Migrating Siemens Masterdrive 6SE7021-0TP70 from EP50 AC/AC

Overview of the Migration Scenario

The 6SE70 SIMOVERT Masterdrives family spans several hardware variants that are not directly interchangeable in the field. Engineers maintaining legacy machines frequently encounter the situation where an installed unit—originally a three-phase AC/AC converter—has to be replaced by a later AC/DC inverter unit, even though the catalog order codes (the 6SE70xx-xxxxx string) look similar at first glance. This application note documents the field-proven migration path from the 6SE7021-0EP50 to the 6SE7021-0TP70, including the DC-bus supply requirement (510–600 V DC), the parameter differences, and the most common PLC identification failure that occurs after the swap.

The reference design uses the SIMOVERT Masterdrives Vector Control firmware and the Compendium manual as the primary documentation baseline. Engineers performing this swap must treat the new unit as a different converter class, not as a "drop-in" replacement of the EP50.

Part Number Identification: 6SE7021-0EP50 vs 6SE7021-0TP70

Both units belong to the 6SE70 SIMOVERT Masterdrives generation, but the last two suffix blocks of the MLFB (Machine-Readable Product Designation) define the converter class, frame, and power stack. The relevant differences are summarized below.

The full SIMOVERT Masterdrives Vector Control documentation—VC332 KOPL AC/AC manual (PDF)—remains the canonical parameter reference for the firmware that runs in these units. When in doubt about a specific parameter, fault code, or terminal assignment, treat that document as the source of truth and verify against the actual drive firmware version stamped on the nameplate of the unit in the cabinet.

Topology Difference: AC/AC Converter vs AC/DC Inverter

The 6SE7021-0EP50 integrates a six-pulse (or twelve-pulse, depending on the frame) line rectifier, a DC link with capacitor bank, and a PWM inverter stage in a single housing. The 380 V AC line is rectified, filtered by the internal DC-link capacitors, and then inverted back to variable-frequency AC for the motor. The end user only sees AC on both sides of the chassis.

The 6SE7021-0TP70, by contrast, exposes the DC link as the line-side interface. There is no internal line rectifier. The unit expects a regulated 510–600 V DC source that already meets the undervoltage/overvoltage window defined by the firmware (typically 360 V DC minimum, 750 V DC maximum, with the nominal operating range centered around 540 V DC for 380 V AC class systems or 600 V DC for 400/415 V AC class systems). The PWM inverter stage then synthesizes the motor-side variable-frequency AC.

The consequences for the cabinet designer are significant:

• A pre-charge circuit must limit the inrush current into the DC-link capacitors of the TP70 when the external DC bus is energized. Without pre-charge, the rectifier feeding the DC bus will see a near-short-circuit condition and trip on overcurrent, or blow the input fuses.
• The DC bus must include a ground-fault monitoring path. A DC link does not self-clear ground faults the way an AC-fed system does, and an undetected ground fault on the DC bus can damage the IGBT modules of the TP70 within milliseconds.
• Braking energy must be handled at the DC bus level (braking chopper + resistor, or a regenerative line module). The TP70 does not have an internal AC-line-side braking path because it has no AC line side.

DC Bus Supply Requirements (510–600 V DC)

The 510–600 V DC requirement is not arbitrary. It corresponds to the rectified peak voltage of a 380–415 V AC three-phase system, minus the typical line-side voltage drop and minus the ripple margin tolerated by the firmware's DC-link monitoring.

Approximate derivation (for verification only, not a substitute for nameplate data):

• Line-to-line RMS voltage: V_LL = 380 V AC (European) or 400–415 V AC (international)
• Rectified average DC voltage: V_DC ≈ 1.35 × V_LL = 1.35 × 380 ≈ 513 V DC
• Rectified peak DC voltage: V_peak = √2 × V_LL = √2 × 380 ≈ 537 V DC
• With ±10 % line tolerance: 1.35 × 418 ≈ 564 V DC average, peak ≈ 591 V DC

The 510–600 V DC window covers the full European 380 V class operating range, including normal line transients. Anything below 510 V DC will trigger an undervoltage fault (F0006 / F0007 family on the OP1S display, depending on firmware); anything above 600 V DC risks tripping the overvoltage threshold of the DC link (F0002 family).

