GE DS200DPCBG1AAA | Mark V DS200 Auxiliary PSU

Description

Product Introduction

The standard DPCBG1A tops out at 2.1 A. That’s fine for a handful of transmitters. But when you’re driving eight solenoid valves that each pull 250 mA inrush, you need more. A pipeline station in Texas kept blowing the fuse on their standard DPCBG1A every time a valve cycle started. The fix was the AAA revision. The DS200DPCBG1AAA is the high-current version of GE’s auxiliary power supply. Same isolation rating. Same convection cooling. But 3.5 A continuous on the +24 V output — 80 watts instead of 50.

What changed inside? Bigger transformer, heavier rectifier diodes, and a redesigned output filter with 2,200 µF of capacitance (up from 1,000 µF). The input range is the same: 85–264 VAC or 100–300 VDC. The board is physically larger — it occupies two slots instead of one because of the transformer size. And it’s heavier. You’ll feel the difference. The “AAA” suffix indicates the high-current variant. Don’t try to swap a G1A into a AAA slot and pull 3.5 A. The G1A will cook.

Key Technical Specifications

Quality Inspection Process (SOP Transparency)

Incoming Verification — First check: the heat sink. The AAA has a taller, finned heat sink that extends above the board edge. Counterfeit boards often use the smaller G1A heat sink. Measure the transformer dimensions: 50 mm x 50 mm x 40 mm tall. The G1A transformer is 40 mm x 40 mm. Visual inspection under magnification: looking for the two output capacitors — they should be 105°C rated, brand name (Rubycon, Nichicon, or United Chemi-Con). Input fuse: 3.15 A, 250 V, slow-blow. Any signs of previous repair? Burn marks around the transformer mounting screws are normal — the screws ground the core. Burn marks elsewhere are not.

Live Functional Test — Test rack uses a variable AC source (0–300 VAC, 10 A) and a load bank capable of 5 A. Input at 120 VAC. Power-on: output reaches 24.00 V ±2% within 300 ms — the larger output capacitors slow the rise time. Ramp the load from 0 to 3.5 A in 0.5 A steps. Voltage must stay within 23.5 V to 24.5 V. Then test at 240 VAC and 125 VDC input. Inrush test: apply 240 VAC with the output at full load. Input inrush current should stay below 30 A peak. Short-circuit test: short the output for 10 seconds — the board should current-limit and recover within 2 seconds after the short is removed.

Electrical Parameters — Output ripple at 3.5 A load, 120 VAC input: <120 mV peak-to-peak (measured with oscilloscope, 20 MHz bandwidth). Higher than the G1A because of the higher current, but still within spec. Isolation test: 1500 VAC between input and output for 2 seconds — leakage current below 5 mA. Insulation resistance input to output: >100 MΩ at 500 V DC. Ground continuity from mounting holes to input ground terminal: <0.05 Ω. Measure the output voltage temperature coefficient: from –20°C to +55°C, drift must stay below 0.5%.

Firmware Verification — No firmware. Analog supply. But we do check the overcurrent protection threshold. Ramp the load slowly past 3.5 A. The output should fold back at 4.2 A ±0.3 A. Not exactly 3.5 A — the protection has tolerance. If it doesn’t fold back until 5 A, reject the board.

Final QC & Packaging — QC sticker on the top of the heat sink. Anti-static bag — oversized because the board is double-width. Foam end caps. Double-wall carton. We include a load test report and a thermal image showing the heat sink temperature at full load. The board passes if it holds 3.5 A at 55°C ambient for 2 hours without tripping.

Field Replacement Pitfalls

Slot Allocation — The AAA is double-wide. It occupies two adjacent slots in the Mark V card file. I’ve seen techs try to install it in a single slot. It doesn’t fit. They force it. The edge connector bends. Verify you have two empty adjacent slots before ordering. A refinery in Louisiana didn’t check. They had to reshuffle six I/O boards to make room. The outage extended by four hours.

