Liquid dispenser guidance please

Hi. I am looking to automate a task in our lab and I would greatly appreciate some of your thoughts on how to best do this.

I don’t have much experience with liquid handlers or dispensers, but do have a fair bit of experience with other lab automation solutions (mainly sample changers) and I can write code in Python and some C#.

Basically, what we would like to do is to add a fixed volume (5-25 uL) to an “assay-ready” 384 well plate. Our main considerations are going to be cost and accuracy and we are less focused on speed.

I have been looking around and came across robots such as Certus Flex, Thermofisher Multidrop, Formulatrix Mantis and Tempest, Integra Viaflo, Opentrons OT-2 and Flex. These all look like they should be able to do what we want. If any of you has used robotics for a task similar than what we want to do (with the above listed or other hardware) and would share their thoughts, that would be greatly appreciated.

2 Likes

If the reagent is cheap and plentiful nothing will be faster and easier than a Multidrop. I love Multidrops for bulk reagent dispensing.

4 Likes

I second what was already stated and would say that another viable option for you is the Viaflo, great customer service too!

I will also add a vote for multidrop combi. Fast, accurate, will not break down, and easy to integrate with lots of robots. The interface is intuitive and you are right in the middle of the best use-case.

Tempest can be good for doing variable volumes, small volumes, and multiple reagents, and the dead volume is smaller, but they are much more expensive and a lot bigger. I don’t have direct experience with Certus Flex, but they seem fast and flexible and similar price point to Tempest.

I would go with a pump-style dispenser and save the cost of tips in your process.

1 Like

$16k Boom

4 Likes

Multidrop, Certus Flex, Mantis, Tempest, D2 are all good shouts for this depending on your budget, reagent types and reagent costs. If it is expensive the mantis will pay for itself over time with a much lower dead volume compared to the others. If you need multiple plates then the Tempest with stackers will enable more walk away time. If you need a relatively inexpensive but fairly inflexible workhorse the Multidrop will do the job. Most of these systems are also already well integrated with the common schedulers if that was something you’re considering for future plans.

The Flex could certainly do the job for you, if there’s value in having a more “typical” liquid handler for other tasks in the lab and your budget doesn’t stretch to having both a pipetting system AND a dispenser. As mentioned above, one of the big advantages of a dispenser for this kind of job is not needing to use tips.

Happy to have a chat through it with you or put you in touch with someone else here at Opentrons in your local area. (I’m UK/Eur) based.

1 Like

Oh word :eyes::eyes::eyes:

The other thing to really call out here is for the Mantis and the D2 dispense times per plate are going to be measured in minutes for a 384w plate.

Tempest/Combi/CertusFlex will be much faster per plate overall.

This doesn’t matter for a lot of people (and you mention not caring as much about speed) but if you’re manually loading it and it takes 2-3 minutes to do the dispense, there’s enough waiting time to be annoying. Can’t do much else while you’re waiting for the dispense in these cases.

Big supporter of tip-free systems here! Both the environment and your purse will be grateful in the long-term run.
I can recommend the Tempest and Echos from own experience. The CertusFlex is a nice instrument as well.

A big thanks to everybody for sharing their experience. This is really useful for me and much appreciated.

Even though I said that speed wasn’t really an issue for us, I hadn’t considered that longer waiting times can be annoying and lead to inefficient workflows when using a dispenser as a stand-alone tool (rather than integrated into a workcell).

A couple of follow up questions come to mind:
What is the dead volume of a multidrop? Can that dead volume be recovered by “reversing” the pumping direction?
Can a multidrop be operated in a cold room (or fridge)? Or is is better to just keep the dispensing solution on ice and operate the machine at room temperature?

Considering a system like the Viaflo with a 96 (or even 384) head: Would it be possible to use the same tips for multiple dispenses? I.e. is it possible to accurately “jet dispense” thereby avoiding contamination of the tips?

1 Like

The multidrop dead volume can be found here: https://www.thermofisher.com/TFS-Assets/LCD/Product-Guides/D20216.pdf looks like it is cassette dependent (1-24 mL). You can reverse the pump and get this volume back into the source labware you use.

As for using it in cold temperatures, I would venture it is easier to keep the solution on ice while dispensing, I would ask Combi’s support about different fluid types/viscosities to get a better handle on this as well.

The VIAFLO has a multidispense function that you speak of, and more information on it can be found here: VIAFLO 96 & 384 | 24, 96 And 384 Channel Pipettes | INTEGRA.

Operating range per the manual is 10-40c.