Building a 510–600 V DC Bus from 3-Phase 380 V AC

To reuse the existing 380 V AC feed that the EP50 was connected to, the cabinet must be reworked to insert a DC-bus supply stage between the AC line and the TP70. Three practical options exist:

Option 1: Six-Pulse Diode Rectifier + Pre-Charge + Braking Chopper

The lowest-cost rebuild. A three-phase diode bridge (typically a SEMIKRON, Epcos, or equivalent module rated for at least 1.5× the DC link current) feeds a DC-link capacitor bank through a pre-charge resistor that is bypassed by a contactor after the bus reaches 80 % of nominal. A braking chopper and resistor are mandatory if the motor overhauls the load (vertical axes, hoists, decelerating high-inertia loads).

Option 2: Active Line Module (ALM) with IGBT Rectifier

The preferred solution when the cabinet must remain on a common DC bus with multiple drives, when regenerative braking is required, or when line harmonics must meet IEEE 519 / EN 61000-3-12. The ALM regulates the DC bus to a fixed value (typically 600 V DC for 400 V AC class) and can push energy back into the line. Pre-charge and ground-fault monitoring are integrated. This is the most expensive option but the most robust.

Option 3: Basic Line Module (BLM) with External Pre-Charge

A middle ground. The BLM is a six-pulse diode rectifier with integrated current measurement, but without active line-side switching. It accepts an external pre-charge circuit and a braking chopper must be added if regeneration is possible. The BLM does not meet strict harmonic limits on its own; it is appropriate for installations where harmonic compliance is not a contractual requirement.

For the EP50 → TP70 swap, Option 1 is the most common in brownfield retrofits where the cabinet footprint is fixed. Option 2 is the most common in greenfield cabinets or in retrofits where multiple drives are sharing a common bus.

Parameter Migration Considerations

The firmware baseline for the 6SE70 family is documented in the Vector Control manual. Key parameters that must be reviewed during migration include:

Engineers should dump the parameter set from the EP50 before decommissioning it (using DriveMonitor or SIMOVIS) and then treat that file as a starting reference, not as a direct upload. Most parameters will need to be re-entered or re-derived for the TP70. The peer-link addresses, USS parameter numbers, and PROFIBUS GSD file do not change between the two units, which is the only reason a partial parameter copy is meaningful.

PLC Communication and Identification Failures

The most common field complaint after a TP70 swap is that the supervisory PLC reports the drive as "not identified," "slave failure," or "configuration error," even though the TP70 powers up and shows a healthy OP1S / OP2S display. The root cause is almost always one of three issues.

Cause 1: PROFIBUS GSD Mismatch

The 6SE70 GSD file references a specific PPO type and module order. If the PLC program was written against the GSD for the EP50, and the TP70 ships with a different PPO type or a different module family identifier, the PLC will not establish a cyclic data exchange and will mark the slave as failed even though acyclic communication works. Load the correct GSD for the TP70 firmware version (visible in P060 / P070 firmware code on the OP1S) and re-insert the slave into the PROFIBUS configuration.

Cause 2: USS / Peer-Link Address Conflict

If the drive is on a USS bus (not PROFIBUS), the USS address in P683 / P684 must be unique on the bus. After a swap, the TP70 will default to address 0 or to the last-saved address from the factory, which can collide with another drive on the bus. Change P683 to a free address and power-cycle the drive so the new address takes effect.

Cause 3: PLC Program Hard-Codes the Slave Family

Some legacy PLC programs use SFB / SFC calls (on S7) or explicit MSG instructions (on ControlLogix) that include the drive's order number or family code in the data block. If the PLC program is binary-only and the source code is not available, the integrator must either reverse-engineer the data block to identify the hard-coded fields, or accept that the PLC will not recognize the TP70 and add an external protocol converter (e.g., a Anybus X-gateway) that presents the TP70 to the PLC as if it were the original EP50.

