For our Tecan Fluent, the arm furthest to the left is the FCA, the MCA is in the middle, and the RGA is on the right. I wanted to use the FCA on the far right side of the deck; however, the software has a hard time moving the RGA and MCA out of the way and usually throws an error since it expects a collision. To get around this problem, I included “move arm” commands in my script to move the RGA and MCA out of the way for the FCA. This seems to affect the accuracy of the FCA (particularly channel 1). It can take a while for the FCA to recover from this. At first, I could just turn off the instrument overnight, and it was fine the next day. However, recently, it’s been taking days to recover from this.
To test this, I run a basic script and check the volume of the liquid using a micropipette (confirmed to be accurate, checked gravimetrically). Specifically, I pipette 150 uL using the 200-uL conductive tips. For this basic script, the FCA is on the left side of the instrument. Before moving the FCA to the far right, the volumes are accurate. I reran this test script after the FCA was moved to the far right position, and the volumes (usually the first channel) were off by about 10 uL.
I asked one of my coworkers and he said he was able to use the FCA on the far right of the deck without any issues. However, it seems like the accuracy is thrown off anytime I try this.
Hi, is it a liquid-FCA or an air-FCA? If it‘s a liquid FCA it might be an effect of air entering the lines connected to the channel, specifically channel 1.
What an interesting issue. Is the accuracy still poor if you move the RGA and MCA arms manually in zero-G instead of using the move commands? Are the nests/labware on the far left and the far right identical? What about the y-position of the nests? Channel alignment relative to the well can differ as the FCA travels inward and outward.
I haven’t tested using the RGA and MCA in zero-G. I believe the nests are all the same. Unfortunately, I am hesitant to run any tests that could affect the FCA, since it takes a long time to “recover” (I’d be more inclined to run tests if the problem could be fixed by restarting the instrument).
I am using free dispense pipetting. The nests on the right side of the deck are lower than on the left side. I will consider testing this but unfortunately, I hesitate to just because the FCA is back to normal now, and I don’t want to mess it up again.
It’s strange that the accuracy would be semi-permanently affected and then take awhile to recover. To me, that makes it less likely to be a teaching/position issue. Have you done an inline-filter check recently? Has the helpdesk given you any ideas?
I have tested the accuracy using a balance. The FCA suddenly becomes worse after trying to pipette on the far right (tested gravimetrically). The weird thing is that it stays incorrect, even if you use it on the left side of the deck.
I did reach out to Tecan Fluent support. I think the FCA just needs maintenance, and some of the parts replaced. They will be sending someone onsite to do maintenance and fix this issue.
I might not be fully understanding and it may have nothing to do with your pipetting problem but it’s weird to me that the software is preventing the arm from reaching physically accessible positions without a workaround. I would double check your worktable configuration and make sure there aren’t any incorrectly configured components. I feel like direct arm movements bypass the pathfinder algorithm
When I reached out to the company, they actually said this is because of a bug in our software (version 3.4) and that upgrading it should fix it. Also, for the “move arm” command, the instrument will calculate whether or not it can move it to the destination, and has stopped me several times from potentially accidently moving it to far.
This is what support at Tecan Fluent said about this: “I do see two bug that were found in the FluentControl 3.4 and corrected in later versions of the software:
120382 PathFinder cannot evade MCA384 arm as it does not use all available range
120985 RGA contouring error when evading MCA384
Users also noted that the arms could take an extremely long time to move sometimes as well during basic movements sometimes as well. The only way to correct these bugs would be to upgrade FluentControl to a newer version.”
Certainly seems like it is a bug, but I’m not sure how this bug could semi-permanently affect the FCA accuracy. Hope the update fixes your problem though, and thanks for the info.
Any chance that the extreme right position could be affecting how the energy chain (image) sits in the track or otherwise pulls on the mount or connectors of the FCA arm? I’ve seen scenarios where this chain is shortened or lengthened for one reason or another, and this can be problematic at the extreme left/right of the deck.
On the Fluent, the chain will be near the front of the arm so you’ll have to use a mirror or camera or stick your whole head in the machine to really get a good look at it.
