Are your sample volumes really
what you think they are?

VolumeSense® measures the actual volume in every well of your own plate — no dyes, no custom consumables, no balance.
40 seconds for a 96-well plate read.

A major genomics research institute has used our instrument for liquid handler verification for over two years. Installed in clinical diagnostics, genomics, and research labs.

Send us a plate View performance data
MEASURED DATA · 96-CHANNEL DISPENSE · 10 µL · 4 REPLICATES · CORNING 3695 GLOBAL RESULT: PASS
Mean9.89 µL
SD0.13 µL
Inaccuracy−1.12%
CV1.31%

Mean volume per tip

MEAN OF 4 REPLICATES IN µL

1 2 3 4 5 6 7 8 9 10 11 12 A 9.77 9.76 9.83 9.75 9.75 9.72 9.76 9.70 9.84 9.76 9.74 9.87 B 9.76 9.79 9.83 9.79 9.86 9.87 9.77 9.78 9.78 9.76 9.78 9.70 C 9.80 9.86 9.85 9.85 9.85 9.83 10.31 9.80 9.89 9.82 9.87 9.80 D 9.82 9.82 9.96 9.95 9.89 9.88 9.89 9.83 9.84 9.83 9.93 9.86 E 9.87 9.90 9.89 9.88 9.83 9.99 9.96 9.91 9.87 9.90 9.85 9.92 F 9.90 9.93 9.92 9.91 9.95 9.95 10.08 9.95 9.89 9.93 9.95 9.98 G 9.96 9.96 9.94 9.96 9.92 9.94 10.00 10.01 9.95 10.06 9.93 9.97 H 10.03 9.96 9.99 9.98 9.67 10.17 10.01 10.03 9.98 9.99 9.98 9.98
below target above target outside 3 SD

Coefficient of variation per tip

REPLICATE-TO-REPLICATE SPREAD IN % · 5.00% CUTOFF

1 2 3 4 5 6 7 8 9 10 11 12 A 0.97 0.70 0.95 0.98 0.61 0.62 0.68 0.89 1.06 0.78 1.06 0.93 B 0.62 0.82 0.55 0.54 1.11 0.70 0.93 0.75 0.88 0.87 0.90 1.03 C 0.46 0.67 0.76 0.59 0.66 0.59 2.33 0.69 0.89 0.96 0.99 1.10 D 1.24 0.68 0.67 0.55 0.71 0.66 0.74 0.77 0.83 0.88 0.91 0.86 E 0.95 0.98 0.78 0.92 0.43 0.77 0.70 0.61 0.59 0.94 0.78 1.12 F 0.90 0.89 1.03 0.95 0.91 1.00 0.91 0.68 1.29 0.81 0.85 0.83 G 0.94 0.91 0.89 0.90 1.33 0.75 0.70 1.10 0.76 0.77 0.69 0.68 H 1.16 1.12 0.92 1.00 2.58 1.16 0.84 0.76 0.42 0.55 1.23 0.93
0.42% → 2.58% all 96 tips inside cutoff

4 chained 10 µL dispenses following an initial 70 µL prefill. Independent gravimetry: 9.89 µL, matches the VolumeSense mean. 4 individual dispense mean volumes (inaccuracy against gravimetry): 9.81 µL (+0.07%), 9.86 µL (−0.56%), 9.97 µL (+0.43%), 9.91 µL (−0.08%).

Why measure the well instead of the dispense?

Legacy methods verify a liquid handler by indirectly measuring what it transferred. VolumeSense directly measures what is in the well. The difference matters in three places.

Multichannel heads

Channel-level gravimetric verification is limited to 8 channels. As 96- and 384-channel heads become standard, the reference method can no longer report channel-specific performance. VolumeSense directly reads every channel of every dispense, for your liquid – no dye required.

Absolute volume, not ratios

Photometric methods report the ratio between the absorbances of two dye solutions. They cannot distinguish a fault affecting both dispenses equally — and cannot see what happens to the plate afterward. Evaporation, incomplete transfers, and post-dispense losses are all visible to a direct measurement yet disappear in a ratio.

No custom consumables

No dye inventory, no expiration tracking, no proprietary reference plates. One customer eliminated approximately $20,000 per year in consumable spend.

Performance

96 flat bottom384 flat bottom96 PCR384 PCR
Minimum well volume≥40 µL≥8 µL≥5 µL≥3 µL
Repeatability50 nL30 nL**
Read time40 s115 s40 s115 s

Minimum well volume is set by the requirement for a wall-to-wall meniscus and varies by labware geometry and liquid. Repeatability scales with well cross section and is stated for standard well geometries. *In conical wells the cross section changes with depth, so repeatability varies with fill height. In our testing, VolumeSense agrees with independent gravimetric determination of the same dispense to within 1%.

Near-infrared interferometry

VolumeSense uses low-coherence interferometry — an established method in clinical imaging and the precision metrology of solid surfaces — to measure the meniscus elevation in each well: we shine a near-infrared beam onto any liquid surface. The reflected light interferes with a reference beam, and the resulting interference pattern encodes meniscus elevation, which, together with the measured geometry of the labware, defines the volume – no reference quantities required. Well density does not limit the method. The same measurement reads plates from 24 to 1536 wells.

The near-infrared light we use is at a wavelength where photodamage to well contents is essentially zero. Samples are never exposed to white light: fluorescent tags, photosensitive compounds, and cell-based assays remain unaffected.

Only photons touch the sample. Nothing is added to it. The plate can be measured again at any point in the workflow.

The measurement principle behind VolumeSense is also described in the peer-reviewed literature
and covered by issued patents in the United States, Europe, and Canada. Publications and patents →

Three ways to see it

Start wherever you like. Each one is a real look at the instrument on real labware.

No commitment

Send us a plate

Send your labware, your liquid, or both. If you're missing one, we've got some standards on hand. We'll run some plates and share what we find.

Start here →

30–60 minutes

See it live

A video call with the people who built it. Real instrument, real plates, and your questions answered as they come up.

Book a call →

Half a day

Visit us in Hudson

Bring your labware and liquids to our lab in Hudson, Massachusetts, and spend half a day running them yourself.

Arrange a visit →