I wanted to post this here as a sort of an answer to a question that still has some questions. We have a Tecan Evoware with an 8 channel LiHa and we were writing a Library Normalization script. Since the script was for normalization, we were using worklists to accomplish telling the instrument where to pick up volume and how much and where to dispense it. The team we were writing the script for asked if there could be a sip and spit of the sample as a little mix step to ensure all of the sample was being dispensed into the new well.
Unfortunately that is where we ran into an invalid operand error…My coworker and I read up on the invalid operand and tried many different variations of what the Help file suggested to no avail.
Luckily I was going to Tecan training the next week and so I brought the issue up to the instructor and they suggested that I try and move the position of the plate on the deck because possibly the tips did not have enough room to spread out. This also did not help. So I submitted a ticket to Tecan and after a few weeks of them troubleshooting, them giving me suggestions that did not work or were unclear, they brought this issue to their expert team where they were finally able to come back and tell me that if I just double my aspiration speed that the invalid operand error would go away.
So I took my test script, and went ahead and doubled the aspiration speed and the instrument was able to accomplish the single mix step. Now I would like to note that this error was only occurring with the 50uL tips and would not occur for the 10uL tips.
So my questions are still, why would doubling the aspiration speed fix this error and why would this only be an error for the 50uL tips and not the 10uL tips?
Is there any other work around for mixing when using a worklist?
A dropbox could work - really only need 10 or so lines prior to the error. If you have a timestamp, you can ctrl+f for that or usually searching for “ERROR” or “INVALID OPERAND” will get you to the relevant entries.
Crap I thought I still had a copy of the LiHa firmware commands but I can’t track it down. I was meaning to upload a copy to the wiki for a while, but I might have left it on my old work computer before leaving.
I can’t confirm without those codes, but to me it looks as if there’s a movement that is outside of allowed ranges. This would make sense why the 50tips were triggering the error but not the 10 as the dimensions are different. Not sure why the pipette speed would fix this though… If the expert line had an answer that the helpline couldn’t find, it might just be a strange EVOware “feature” that makes it fun to work with.
Something about that MTR command when the instrument responds with C5,28 is the source of the invalid operand. I believe the R in MTR is for “relative”, so likely there’s a relative movement being sent that the instrument can’t actually execute - possibly a plunger?
Sorry about the false hope; I just can’t remember what enough of these firmware commands are off the top of my head to really get a perfect picture.
Generally when I think of issues with relative moves & pipetting, I suspect that something is off with the software definition. Another indicator would be that if it works for the same tip but with a fixed movement then yes it’s possible that you’re getting some funky software backend action.
If this is the case - maybe a discrepancy between instrument well dimensions and the actual labware? Not sure what the total volume of the well is, but with the pipetting command showing 15-20uL I imagine that this mix must be close to well bottom.
If @randob could confirm if this is a standard or custom labware, that might be the key to this problem. Alternatively, just use a fixed height like @Optimize recommended - just getting a working solution can be waaaaay more valuable than spending precious time to find the perfect solution.
I love to trust but always confirm. I think if it’s detecting at a lower height than expected, it may freak out anyway given that small variations can exist for well dimension from SBS format to format.
Depends on the vendor! I typically like Tecan teachings especially on the Fluent - and 3rd party like BioShake/Alpaqua usually polish their own import files really well. My current robot - Biomek - seems to add a massive safety buffer to everything which is frustrating as hell and I have to reteach everything. That and the gripper arm has a small shelf on the actual paddles which is actually really well matched up to the ALPs but is difficult to use with custom labware.
Your comment on detecting at a lower height also made me wonder if submerge depth could have led to this error. Mixing parameters are freaking weird in EVOware - aren’t they defined in the multidispense tab of the LC editor?
