Leaking Span-8

Has anyone seen leaking Span-8 tips on disposable tips that persist even after a service and lowering flow rates for aspirate/dispenses to <150 uL/s?

I have scoured the hardware manual and IFU to find better mitigations but everything I have done in them has not aided in preventing this. Any help is greatly appreciated!

@Kastronaut Try checking the Span8 supply lines fittings at the pumps for tightness. You can also trim the ends of the tubing at the mandrels and reseat them. Let me know if you have any questions.


This should theoretically be covered under your service visit too, but have you checked to see if the thumbscrews in the syringe bank are tight? We had a similar issue, and making sure the thumbscrews were tightened completed removed the problem. We check these on a monthly basis and tighten. They definitely came loose at least ~ 6 months after our last PM.



Thank you both for those suggestions! I checked the supply lines and they feel snug, so I’ve ruled that out. I’ll take another look at the ends of the tubing in the mandrels to see if they need reseating as well.

I’ve also ensured the thumbscrews are all finger tight, we did have one become loose last month that we tightened before we saw any issues. This is seen almost exclusively with high volume tips (1025 filtered) which we mix ~500 uL at most.

It’s really not a bad idea to take the whole pump bank cover off (the light blue highlight in @UCantBcereus’s pixture) to access and visualize the syringes better. It’s just 1 screw to remove and 6 screws to loosen and you can access inlet/outlet connectors, syringe to valve, and thumbscrew to syringe plunger a whole lot easier.

Is the leaking happening with simple water-like reagent? And is it possible to seat the tips with slightly more pressure,? I had similar issues and there was no single thing I tried that resolved the problem - just periodic fine-tuning as new hardware issues popped up.

1 Like

We do the same exact kind of mixing with the same tips for one of our scripts. Its been a while since we used it, but at the time the instrument was installed and we were working on the protocol, we ended up doing 2 things that I think helped with this.

  1. Rather than use a loop of 10X aspirate/dispenses, we used a 0 uL aspirate step, and Used the “Mix prior to aspirate” option to mix the 500 uL there. We ended up aspirating at 100 uL and dispensing at 250 uL/sec. I think it was a way to get the Biomek to actually allow the higher dispense speed without getting a firmware error.

  1. After each of these mix steps, we had to drop tips and run a “Wash Tips” command. I don’t remember the reason why this occurs, but it was explained to me that mixing large volumes fast can effectively collapse the STAG in the SPan-8 pod, which would lead to leaking of the system liquid. Adding in the “Wash Tips” command seemed to be a way to “home all axes” and return the liquid lines to a good state/reestablish a STAG after the mixing.We saw leaking when mixing at those volumes until we added that in.This was ok for us since it was for sample dilutions, and we needed to use new tips anyway . Looking through the script, I even see it in other spots after we mix smaller volumes as well (125 uL). This wouldn’t work if you wanted to use the same tips after the mix step, such as a serial dilution. It also adds to the runtime, if what you’re doing is very time sensitive.

Hope this helps!

1 Like

I had it explained to me recently that with large volume syringes (1000uL), the system liquid tubing is small diameter which means that fast pipetting speeds at high volume can break apart the system liquid column creating pressure within the tubing that can effect the accuracy of volume delivered. I visualized it as akin to adding an extra spring to a mechanical system. I was basically advised to keep aspiration speed <150uL/s for 1000uL syringes. This issue presented in my own instrument as system liquid being dispensed on top of tip filters from the mandrels when discarding excess tip volume.

Another tip for mixing is to duplicate the mixing step in the template and having the first command run for C__MixCount-1 and the second command for 1 loop. It allows a bit extra control over the last mix where you may want to slow down speeds to remove all residual volume and whatnot. A “first-loop” command could also be used to overaspirate volume if you wanted to avoid bubbles in a reagent with detergent, as an example.

1 Like

Thanks for the better description that I could give @evwolfson !

The same thing happened with us, its a shame it just doesn’t affect accuracy (which for mixing large volumes isn’t a big deal), but also can cause that kind of leaking onto the filter tip.

That protocol that i pulled my technique from is going to be used sometime soon, so I’ll have someone confirm that the 250 uL/sec dispense speed doesn’t cause that leaking and confirm here.