Hi there,
I would like to use 1000ul or 300lul filter tips to aspirate water from a reservoir and aliquot 200ul in lots of wells with the same tips on a STAR. I would like to have the best volume accuracy as possible. I’ve experimented with single channels and repeated single transfers to fill an empty plate with 200ul. However even after tampering with the correction curve I still can have dispenses in the range of 195-205ul.
Is there demo method available that I could use as an example?
How accurate are you really trying to get? How many total tests have you done?
Hello, ideally I would like to get within a deviation of ±1-2ul from the target volume but be consistent within this range without any outliers and changing tips. I’ve tried several things: clld sensitivity, level, liquid class and correction curve changes. So I was wondering if there is a demo method that I can have as an example that would help or have as reference.
Hi!
Just so that i understand you correctly, Is this a Multidispense with re-used tips or single transfers (1:1) with reusing of tips?
Would you mind posting the LC that you gotten best results with?
Hello,
Single transfers 1:1 with reusing of tips. I’ve initially tried changing tips for each transfer and did achieve very accurate results (less than ±1ul deviation from target), however I would like to achieve similar or close volume accuracy but reusing the same tips.
it is likely you will need to define multiple liquid classes for different cycles through the dispense loop,
loop 1 will be using a “dry tip” to aspirate & dispense - which invariably leads to a slightly lower volume result,
now the tip is wet, pipetting behaviour will be different & will need to be handled in a different liquid class
each liquid class will also need to have parameters defined to ensure a constant volume remains in the tip after each dispense - to equalize the tip conditions to achieve the ±1ul tolerance - and it might also be necessary to calibrate each channel individually to ensure the channels themselves achieve tolerances within that range
Hi @MSMDG,
I have had to do something very similar where volume accuracy on a rather large volume was of utmost importance. The goal here is to minimize any type of factor that could add to the “random walk” effect of volumes dispensed as you progress through the plate with the same tips.
I would start with a new and unique liquid class for this step where many of the parameters are zeroed out - air transport volume, blowout volume, over-aspirate volume all set to 0 to minimize cumulative inaccuracies from consecutive aspirate and dispense steps.
Stick with 300µL tips since the bore is smaller than that of 1000µL tips and the air gap (which is compressible) above the liquid and below the plunger is smaller. Aspirate a first aliquot of greater volume than you need - if you need 200µLs in the plate then aspirate 220µLs. Then dispense 200µLs back into the reservoir, effectively pre-wetting the tip while leaving a small amount still in the tip.
Now perform your aliquots to the target plate using a fixed height with no liquid following. Pick a height where the 200µLs will end up submerging the end of the tip just a bit.
I’d say all other liquid class parameters can be generally optimized for your convenience. Don’t set flow rates super fast to add to inaccuracy and not too slow making the protocol take super long. Then adjust your correction curve to make the volume as accurate as possible.
Matt
Thank you for the informative guide.
I’ve tried somewhat close to this but not in this exact combination. I also had liquid following and clld turned on. Thank you for this, I’ll be sure to test it out.