Heat Sink Clearance — The tall heat sink needs clearance above the board. The card file cover must have at least 15 mm of clearance from the top of the heat sink. Some Mark V cabinets have a lower cover that sits closer. Measure before installation. I’ve seen AAA boards with scratched heat sink fins because the cover rubbed. The scratches don’t kill the board, but they reduce cooling efficiency by about 5%. Use a spacer or replace the cover. A compressor station in Oklahoma added rubber spacers to the cover mounting screws. Gained 10 mm. Problem solved.

Output Capacitance Interaction — The AAA has 2,200 µF on the output. That’s a lot of capacitance. When you power up a solenoid valve bank, the inrush current can be 2 to 3 times the steady-state current. The AAA handles it — the 5 A peak rating covers most inrush events. But if you have extremely capacitive loads (some transmitters have input capacitance), the AAA may see the load as a short circuit at startup. Add a precharge resistor or a soft-start module. I watched a solar farm in California connect 20 transmitters to one AAA. The total input capacitance was 1,000 µF. The AAA tripped on overcurrent at every power-on. Added a 10 Ω, 10 W resistor in series with the output for 1 second at startup. Solved.

Ground Loops with Multiple Supplies — The AAA output is floating, same as the G1A. If you use one AAA to power field devices on the left side of a skid and another AAA to power devices on the right side, the two supplies can have different ground potentials. The difference comes from the AC input wiring — voltage drop along the neutral line. I’ve measured 2 V AC between the COM terminals of two AAA boards in the same cabinet. Tie the COM terminals together with a heavy wire (12 AWG). Then ground that common point at one location only. A chemical plant in Texas had erratic 4–20 mA signals because of a 1.5 V potential difference between two supplies. Tied the COM terminals together. Signals cleared up.

Convection Cooling in Double-Wide Format — Two slots mean more surface area for cooling. That’s good. But it also means the AAA blocks airflow to adjacent slots. The board below the AAA gets less convection cooling because the AAA’s transformer sits directly above it. I’ve measured the board temperature in the slot directly below a AAA running at full load. The board below ran 8°C warmer than normal. Leave the slot below the AAA empty. Don’t put a heat-sensitive board — like an analog input module — directly underneath. A pipeline station in Wyoming put a thermocouple input board below their AAA. The thermocouple readings drifted by 2°C as the AAA load varied. Moved the thermocouple board three slots away. Drift disappeared.

Get these five right and you’ll cut rework time by 90%.

New Original vs. Refurbished: Why It Matters

What “New Original (New Surplus)” means — This DS200DPCBG1AAA came from GE’s production run for high-current applications. GE manufactured fewer AAA boards than G1A boards — maybe 5% of total auxiliary PSU production. Zero operating hours. The transformer is fresh — no insulation degradation. The output capacitors are new, with full 2,200 µF capacitance (not the 1,800 µF you get from aged caps). The heat sink fins are clean. This is the board for high-load applications where the G1A struggles.

Refurbished risk in plain terms — Refurbished AAA boards are often G1A boards with a bigger heat sink bolted on. The transformer is still the smaller 40 mm unit. The output capacitors are still 1,000 µF. But the label says “AAA.” We tested six “refurbished AAA” boards from online sellers. Four were relabeled G1A boards. Two were actual AAA boards but with output capacitors replaced by 85°C units. The fake AAA boards failed our 3.5 A load test within 15 minutes — output voltage dropped to 20 V, then the thermal shutdown tripped. The ones with 85°C capacitors ran at full load but the capacitors bulged after 24 hours.

Real cost of a refurbished failure — A water treatment plant in California bought four “refurbished AAA” boards at 900 each. They installed one to power a bank of 12 solenoid valves. The board failed after three months — output capacitor shorted, took out the input fuse, dropped the valves closed. The plant lost flow control. Regulatory fine for discharge violation: 75,000. Emergency board replacement: 1,200. Lost production: 40,000. The four refurbished boards cost 3,600 total. New surplus AAA boards would have cost 6,000. The 2,400 “savings” cost them 116,200.