I’d imagine the main issue with this would be material components of the dispensing pieces. How do the material properties of the silicone tubing in the cold room impact accuracy/precision? As @Kastronaut mentioned, there’s also the viscosity and fluid type component making the dispensing more difficult.

The control screen may not work as well below the operating range either.

1 Like

yes, you can recover the volume from a multidrop combi, as you can for any peristaltic pump. I can also vote for a multidrop combi. I have tested accuracy and precision in a paper during my phd thesis and for multidrop it was really good!

3 Likes

Thanks for sharing @Bammes ! Are you able to share pdf?

I’m interested in knowing more about how dispensers preform over time. I’ve heard that tubing in peristaltic pumps can get stretched out with use, leading to changes in the volume dispensed. Is this something you quantified or observed during your studies? Maybe the Inter-Run CV of a dispenser is x% if tested on the same day, but how much does the CV increase if comparing performance at the beginning vs end of a multidrop cassette’s lifetime?

This is something I personally did not investiagate. However we use a BioTek Microplate Dispensers, which is a very good alternativ to the Multidrop btw!
This cassette we used for years and it works fine for us. But we are also not interested in accuracy in this case/application. I gues you’d need to reach out to the suppliers with those questions.

Sure, but I can not upload pdf files here. Please send me a PM with your email. I can also share my dissertation with some of the content e.g. in ch.3.

1 Like

Thanks! PM sent.

I agree that BioTek MultiFlos and Thermo Multidrops are both good dispensers. The MultiFlo FXs also have optional modules if you need additional functionality from your dispenser, but it sounds like OP doesn’t need the extras.

Has anyone here used the Integra Welljet? It’s the successor to the Viafill which was wisely rebranded because it was way too easy to confuse with the Viaflo. In my experience, Integra systems are competitive on cost, which is one of the main concerns for OP.

@labware, Do you need to mix the samples in the well with the solution you are dispensing? If not, I would try to go with a tip-less dispenser, because as @csthauer said above you will save money on consumables and use less plastic.

In addition to keeping the dispensing solution on ice, if your samples are very temperature sensitive, you could use a cold block that fits your destination plate and equipment nest.

We gave the Viafill a solid try about 5 years ago. We were upgrading from WellMates, and the design was similar. I liked the 16-channel cassettes, but we ran into some things we didn’t like so much about it. I’m sure it has improved, but the Multidrops and Multiflos were always great.

1 Like

Cost is obviously incredibly subjective from lab to lab (esp. when you roll in labor costs).

Since that was listed as a main consideration, if the systems you’re finding are out of budget, don’t forget about semi-automated solutions.

Pros: really inexpensive. (like p-card purchase that doesn’t need approvals sort of inexpensive) Really accurate (depending on the system. Hamilton’s ML600 is literally 99% accurate with it’s dispenses because it is syringe based)

Cons: not fully automated. Dispenses/dosing is automated, but a user has to move the probe from well to well. So faster and more accurate than manual, but doesn’t enable walk-away dispensing.

If possible, go fully automated–nothing beats getting to walk away from an instrument while it does the grunt work for you!

But if budgets really are tight (or throughput is low-to-medium), you can get a semi-automated system for 4-8k, depending on configuration. [Most I see are around 5k).

If you’re looking at something like Viaflow or OT-2, the Prep is worth considering as well–you get the pipetting tech of a Star in the size/ballpark price of an OT-2 (LLD, etc.). But that is going to be most useful if you need flexibility across multiple applications. People use it for loading plates, but if you can get a dedicated dispenser it is probably less expensive/faster if you don’t need the option to have other applications automated as well.

Similar to what Dan said, I’m a Hamilton guy, so I can get you connected to the right folks over here if needed.

But I’d look to a truly automated dispenser system first and have this as an ace in the hole if budgets are really tight.

I’ve definitely observed this with our multiflo. Specifically the volume dispensed is now higher than what is entered via the software, my understanding though is as tubes get stretched you can recalibrate the device to bring the volumes back in sync with the software. We haven’t had the time/motivation to attempt this but I have an acquaintance who does it routinely for theirs to keep it in spec.

I haven’t checked for CVs but my expectation would be that they would be similar to a new cassette, but I haven’t checked so can’t really comment

I don’t remember if the multiflo cassettes have them, but multidrop combi cassettes have a set of crews that adjust the tension on the tubing so you can recalibrate the cassette to stay in spec. There is a little door that slides open to reveal the screws. We never did this, but the option was there.

1 Like