- name
- scan-planning
- description
- Expertise in planning and configuring data acquisition scans
## Scan Planning Expertise
When helping with scan planning and data acquisition:
### Scan Type Selection
- **Linear scans**: Best for simple parameter sweeps, characterization
- **Grid scans**: For 2D mapping, finding optimal conditions
- **Spiral scans**: Efficient for circular regions, often faster than grids
- **Adaptive scans**: When you need to focus on regions of interest
- **Count scans**: For time-resolved measurements at fixed position
### Parameter Optimization
- Consider the trade-off between resolution (step size) and time
- Estimate total scan time: (num_points × dwell_time) + overhead
- Account for motor movement time in estimates
- Use coarse scans first to identify regions of interest
### Data Quality Considerations
- Dwell time: longer = better statistics, shorter = faster
- Typical ranges: 0.1s (fast survey) to 10s (high-quality data)
- Consider detector saturation limits
- Account for beam intensity variations over time
### Common Scan Patterns
- Energy scans: step through photon energies (XAS, RIXS)
- Position scans: map sample or beam position
- Angle scans: rocking curves, reflectivity
- Time scans: monitor stability or dynamics
### Best Practices
- Always specify: motor, start, stop, num_points, detectors
- Set appropriate metadata (sample name, conditions)
- Consider running a quick test scan first
- Plan for data backup and documentation
### Communication
- Summarize the scan configuration before starting
- Estimate total measurement time
- Suggest how to verify data quality during acquisition
- Recommend follow-up scans based on initial results
View on GitHub