Field Commissioning Procedure

The following procedure is the conservative order of operations for energizing a TP70 for the first time after an EP50 removal. Do not skip steps.

1. Lock out and tag out the upstream AC feed. Verify zero potential on all three phases with a known-good voltage tester.
2. Confirm the new TP70 is mechanically seated and that the DC-link capacitors are discharged (the unit must sit for at least five minutes after power removal before contact).
3. Inspect the DC-bus wiring: positive rail, negative rail, ground. Verify the pre-charge resistor is in series with the contactor, not bypassed.
4. Remove all motor leads from the TP70 output terminals (U2, V2, W2). The motor must be disconnected for the first energization to prevent an uncontrolled start.
5. Energize the pre-charge circuit only. Measure the DC-link voltage with a multimeter at the test points. It should ramp to roughly 80 % of nominal over 2–5 seconds, at which point the pre-charge contactor is allowed to close and the main rectifier takes over.
6. Verify the DC-link voltage is within 510–600 V DC. If it is outside this window, do not proceed. Investigate the line-side rectifier or the ALM setpoint before continuing.
7. Apply control power (24 V DC) to the TP70 electronics. The OP1S / OP2S should boot and display the drive state. Record the firmware version, the parameter set number, and any active faults.
8. Clear the fault memory (P940 or via the OP1S menu). Re-acknowledge. The drive should idle in state "Ready to Run" (status "00.0").
9. Connect a motor (preferably a no-load test motor of the same frame, or the actual motor with the shaft decoupled) and run a no-load rotation test. Use a low fixed setpoint (P125 = 5 %) and a short ramp (P462 = 2 s).
10. Verify the rotation direction matches the commanded direction. If reversed, swap any two of the motor leads at U2, V2, W2 — never swap the DC-link polarity.
11. Reconnect the fieldbus (PROFIBUS / USS). Verify the PLC sees the slave as "healthy." If not, return to the PLC Communication section above.
12. Ramp the drive to application speed under no load, monitoring motor current, DC-link voltage ripple, and IGBT temperature. If any parameter drifts more than ±10 % from the no-load baseline, stop and investigate.
13. Re-couple the load. Run a loaded commissioning pass at 25 %, 50 %, 75 %, and 100 % of rated speed, with a torque load where the application allows.

Fault Diagnostics and KTY84 Monitoring

The TP70 firmware monitors the motor temperature through a KTY84 sensor (PTC-compatible silicon sensor with linear resistance-temperature characteristic). The sensor is wired to the encoder / temperature feedback connector on the TP70 control board. The VC332 manual specifies the terminals as:

• -X103:29, -X103:30 — KTY84 input on the standard CUSA / CUVC control board (slot A)
• -X104:29, -X104:30 — KTY84 input on the optional second control board (slot B), if installed

A fault on the temperature channel typically reports as F0015 (motor temperature exceeded) or F0016 (sensor open / short) on the OP1S. When the fault appears immediately after a TP70 swap, the most common root causes are:

Verification Checklist

After commissioning is complete, walk the following verification list with the maintenance team:

1. DC-link voltage steady-state: 510–600 V DC under all load conditions, with ripple under 5 % of nominal.
2. Pre-charge operation: pre-charge contactor closes within 5 seconds of AC line energization, no audible contactor chatter, no inrush fuse failures.
3. Motor rotation direction matches command for both forward and reverse.
4. Motor current at rated load matches the nameplate FLA within ±5 %.
5. KTY84 temperature reading tracks the motor frame temperature within ±5 °C at steady state.
6. PROFIBUS / USS communication: cyclic data exchange is active, no diagnostic interrupts logged in the PLC, PZD update time under 4 ms for a 6-word PPO.
7. Fault memory is clear (P940 = 0) at the end of commissioning, and a fresh fault log is started for the post-commissioning period.
8. All parameters are saved to the EPROM (P971 = 1, then a 5-second wait for the save to complete).
9. Parameter set dumped to DriveMonitor / SIMOVIS and stored on the maintenance server under a filename that includes the date and the cabinet ID.
10. Documentation updated: cabinet drawing reflects the new DC-bus rectifier, braking chopper, and pre-charge contactor; the drive replacement is logged in the asset management system.

