Opentrons: bug: API2 "Cannot aspirate more than pipette max volume"

[dany-fu]

Overview

The mix, transfer with mix_before or mix_after args, and aspirate functions throw "Cannot aspirate more than pipette max volume" even if the volume is less than or equal to pipette.max_volume.

Current behavior

The transfer method throws AssertionError: Cannot aspirate more than pipette max volume when mix_before or mix_after is specified, even if I pass in pipette.max_volume as the second parameter of the tuple.

Using the atomic commands mix and aspirate throw the same error.

Steps to reproduce

# throws AssertionError: Cannot aspirate more than pipette max volume
    pipette.transfer(v, source, dest,
                     new_tip='never',
                     mix_before=(5, pipette.max_volume),
                     mix_after=(5, pipette.max_volume),
                     blow_out=True)

or

pipette.mix(repetitions=5, volume=pipette.max_volume)

or

pipette.aspirate(volume=pipette.max_volume, location=source[0])

Error occurs both on the command line using opentrons_simulate on MacOS and on the Opentrons app on Windows.

Expected behavior

No errors should be thrown

┆Issue is synchronized with this Wrike Task by Unito

[nickcrider]

Was not able to reproduce on 3.18.1

metadata = {'apiLevel': '2.0'}

### Run Protocol ###
def run(protocol: protocol_api.ProtocolContext):

    # tipracks
    tiprack_multi = protocol.load_labware('opentrons_96_tiprack_300ul', '10')
    tiprack_single = protocol.load_labware('opentrons_96_tiprack_1000ul', '11')

    # labware
    plate = protocol.load_labware('nest_96_wellplate_100ul_pcr_full_skirt', '1')

    # pipettes
    p_multi = protocol.load_instrument('p300_multi', 'right', tip_racks=[tiprack_multi])
    p_single = protocol.load_instrument('p1000_single', 'left', tip_racks=[tiprack_single])

    # commands

    protocol.comment("single Max vol is: {}".format(p_single.max_volume))
    protocol.comment("multi Max vol is: {}".format(p_multi.max_volume))

    p_single.pick_up_tip()
    p_single.aspirate(volume=p_single.max_volume, location=plate['A1'])
    p_single.dispense(volume=p_single.max_volume)
    p_single.mix(volume=p_single.max_volume, location=plate['A1'], repetitions=1)
    p_single.transfer(p_single.max_volume, plate['A1'], plate['A2'],
                     new_tip='never',
                     mix_before=(5, p_single.max_volume),
                     mix_after=(5, p_single.max_volume),
                     blow_out=True)

    p_multi.pick_up_tip()
    p_multi.aspirate(volume=p_multi.max_volume, location=plate['A1'])
    p_multi.dispense(volume=p_multi.max_volume)
    p_multi.mix(volume=p_multi.max_volume, location=plate['A1'], repetitions=1)
    p_multi.transfer(p_multi.max_volume, plate['A1'], plate['A2'],
                     new_tip='never',
                     mix_before=(5, p_multi.max_volume),
                     mix_after=(5, p_multi.max_volume),
                     blow_out=True)

Simulates successfully.

[SyntaxColoring]

@dany-fu: Could you post your full protocol file? I've seen this error before when the wrong tips were accidentally loaded.

[dany-fu]

@SyntaxColoring thanks for the hint.

It looks like the problm is that I'm loading opentrons_96_filtertiprack_200ul into a p300_multi_gen2 pipette (I have to use the 200 because the 300 does not have a filtered version). So when I call pipette.max_volume it returns 275 which exceeds the tip's max volume. That was a fundamental misunderstanding on my part about what max_volume actually measured.

I tested this by loading in opentrons_96_tiprack_300ul and my protocol ran without errors.

For now, I manually lowered the max_volume to 175, but is there a programmatic way of getting the tip's max volume? It threw the same error if I specify 200, so I just guessed 175. But more importantly, it's being used in a function that's potentially reused across different tip sizes.

[SyntaxColoring]

@dany-fu Sorry for the delayed reply!

---

So when I call pipette.max_volume it returns 275 which exceeds the tip's max volume. That was a fundamental misunderstanding on my part about what max_volume actually measured.

Yeah, we should clarify that in the documentation. I've added it to our to-do list.

---

is there a programmatic way of getting the tip's max volume?

I don't believe there's an official way, but there might be a hacky way if you're willing to rely on internal, unsupported implementation details. I haven't had a change to look into this very deeply yet.

