GE DS200GDPAG1AKF | Mark V DS200 Pulse Input Board

Description

Product Core Brief

The AJF did 200 kHz. That was the fastest. Until a turbine manufacturer needed 240 kHz from a high-resolution encoder. GE responded with the AKF. The DS200GDPAG1AKF is the fastest pulse input board ever made for the Mark V DS200. Six channels. 250 kHz maximum. 1500 VAC channel-to-channel isolation. 24 V or 5 V inputs. 32-bit counters. Hardware latch. The input comparators have 5 ns rise time. The isolation amplifiers have 800 kHz bandwidth. The board has a larger heatsink than the AJF — 20 mm tall.

The board has six ultra-wide-band isolation amplifiers. The “AKF” suffix indicates the top-speed version. The board has six yellow LEDs and one yellow LATCH LED. The terminal block has 13 positions. The board draws 550 mA on the +5 V rail — the highest of any Mark V I/O module. The operating temperature range is 0-50°C (no coating — the heatsink is too large for the coating process). This board is for speed, not for corrosion resistance.

Key Technical Specifications

**Quality Inspection Process (SOP Transparency)

Incoming Verification — Visual inspection first. The board has six ultra-wide-band isolation amplifiers — they’re larger than the AJF version (25 mm × 18 mm). The heatsink is taller (20 mm). The input comparators are the fastest available — 5 ns rise time. The board has no conformal coating — the heatsink covers too much area. Counterfeit boards sometimes use AJF boards with a taller glued-on heatsink.

Live Functional Test — Test rack uses a 250 MHz pulse generator and a high-speed oscilloscope. Test channel 1 at 250 kHz for 2 hours. Zero missed pulses. Test at 260 kHz — the board should miss some pulses but not fail catastrophically.

Quadrature X4 mode at 62.5 kHz encoder cycles (250 kHz internal). Simulate 1,000,000 cycles. Counter should read 4,000,000 counts. Reverse direction. Counter decrements.

Isolation test: apply 1500 VAC between channel 1 and channel 2. Leakage below 5 mA. High-frequency isolation test: apply a 800 kHz, 10 V signal to channel 1 input. Read channel 2. The coupled signal should be below 20 mV.

Latch test at 250 kHz: send a latch pulse. Counters freeze within 50 ns.

Test all six channels simultaneously at 250 kHz. Run for 2 hours. Monitor for crosstalk or overheating. The heatsink should stay below 80°C.

Electrical Parameters — Input threshold: 24 V mode turn-on 15 V ±0.2 V (tighter spec than AJF). Propagation delay: 50 ns typical.

Firmware Verification — The firmware version is printed on a sticker. Version 6.0 or later. V6.0 adds the 250 kHz capability and the 0.25 ms update. Connect via the backplane. The signature is 0xGD60.

Final QC & Packaging — QC sticker on the metal bracket. Frequency test report — 250 kHz for 2 hours, zero missed pulses. Isolation test report. Latch timing capture. Thermal image at full load. Anti-static bag. Foam-lined carton with cutout for the tall heatsink.

Field Replacement Pitfalls

Heatsink Clearance — The heatsink is 20 mm tall — much taller than standard boards. Many Mark V cabinets have only 15 mm of clearance above the card file. Measure your cabinet clearance before ordering. A power plant in Indiana had to leave the card file cover off. The board ran fine, but dust was an issue. Cleaned the heatsink monthly. Some cabinets have a deeper cover option — part number 531X111. Order it.

Power Supply Sizing — The board draws 550 mA on the +5 V rail at 25°C. At 50°C, it draws 600 mA. In a rack with four of these boards, the +5 V draw is 2.4 A. Add a processor board (1.5 A) and other I/O, and you’re at 5 A — still within the 8 A rating. But the PSU derates at high temperature. Calculate your power budget including derating. A compressor station in Oklahoma had four AKF boards in a 50°C cabinet. The PSU’s 8 A rating derated to 6.5 A. Total draw was 6 A — acceptable but no headroom. Added a second PSU.

Cable Length at 250 kHz — At 250 kHz, cable length is very critical. For 24 V signals, 25 meters maximum. For 5 V signals, 1 meter maximum. The higher frequency means faster edges. Use high-quality shielded cable — Belden 9842 or better. Keep cables as short as absolutely possible. A paper mill in Wisconsin had a 50-meter cable on a 24 V, 230 kHz encoder. The signal had severe reflections. The board missed 10% of pulses. Shortened the cable to 20 meters. Counting became accurate.

Minimum Pulse Width — At 250 kHz with 50% duty cycle, the pulse width is 2 µs. That’s fine. But if your encoder has a duty cycle variation (40% to 60%), the minimum pulse width could be 1.6 µs. The board needs at least 1.5 µs to recognize a pulse. Check your encoder’s duty cycle spec. A mining operation in Chile had an encoder with 35% duty cycle. At 250 kHz, the low pulse was 2.6 µs, the high pulse was 1.4 µs (35% of 4 µs). The board missed the narrow pulses. Reduced the frequency to 200 kHz. Duty cycle became acceptable.

