, valid at 10% duty, and the handbook states explicitly that the continuous forward current must not be exceeded for steady-state operation. At 3535 the peak allowance is 1,400 mA against a 700 mA continuous rating, so using the peak figure as a continuous set point doubles the dissipation.

Is Tj 150 °C usable for the 5050?

The handbook rates junction temperature in two groups: Tj 150 °C for the 3535 and multi-colour packages, described as a series design limit, and Tj 125 °C for the 5050, 7070 and 1860 packages, stated as both the official rating and the recommended limit. The 5050 belongs to the 125 °C group, and because that figure is also the recommendation, a design is expected to stay at or below it.

Can the junction reach 125 °C when the ambient is below that?

Yes. Junction temperature is the die temperature, not the ambient, and it is higher than the ambient by the temperature rise across the thermal path at the current being used. That rise is why the 5050 and 7070 are limited to 125 °C while the operating range extends to +60 °C ambient: the remaining budget has to cover the path from junction to ambient, not just the ambient itself.

Does the 2,000 V ESD rating mean the device is robust to static discharge?

It means the device withstands 2,000 V in the Human Body Model, and the handbook states at the same time that the device is ESD-sensitive and requires standard precautions: a grounded wrist strap, a dissipative work surface and ionised air during assembly. The rating is a qualification figure, not an exemption from handling discipline.

Which parameters should I ask for when the datasheet is not available?

For any ceramic part, the forward-current rating for the exact die, the continuous and peak forward current, the power dissipation, the junction temperature limit for its group, and the thermal resistance if the package is not a 5050. Those five cover the limits that a design can violate in normal operation. The full set, including reverse voltage, temperature ranges, reflow and ESD, is in the consolidated rating table in the handbook.