Training rig · 8 September 2026 · Water_Dev1 · watertime in the working VI.
After the line is primed and bubble-free, delivered volume is linear with pulse length above ~100 ms:
µl / pulse = 0.0350 × t(ms) − 0.48
R² = 0.998. About 14 ms of valve opening delay, then 0.035 µl per ms (~30 ms per extra µl). Do not use a single ratio through the origin; 50 ms is less efficient.
| Target drop | From the fit | Set in LabVIEW |
|---|---|---|
| 4 µl | 128 ms | 130 ms |
| 6 µl | 185 ms | 185 ms |
| 8 µl | 242 ms | 240 ms |
The old bench default of 200 ms was 6.6 µl, not 4–6 µl.
Cup weight after 50 pulses (empty cup 2.94 g):
Same data as microlitres per pulse, with the linear fit and the 4 / 6 / 8 µl targets:
Empty cup: 2.94 ± 0.02 g. Fifty pulses per condition, caught in the cup under the spout (tip not scraped). 1 g = 1000 µl.
| Pulse (ms) | Cup (g) | Water (g) | Total (µl) | µl / pulse | ms / µl |
|---|---|---|---|---|---|
| 50 | 3.00 | 0.06 | 60 | 1.2 | 41.7 |
| 100 | 3.10 | 0.16 | 160 | 3.2 | 31.3 |
| 150 | 3.17 | 0.23 | 230 | 4.6 | 32.6 |
| 200 | 3.27 | 0.33 | 330 | 6.6 | 30.3 |
| 300 | 3.44 | 0.50 | 500 | 10.0 | 30.0 |
Tare uncertainty ±0.02 g is about ±0.4 µl / pulse. Fine for choosing 130 / 185 / 240 ms.
- Measure mass in a cup under the spout. Do not try to wick the lick tube with a capillary; the tip is meant to hold a film.
- Prime until bubble-free before a run. Short pulses fail if there is air in the line (seen 18/08/26).
- Recalibrate if the valve, tubing, height of the reservoir, or needle/spout tip changes.
watertime is still a per-trial parameter-file field (column in the 21-wide
file), not a new front-panel control. Session-constant cue / delay / spout
voltages stay on the front panel until that map is confirmed.