user: factoryops · 13 days ago Pendant is slow but gives best real-world accuracy. For production, combine both methods.

user: raj_robot · 15 days ago But you still need fine touch-ups on real robot. Digital twin is never 100% same as actual.

user: cadguru · 16 days ago Offline programming (RoboDK, Process Simulate) is faster for complex paths. You simulate collisions also.

Our pick path is curved and needs accuracy. Pendant teaching feels slow.

user: ee_mani · 10 days ago Be careful: supercaps need balancing circuits and proper cutoff. Not plug-and-play.

user: robotstudent · 11 days ago Hybrid (battery + supercap bank) works well. Many AGVs use this for smooth power.

user: electro_guy · 12 days ago Supercaps good for peak current (motors start-up), but energy density too low for long runs.

Building a delivery robot. Need quick bursts of power. Are supercaps worth adding?

user: sensorspace · 25 days ago Also lack of standardized components. Import duty on sensors, LiDAR etc increases BOM.

user: manoj_mech · 27 days ago Most founders underestimate maintenance and service requirement. Robot business is not one-time sale.

user: startupjunkie · 28 days ago Hardware margins are low + long sales cycle. Customers take months to decide.

I notice many Indian robotics companies shutdown after 2–3 years. Anyone know the real causes?

user: bharath_auto · 17 days ago If cycle time not very fast, cobot with linear slide may also work. Maintenance easier.

user: makerlabs · 19 days ago Check for backlash in Z axis. Many budget SCARAs fail in vertical accuracy.

user: pcbman · 20 days ago Epson LS series is reliable. Local brands cheaper but repeatability not same. Try to check ±0.02 mm spec minimum.

We need a SCARA for PCB loading but budget is low. Looking at Epson vs some Indian manufacturers.

user: vish_robotics · 14 days ago Also need good lighting. Depth cameras give noisy data in dark godowns.

user: deepak_sorter · 16 days ago Try suction grippers with large pads. Robot doesn’t need perfect geometry then.

user: harshvision · 17 days ago 3D picking works if box edges are not crushed. Many warehouses here reuse boxes too much — robots fail to detect exact corners.

Thinking to deploy robot for pick-and-place of cartons using 3D cameras. Concern: carton quality varies a lot here.