What we provide as proof — GE packing slip showing the “AAA” suffix and double-width configuration. Transformer dimensions — we photograph the transformer with a ruler. Output capacitor brand and temperature rating — we photograph the capacitor label (105°C, Rubycon or Nichicon). Load test report at 3.5 A for 2 hours with temperature logging. Output ripple measurement at full load. Inrush current capture from the oscilloscope.

Pricing context — Our price sits 30–40% above refurbished boards (most of which are fake) and 10–15% below GE’s last list price. The premium covers the genuine double-width transformer, the full 3.5 A load test, a 12-month warranty, and the certainty that you’re not buying a G1A with a fake label.

Performance Benchmarks & Test Results

Load regulation — 24.1 V at 0 A, 23.8 V at 3.5 A — 1.2% drop. Line regulation: 23.9 V at 85 VAC input, 24.1 V at 264 VAC input — 0.8% change. The AAA holds regulation slightly worse than the G1A because of the higher current, but still well within the ±2% spec. Test conditions: 120 VAC input, 25°C ambient.

Output ripple — At 3.5 A load, 120 VAC input: 95 mV peak-to-peak (spec: <120 mV). At 85 VAC input: 112 mV — close to the limit but acceptable. At 264 VAC input: 78 mV. Ripple frequency is 120 kHz, same as the G1A. Temperature effect: at 55°C ambient, ripple increases to 115 mV — still within spec.

Efficiency — 85% at 3.5 A load, 120 VAC input. 83% at 3.5 A load, 85 VAC input. 86% at 3.5 A load, 264 VAC input. At light load (0.5 A), efficiency drops to 72% — the transformer core losses dominate at low current. If your load is consistently under 1 A, the G1A is a better choice. The AAA wastes power as heat.

Thermal performance — At 25°C ambient, full load (3.5 A), the heat sink temperature stabilizes at 58°C after 60 minutes. The transformer core runs at 68°C. The output rectifier diodes run at 75°C. At 55°C ambient, full load, heat sink temperature hits 82°C — close to the 85°C thermal shutdown threshold. Derate output current by 8% per degree above 50°C ambient. At 55°C, maximum continuous output is 3.0 A. At 60°C, 2.5 A. Don’t run full load in a hot cabinet.

Transient response — Load step from 1 A to 3.5 A: output dips to 22.5 V for 500 µs, recovers to 23.8 V within 2 ms. The large output capacitors help but don’t eliminate the dip. Load step from 3.5 A to 1 A: output overshoots to 25.5 V for 400 µs, settles within 1.5 ms. The overshoot is higher than the G1A because of the larger output capacitance. 25.5 V is safe for most 24 V devices (typical absolute max is 28 V to 30 V). But check your transmitters. Some old analog transmitters have a 26 V absolute maximum.

Inrush current — Output inrush at power-on with full load: the AAA delivers up to 5 A for 5 seconds. That’s intentional. The output capacitors charge gradually through the regulator. But input inrush at 240 VAC with output at 3.5 A: 28 A peak for 2 ms. That’s within the 30 A rating of the input bridge. The slow-blow input fuse (3.15 A) handles it. But upstream breakers? A 3 A fast-blow breaker may trip on inrush. Use a time-delay breaker (C-curve or D-curve). A wastewater plant in Florida used fast-blow breakers. The AAA tripped the breaker at every power-on. Switched to D-curve breakers. Problem solved.

Reliability — GE’s published MTBF for the DPCBG1AAA: 300,000 hours (ground fixed, 40°C ambient). Lower than the G1A because of higher thermal stress. In real service, expect 50,000 to 70,000 hours before capacitor degradation — roughly 6 to 8 years. The transformer will outlast everything else. The fanless design helps. No moving parts except electrons. The AAA is a workhorse. Just keep it cool and don’t overload it. And for God’s sake, don’t buy a refurbished one.

EI813F
FI810F
FI820F
FI830F
FI840F