Hi everyone,
I’m wondering if anyone has experience with Method Recovery on Tecan Fluent systems? We occasionally use it on our Fluents, and it works really well. However, I’m curious if it’s possible to resume a recovery run on a different robot in the event that the original one goes down? I’ve asked Tecan, and their response was that it’s not possible as the data is stored in a database. But maybe there’s a workaround or another solution?
Looking forward to hearing your thoughts! 
One option is to record the processing status of each labware and use that for error recovery.
Let’s say your process has steps A, B, and C. Each time a labware passes through any of those steps, you update the status of that labware as having completed a step and store information in a file. If you are in recovery run, you can read the status of a labware and skip the steps it has already been through. If you had to run a recovery on another machine, you can copy the status files and go run the script.
This approach would not fully work if the interruption (machine crashing, script aborting) happens in the middle of a step – which usually the case. However, you might be able to do partial recovery.
Also, this is a high-level idea, so not sure how much of effort is needed to implement something like this.
Good luck!
To add to this because I’ve recently been taking a closer look at the Fluent Control Database. The database (backend) of FC stores all the data around objects like worktable, liquid classes, etc. but it also keeps a record of any actions taken by a user and information being passed back from the robot during runtime in FC. This gives the application its undo-redo functionality but means that each database is tied directly/locally to a single install instance of FC. This database is also locked down and nobody outside of Tecan can access it outside of just using the FC interface. There is the ability to modify the database data directly but FC doesn’t like that and you’ll more than likely break something. So in summary there really isn’t a way for a separate Tecan to run a recovered script given the differences b/w the databases for each FC instance. What @solomon suggested is probably your best bet to try and implement this kind of functionality.
Hi solomon and Sam,
Thanks so much for your responses.
Solomon, your idea of tracking the status for each labware makes a lot of sense and seems like a really practical approach. I’ll definitely look into it further. And Sam, thanks for explaining more about the Fluent Control database. That really helps clarify why it’s tough to transfer recovery data to another robot.
I really appreciate you both taking the time to respond - thanks for the help!