TLP250 replacement: the HCPL-3120, TLP352 and gate-driver optocoupler options compared
The Toshiba TLP250 gate-driver optocoupler is obsolete, and every still-made cross drives more current and adds UVLO.
The Toshiba TLP250 isolated gate-driver optocoupler is obsolete, and every still-made cross a search returns drives more current and adds undervoltage lockout. Same 8-pin DIP, different gate drive: the cross-vendor HCPL-3120 is the closest pin match, Toshiba's own TLP352 changes pin 7, and the popular TLP350 cross is itself obsolete. The sections below work through those differences, part by part.
Status // The part in question
Lifecycle status
The TLP250 is obsolete. The TLP250 and TLP250F read Obsolete at DigiKey and Mouser and are no longer manufactured. No primary Toshiba discontinuation notice number was located; the distributor lifecycle flag is the source. Confirm the exact orderable you use, the TLP250H is a separate, later device (2015 datasheet) listed active, a different part, not the same silicon.
Buying remaining stock only defers requalification, it does not remove it. Plan a last-time-buy alongside the design-in of one named cross, and budget the requalification either way.
Current stock and any last-time-buy window are dated data for the email, not this page.
Candidates // Why these
The candidate set
The parts engineers actually cross-shop for the TLP250, each with its own fit-catch to validate in your design:
Catch: pin 7 and the electrical shift. 2.5A peak, UVLO, and a fast 200ns delay. Pin 7 is a no-connect where the TLP250 has a second output, check any trace tied to pin 7. Re-check the gate resistor, timing, and that the rail clears the UVLO threshold (~12.5V typ).
Catch: electrical, not pinout. The closest match to the TLP250: same second output on pin 7, pin 5 the output return either way. Active and widely stocked. What differs is 2.0-2.5A peak and UVLO (~12.3V typ), re-tune the gate drive and confirm the VEE return.
Catch: it is also obsolete. The classic 2.5A cross a tool returns for the TLP250, but it reads Obsolete at DigiKey and Mouser, crossing to it moves one end-of-life part to another. Toshiba's current part is the TLP352.
Catch: it drives less current. Active, but 0.6A peak (below the TLP250's 1.5A) with no UVLO. Built for smaller IGBT gates, it fits a smaller gate, not a like-for-like replacement for a 1.5A design.
Catch: technology and footprint change. These fill the isolated-gate-drive role without an LED. TI markets the UCC23513 (SO-6) as a pin-to-pin upgrade for 6-pin opto drivers, but against a TLP250 8-pin DIP it is a different footprint; the Si823x is SOIC and not pin-compatible. Consider them when the board is opening for a redesign anyway.
Pinout // Pin 7 is the tell
Same package, not the same gate drive
Three isolated gate-driver optocouplers share the exact 8-pin DIP outline, so a cross-reference on "8-pin gate-driver optocoupler" reads them as interchangeable. The non-obvious result: the cross-vendor HCPL-3120 lands pin-for-pin on the TLP250, while Toshiba's own TLP352 changes pin 7 from a second output to a no-connect, the same-vendor part needs a trace check the cross-vendor part does not.
8-pin gate-driver optocoupler DIP · pins 1 to 4, 6, 8 shared · pin 5 and pin 7 differ
Pin color,green same function · amber label differs (Vee and GND both the output return) · red output pin becomes a no-connect
Pinout is necessary, not sufficient. Suggested candidates to validate in your own design, not drop-in equivalents. Form, fit and function is a property of your system.
Parametrics // Side by side
Comparison table
| TLP250 | TLP350 | TLP351 | TLP352 | HCPL-3120 | |
|---|---|---|---|---|---|
| Lifecycle | Obsolete | Obsolete | Active | Active | Active |
| Peak I | ±1.5A | ±2.5A | ±0.6A | ±2.5A | 2.0-2.5A |
| Supply V | 10-35 | 15-30 | 10-30 | 15-30 | 15-30 |
| UVLO | None | Yes | None | Yes | Yes |
| Isolation | 2500 Vrms | 3750 | 3750 | 3750 | 3750 |
| Prop delay | 0.5µs | 500ns | 700ns | 200ns | 500ns |
| Pin 7 | Vo (2nd out) | NC | NC | NC | Vo (2nd out) |
| Package | 8-DIP | 8-DIP | 8-DIP | 8-DIP | 8-DIP |
The live crosses sit at higher drive current and 3750Vrms with UVLO, where the TLP250 is 1.5A and 2500Vrms with no UVLO. On pin 7 the TLP250 carries a second output; the Toshiba successors move it to a no-connect, while the Broadcom HCPL-3120 keeps it. Lifecycle flags are distributor/vendor-reported, not independently audited.