Latch Input Speed — The latch input response is 50 ns. That’s extremely fast. But the latch signal must be clean. A slow rising edge (over 100 ns) may cause metastability — the latch may capture an indeterminate state. Use a fast latch signal with rise time under 20 ns. A refinery in Texas used a PLC output with 200 ns rise time. The latch occasionally captured the wrong counter value. Added a Schmitt trigger between the PLC and the latch input. Rise time improved to 10 ns. Latch became reliable.

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 DS200GDPAG1AKF came from GE’s ultra-high-speed pulse input production line. GE manufactured very few of these — the 250 kHz isolated version is the rarest Mark V I/O module. Zero operating hours. The ultra-wide-band isolation amplifiers are fresh. The heatsink has never been warm. This is the fastest pulse input board ever made for the Mark V.

Refurbished risk in plain terms — Refurbished AKF boards are almost always AJF boards with a taller heatsink glued on and a new label. The isolation amplifiers are still the 500 kHz version, not the 800 kHz version. The maximum frequency is still 200 kHz. We tested one “refurbished GDPAG1AKF” board from an online seller. It had the AJF isolation amplifiers with a taller glued-on heatsink. The board failed at 230 kHz — missed 15% of pulses. The seller claimed “250 kHz” but couldn’t provide a test report.

Real cost of a refurbished failure — A high-speed turbine manufacturer in Germany bought two refurbished AKF boards at 2,500 each. They installed one on a test stand encoder running at 240 kHz. The board’s fake ultra-wide-band isolation amplifiers missed 8% of pulses. The turbine speed reading was wrong. The test was invalid. Rework cost: 100,000. The two refurbished boards cost 5,000 total. New surplus would have cost 7,500. The 2,500 “savings” cost them 100,000.

What we provide as proof — GE packing slip showing the AKF suffix. Ultra-wide-band isolation amplifier verification — we photograph the larger modules and measure their bandwidth (must exceed 750 kHz). Frequency test at 250 kHz for 2 hours — zero missed pulses. Isolation test at 1500 VAC. Latch timing capture. Heatsink temperature measurement at full load (must stay below 85°C).

Pricing context — Our price sits 30–40% above refurbished boards (which are fake) and 10–15% below GE’s last list price. The premium covers genuine 800 kHz isolation amplifiers, the tall heatsink, full 250 kHz testing, a 12-month warranty, and the certainty that your 240 kHz encoder will be counted accurately.

Performance Benchmarks & Test Results

Maximum frequency — 252 kHz at 25°C, all six channels active, zero missed pulses. At 260 kHz, the board misses about 1 pulse per 10,000.

Propagation delay — 52 ns typical. The fastest of any GDPAG board.

Isolation bandwidth -3 dB at 820 kHz. The board passes 250 kHz signals with minimal distortion.

Latch response — 48 ns typical.

Update rate — 0.26 ms typical for all six channels — 4 updates per millisecond.

Input threshold precision — 24 V mode: 15.03 V ±0.1 V. 5 V mode: 2.52 V ±0.03 V. Extremely precise.

Minimum pulse width — 1.5 µs high, 1.5 µs low at 250 kHz. At 270 kHz, the board needs 1.8 µs.

Crosstalk at 250 kHz — Channel 1 at 250 kHz, channel 2 reads 0.005% of full scale. The ultra-wide-band isolation is excellent.

Power consumption — 550 mA at +5 V (2.75 watts) at 25°C. At 50°C, 600 mA (3 watts).

Thermal performance — At 25°C ambient, the isolation amplifiers run at 48°C with the heatsink. At 50°C ambient, they hit 78°C — within their 85°C rating. The heatsink is effective.

Reliability — GE’s published MTBF for the GDPAG1AKF: 130,000 hours (ground fixed, 40°C ambient). The AKF is for the fastest of the fast. When a turbine encoder spins at 15,000 RPM with 1000 pulses per revolution (15,000/60 × 1000 = 250 kHz). When a flowmeter pulses at 240 kHz. It’s the top. The rarest. The most expensive. The most power-hungry. It needs a tall heatsink and short cables and a clean latch signal. But it delivers 250 kHz with full isolation. No other Mark V board can do that. Just measure your cabinet clearance. Keep cables under 25 meters. Watch your power budget. And don’t buy refurbished. The fake boards have slow amplifiers, fake heatsinks, and missing bandwidth. And you won’t know until the turbine speed is wrong. At 3 AM. On a test stand. In Germany. Ask me how I know.

ABB DSTA156 PLC DCS
ABB IP854 3BSE025349R1 PLC DCS
ABB 3BSE025349R1 PLC DCS