---

It threw the same error if I specify 200

I was unable to reproduce that. Here's my test script, tested with opentrons_simulate v3.18.1.

metadata = {"apiLevel": "2.3"}

def run(protocol):
    tip_rack = protocol.load_labware("opentrons_96_filtertiprack_200ul", 1)
    plate = protocol.load_labware("biorad_96_wellplate_200ul_pcr", 2)
    pipette = protocol.load_instrument("p300_multi_gen2", mount="left", tip_racks=[tip_rack])
    pipette.pick_up_tip()
    pipette.mix(3, 200, plate["A1"])
    pipette.return_tip()

This outputs:

Picking up tip from A1 of Opentrons 96 Filter Tip Rack 200 µL on 1
Mixing 3 times with a volume of 200.0 ul
    Aspirating 200.0 uL from A1 of Bio-Rad 96 Well Plate 200 µL PCR on 2 at 1.0 speed
    Dispensing 200.0 uL into A1 of Bio-Rad 96 Well Plate 200 µL PCR on 2 at 1.0 speed
...

What are we doing differently?

[dany-fu]

@SyntaxColoring thanks for following up on this. I reran my script with 200 and could not replicate my own error, so the first one I encountered must have been caused by some other bug.

The Pipette class has a working_volume property which would be exactly what I need, anyway I can get access to that? I tried pipette.working_volume() but that just throws 'InstrumentContext' object has no attribute 'working_volume'

[ahiuchingau]

Hi @dany-fu! To get the working volume from the Pipette class, you can use pipette.hw_pipette['working_volume'] in your protocol.

[SyntaxColoring]

To elaborate on that workaround, since it's hw_pipette isn't part of the stable API, I think that it's technically unsupported. It might change without warning on some software update. But we don't "expect" it to.

Same vein as https://github.com/Opentrons/opentrons/issues/4838.

[asogluizzo]

Need additional logs to debug this

[dany-fu]

@asogluizzo we worked out that this was a misunderstanding on what pipette.max_volume actually measured. I thought it was for the tip but it was actually returning the value for the pipette head itself. I was able to resolve by using pipette.hw_pipette['working_volume'] instead.

[asogluizzo]

@dany-fu awesome!! Thank you for the update and we will close this issue now!

[dany-fu]

To elaborate on that workaround, since it's hw_pipette isn't part of the stable API, I think that it's technically unsupported. It might change without warning on some software update. But we don't "expect" it to.

Same vein as #4838.

@SyntaxColoring @ahiuchingau I guess this did change. I can no longer get the tip's max volume by calling pipette.hw_pipette['working_volume']

from opentrons import protocol_api
metadata = {'apiLevel': '2.7'}

def run(protocol: protocol_api.ProtocolContext):

    tiprack_200 = protocol.load_labware('opentrons_96_filtertiprack_200ul', '10')
    tiprack_10 = protocol.load_labware('opentrons_96_filtertiprack_10ul', '11')

    p300 = protocol.load_instrument('p300_multi_gen2', 'right', tip_racks=[tiprack_200])
    p20 = protocol.load_instrument('p20_multi_gen2', 'left', tip_racks=[tiprack_10])

    protocol.comment("Tip 200 Max vol is: {}".format(p300.hw_pipette['working_volume']))
    protocol.comment("Tip 10 Max vol is: {}".format(p20.hw_pipette['working_volume'])) 

Tip 200 Max vol is: 300
Tip 10 Max vol is: 20

Is there a new workaround for when I use a tip that has a smaller volume than the pipette's max?

[SyntaxColoring]

@dany-fu I tested that script in v3.21.2 and v3.19.0 (adjusting the apiLevel as necessary) and I didn't see any change in behavior.

But note that if you insert a p20.pick_up_tip(), the P20's reported max volume does drop down to 10. In other words, currently, hw_pipette['working_volume'] returns min(pipette_volume, tip_volume) if there is a tip attached, or pipette_volume if there isn't. Maybe that's the source of the "change"?

---

As an aside, we remarked on this missing feature recently in https://github.com/Opentrons/opentrons/pull/6741#discussion_r503446404. If you want to open a feature request for a pipette.working_volume property, with some background on how it would help you, I think we'd support that.

[dany-fu]

@SyntaxColoring

But note that if you insert a p20.pick_up_tip(), the P20's reported max volume does drop down to 10. In other words, currently, hw_pipette['working_volume'] returns min(pipette_volume, tip_volume) if there is a tip attached, or pipette_volume if there isn't. Maybe that's the source of the "change"?

I see.... so it has to pick up tip first because returning the tip's max volume. This is different from before but it makes more sense. Thanks!