The gap // What a parametric search hides
What the cross-reference misses
A parametric search for "2.5A IGBT gate-drive optocoupler, 8-DIP, 3750Vrms" returns the TLP350, TLP352 and HCPL-3120 as matches for the TLP250. What it does not surface: that the TLP250 is a 1.5A, 2500Vrms, no-UVLO part while those crosses are higher-current, 3750Vrms, UVLO parts; that the closest-looking cross, the TLP350, is itself obsolete; and that the TLP351 in the same result list drives less current, not more.
The footprint match hides the drive-current step, the added UVLO, the pin 7 change, and the second part's own lifecycle, so the obvious cross can change gate drive strength and startup behavior, or design in a second end-of-life.
Decision // Fit to your design
How to choose
No single winner; a fit to your design and program. Both registers:
Checklist // Before you commit
Before you buy
Match the peak output current to your gate charge and switching-speed target, not just the package. The TLP250 is 1.5A; the TLP352 and HCPL-3120 are 2.5A-class; the TLP351 is 0.6A.
Check for UVLO and confirm your gate-drive supply clears its threshold. The TLP250 has none; the TLP352, TLP350 and HCPL-3120 do.
Confirm the pinout against your board, pin 7 especially. It is a second output on the TLP250 and HCPL-3120, a no-connect on the TLP350, TLP351 and TLP352.
Confirm the candidate's own lifecycle status (active, not obsolete) before standardizing, the TLP350 fails this.
Re-verify creepage and clearance if the isolation rating changes (2500 to 3750 Vrms) and the package body or pin spacing differs.
Requalify in your system. A matching footprint does not qualify the part.
FAQ // Frequently asked
Frequently asked questions
What is the replacement for the TLP250?
There is no single drop-in. The active same-vendor path is the TLP352 (the TLP350 is also obsolete); the common cross-vendor part is the Broadcom HCPL-3120, the closest pin match. Both are 2.5A-class parts with UVLO, so they change the gate drive versus the 1.5A TLP250 and need validation, not a straight swap.
Is the TLP350 a drop-in for the TLP250?
It is the usual cross, at 2.5A, but it is itself obsolete at distribution, so it trades one end-of-life part for another. Look at the TLP352 instead of crossing the TLP250 to it.
Difference between the TLP250 and TLP350?
Peak output current (1.5A vs 2.5A), UVLO (the TLP250 has none, the TLP350 does), isolation (2500Vrms vs 3750Vrms), and pin 7 (a second output on the TLP250, a no-connect on the TLP350). They are not electrically identical.
Can the HCPL-3120 replace the TLP250?
It is a candidate to validate: active, 2.0 to 2.5A peak, UVLO, 3750Vrms, and pin-for-pin with the TLP250 on assignment (pin 5 is the output return either way). Re-tune the gate drive for the higher current and confirm the VEE return; it is not a guaranteed drop-in.
Is the TLP351 the same as the TLP352?
No. They share a pinout, but the TLP351 is a 0.6A part with no UVLO, and the TLP352 is a 2.5A part with UVLO and a faster 200ns delay. The TLP352 is the closer match to a 1.5A TLP250 design.
Signals // Watch List
The Watch List
Two things to carry forward:
The obvious cross is also end-of-life
The TLP350, the part most often returned as the TLP250 upgrade, is itself obsolete at distribution (DigiKey, Mouser); Toshiba's current part is the TLP352. Confirm a candidate's own lifecycle status before crossing to it.
Gate-driver optocouplers carry an LED wear-out path
The isolation LED ages over the part's life, one reason non-optical isolated gate drivers exist. TI's UCC23513 cites a >50-year isolation-barrier life. A like-for-like opto keeps the LED wear mechanism; a technology change removes it, at the cost of a footprint change.
Sources // Public, primary
Sources
- Toshiba TLP352 datasheet (PDF) · active same-vendor gate driver
- Toshiba TLP351 & TLP350 datasheets (PDF) · TLP350 obsolete at distribution
- Toshiba TLP250 datasheet (PDF) · distributor lifecycle flag: Obsolete
- Broadcom HCPL-3120 datasheet (PDF) · 2.0-2.5A gate driver
- TI UCC23513 · product page · non-optical isolated gate driver
- DigiKey / Mouser TLP250 & TLP350 lifecycle and stock pages
Editorially independent and from public sources only; lifecycle, pin and rating data is vendor-reported, not independently audited. Alternates are candidates to validate in your own design. Confirm every detail with the manufacturer before committing.