FA
fanuc_fitter_anil
OP
Jan 8, 2026
Running an ESAB Aristo Mig 5000i paired with a Fanuc R-2000iC/165F robot doing structural weld on 6mm mild steel frames. Every time we do an automatic nozzle/contact-tip change via the reamer station, the TCP (Tool Center Point) drifts by 0.3–0.8mm. We re-touch-up the TCP and it's fine, then drifts again after the next tool change. We're using ESAB's own torch neck (PSF 405W) and the reamer station is a Binzel RTW 2000. The frames are going out of weld spec — fusion line position is wrong and we're getting rejections. Downtime for re-calibration is killing our OEE.

RC
robot_cell_carlos
Jan 9, 2026
This is almost certainly torch neck seating inconsistency — it's the most common hidden problem in robotic ESAB cells and almost nobody checks it properly. The PSF 405W bayonet connection has a 0.05mm locating tolerance but after ~5,000 cycles the mating surfaces wear and you can get 0.2–0.4mm of random seating variation per connection. The fix is two-part: first, replace the torch neck every 8,000–10,000 cycles regardless of visible condition (ESAB's recommended interval is 12,000 but that's for clean environments — production steel cells are dusty and abrasive), and second, add a "TCP verification touch cycle" to your program that runs a 4-point touch-up against a reference pin after every tool change before resuming production. Fanuc has a built-in touch-up macro (UTOOL_MEASURE) that takes about 8 seconds — cheap insurance against 3% scrap.

PR
process_eng_raj
Jan 10, 2026
Also check if your Binzel RTW 2000 reamer blade is making full contact during the cleaning cycle — if the blade is even 0.5mm off-center, it cleans unevenly and leaves a partial spatter slug on one side of the nozzle bore. When the new tip seats, that spatter shifts the contact tip axis just enough to move your TCP. We solved this on a similar cell by adding an air blow-off nozzle (from the robot program, triggered via a spare DO output) that fires into the reamer station for 1.5 seconds before and after each reaming cycle. Also, ESAB released the Aristo Mig 5000iw firmware update 3.12 in Q4 2025 that improved the torch communication protocol and reduced arc-start inconsistency by about 18% in their own benchmark — worth updating if you're still on 3.09 or earlier.

LS
lead_systems_lena
Jan 12, 2026
One thing from a systems perspective — in 2026 you really should be running laser TCP verification, not touch-sensing, if your OEE targets are above 85%. Servo-Robot makes a laser torch calibration unit (TAST-L series) that does full TCP measurement in under 4 seconds without robot contact, and integrates directly with Fanuc R-J3iC and R-30iB+ controllers. The system also monitors TCP drift over time and alerts you when wear is approaching the replacement threshold, so you're doing predictive maintenance instead of reactive. Upfront cost is roughly $8,000–12,000 USD but in a high-volume structural weld cell it pays back in 3–5 months of scrap reduction alone. We installed one on our ESAB/Fanuc line in late 2025 and our TCP-related rejects dropped to essentially zero.