Field-Proven Caveats and Edge Cases

Several recurring issues appear in the field during EP50 → TP70 swaps. The following list captures the most common ones.

• Cooling fan wiring: The TP70 internal cooling fans may run on a different voltage tap than the EP50 (380 V vs 400 V). On a 50 Hz European supply, this rarely matters; on a 60 Hz supply with a higher nominal voltage, the fans can run hot and fail prematurely. Verify the fan tap setting on the TP70 nameplate.
• Encoder feedback board: If the EP50 used a resolver or TTL encoder on the SBR / SBM boards, the TP70 must have the matching option board installed. A blank slot will not accept the EP50's encoder signals. The commissioning will fail with F0051 (encoder error) on the first run command.
• Brake control output: The TP70's brake control logic is gated by different parameters than the EP50. If the application holds a vertical load with a motor brake, verify P625 / P626 / P627 against the application requirement before lifting the load.
• OP1S parameter copy: A parameter set downloaded from an EP50 into a TP70 using an OP1S parameter copy module will partially load. The TP70 will accept the motor data and the peer-link addresses, but will reject the EP50-specific rectifier and DC-link parameters. The drive will fault on power-up. Always edit the parameter set on a PC before uploading it to the TP70.
• EMC filter sizing: The TP70, as a bus-fed device, may have different EMC filter requirements than the line-fed EP50. A filter sized for the EP50's rectifier harmonics may not attenuate the TP70's DC-link ripple. Verify the filter is rated for DC operation, not just AC.

Documentation and Spare-Parts Strategy

Once the migration is complete, update the documentation set so that the next maintenance event is not blocked by the same topology ambiguity. The following artifacts should be in the cabinet folder and in the maintenance database:

• Updated single-line diagram showing the new DC-bus supply stage, the pre-charge circuit, and the braking chopper.
• TP70 parameter set dump, dated and signed off by the commissioning engineer.
• TP70 nameplate photograph showing the MLFB, serial number, and firmware version.
• DC-bus voltage log captured at 25 %, 50 %, 75 %, and 100 % load, with the test date and the motor nameplate data.
• PLC-to-drive telegram capture (PROFIBUS trace or USS log) showing the cyclic data exchange in healthy operation. This is the baseline for future diagnostics.

For spares, keep at minimum one TP70 of the exact MLFB on the shelf, plus one matching CUSA / CUVC control board, plus one KTY84 sensor with the correct lead length and connector. Lead time on a 6SE70 series unit from Siemens is typically 6–12 weeks, and a production line cannot wait that long for a swap.

Summary of Key Engineering Decisions

The EP50 → TP70 migration is a topology change, not a component change. The five engineering decisions that drive the cost and schedule of the swap are:

1. DC-bus supply topology: diode rectifier + chopper (lowest cost), BLM (middle), or ALM (highest cost, best harmonic performance).
2. Pre-charge strategy: integrated pre-charge on the BLM/ALM, or external pre-charge contactor + resistor for the diode rectifier option.
3. PLC integration strategy: rewrite the PLC program (when the source is available and the budget allows), or insert a protocol gateway (when the source is not available and the schedule is short).
4. Braking energy strategy: braking chopper + resistor (dissipative), or ALM with regenerative line-side inverter (recover energy to the AC line).
5. Spare-parts strategy: keep a TP70 on the shelf, or accept the Siemens lead time and plan the spare-parts budget accordingly.

Each of these decisions is independent of the others, but the answers cascade into the cabinet layout, the bill of materials, the PLC program structure, and the maintenance documentation. A well-planned EP50 → TP70 migration can be completed in a single maintenance window; a poorly planned one can drag into weeks of troubleshooting.

FAQ