Bubbles in Pump Tubing


I’m finding that I’m getting several bubbles in my tubing after a day or so. I tried to get the camera to focus, but it’s not super clear. I am diligent about the entire prep process (no bubbles in syringe, withdraw bubbles from cartridge, check for bubbles when priming, check for bubbles stuck in needle tip). Everything looks good, and then this? I have tried tubes from different orders, insulin from different vials. I used Tandem t:slim x2 and Novorapid. I use room temp insulin. What gives?


Bubbles in the tube means there was air in the cartridge.

The technique I use to remove air has 2 steps. (And room temperature insulin, as you mentioned already.)

The first step is to get all the air out of the insulin and syringe. After filling the syringe I hold the syringe pointing up, still in the vial, and hit the syringe sharply with a ball point pen 4 or 5 times. Then inject a little into the syringe, watching for bubbles. Then give the syringe/vial a quarter turn and do the process again (hit, squeeze some bubbles back into the syringe.) After about 3 times the big bubbles would be out but I’ll still be getting a fine mist of champagne bubbles. I keep going until no bubbles at all came out 4 times in a row (one full turn). I think maybe the sharp hits from the ballpoint knock dissolved air out to make the champagne bubbles, because they keep coming even after I don’t see any remaining air at the top of the syringe. Finally I draw the full 300u into the syringe and remove the needle from the vial.

The second step is to get the insulin into the cartridge and get all the air out. With the needle pointing down, I stick it into the cartridge. Pull on the plunger to suck air out of the cartridge. Press on the plunger to force some of the insulin into the cartridge. Pull the plunger back out (bubbles and suds come into the syringe and rise to the top). Push some insulin back into the cartridge. Pull back out (fewer bubbles and suds this time). Keep going until the insulin comes out clean (no remaining champagne bubbles.) Then press the remaining insulin into the cartridge but don’t push any of the air into the cartridge. And remove the needle from the cartridge.

The procedure I use is fussy, but I get all the air out, so I never have bubbles in the tube.


When you are talking about pulling some insulin out of the cartridge and seeing bubbles, then push more insulin in then pull out until I see no bubbles, are you removing the syringe from the cartridge and injecting those bubbles into the air?


No, the needle is pointing down, so when I pull bubbles from the cartridge they rise in the syringe leaving clear insulin at the bottom, and I push some of that back into the cartridge. So with each cycle I gather more air in the top of the syringe. When I’m satisfied that there’s no air left in the cartridge I push the insulin in but stop before pushing any of the air bubble in. I suppose it would be better to withdraw the syringe to expel the bubbles before the final fill of the cartridge, but at least while removing the suds from the cartridge that doesn’t seem to be needed.


That’s one way to do it. I do a different thing, which was suggested to me by my Tandem trainer. But it accomplishes the same thing - getting air out of the cartridge before filling it.

BTW, thanks @bkh for that suggested hypothesis on the causes of the bubbles. That makes sense to me as well, but the proof of the tubing is in the filling (as they sort of say). @beans_betes , please let us know if this seems to solve your problem.

Here’s my method:
Before I put any insulin in the syringe, I put the needle into the little white hole of the cartridge, and just pull back on the syringe until the stopper is at about 1 or 2. You’re just trying to get any little air bubbles out of the cartridge reservoir, and this will do it.

Then I put the cartridge aside and fill the syringe with insulin.

Hope that helps - @beans_betes – how is it going with your pump?


Yes, in some sense I’m going way overboard. For me what you describe is just the first step, because I can get even more air out of the cartridge. Nevertheless, it’s true that Tandem recommends your procedure, and says it is sufficient.


Got it, thanks!


Interesting! I will try this too. Thank you!


Hey @bkh,
If you want to get super fancy you can try keeping your vial at negative pressure. That reduces the amount of air dissolved in the insulin.

I’ve been doing my vials like this for a while. It’s pretty easy to do. Basically you just always take out more insulin than the amount of air you put into the vial.

Here is a reference thread:


I do the same thing. Works well for me.


Off topic but I tried that once with the Omnipod Dash - i.e. squirt insulin in then draw back on the syringe to suck air out. It borked the pod; I suspect the negative pressure on the plunger within the pod delivery system just pulled it out of whack and the pod detected this, causing it to error out.


I believe that you believe you are properly filling the cartridge, but something in the process is out of balance. You are an imperfect human being. or nature, physics, is fouling you up - or both.

I vote for physics because its easier for me to understand and work with than people.

There will always be occasional small bubbles because air can dissolve in insulin. It’s a function of air pressure and temperature.

There is a 20 degree F difference between “room temperature” and your skin under clothing. There is also a lag in the temperature level of the surface of a container vs its contents. (Like the difference between interstitial glucose and blood glucose level.)

Raising fluid temperature will cause some dissolved air to escape. But like a glass of beer, some air won’t bubble out until I drink it.(Hopefully some won’t)

I don’t see any ways to avoid those things, so if it were me, I’d look carefully at what I might be doing or not doing that is fighting physics. I want my insulin to go with the flow.

Ho could air get into a cartridge? How could air get into the insulin?

Either air wasn’t completely withdrawn before introducing the insulin, or it is introduced during the filing.

If the air in that cartridge is removed slowly, it will not be trapped. It will move to toward the low pressure area where you are drawing it out. If I do it too fast, the bladder may not completely collapse, trapping air.

If I inject insulin into the cartridge by introducing it faster than the cartridge can accept it, turbulence will cause micro-bubbles to form (cavitation).

In an empty sealed container like an insulin vial or cartridge a change in fluid within it will cause it to try to expand or contact to match. If there is some air in it, and the container can’t expand fast enough the pressure differential will cause air in the cartridge and air from outside to be pulled into it through any crack, no matter how small.

How? It could happen along the sides of the needle where it penetrates the filler port. It could happen in the syringe between the cylinder wall and the plunger seal. Very tiny bubbles will be dissolved in the insulin. When pressure is relieved bubbles will form and some will stay in suspension a long time, as foam.

That’s why I was taught to roll insulin vials, not shake them.

Eventually the micro-bubbles will join into larger bubbles and rise toward the upper surface. An infusion set has a very large interior surface that promotes micro bubbles joining.

The way to avoid both issues is to take my time when filling the cartridge.

  • Before I start drawing insulin, I’ll move the syringe plunger in and out several times. The lubricant can separate and pool during storage. The action re-lubricates the seal.
  • I’ll avoid any tilting of the syringe relative to the cartridge which could create a weaker seal between needle and filler membrane.
  • After adding the insulin, I’ll tap the cartridge can dislodge any air that was trapped or stuck to the bladder wall as bubbles.
  • I’ll draw back on the plunger half way through the fill. That will cause much of the trapped air to be pulled into the syringe with some insulin. If it appears, I’ll withdraw the needle, invert the syringe and urge the air, then resume.
  • I won’t shake the reservoir or pump between filing it and filling the infusion set
  • I’ll keep the reservoir and pump oriented with filler port up as much as possible through each step of filling the infusion set.

These will give any residual trapped air the best chance of escape, and me the best chance of success.


Just did a site change and ran out of my vial so had to draw insulin out of a penfill, and it actually was far easier to get no bubbles. I may stick to ordering penfills exclusively.

1 Like

I do that when travelling, but I use the pen to fill the syringe rather than drawing the insulin out of the pen with the syringe. I’ve tried it both ways and pretty much I can’t suck stuff out of the pen; for me it just fails, I get a massive vacuum, lots of vacuum bubbles and no insulin. This is for Fiasp Novopens, they may be mechanically different to others.

When I do this I’m not creating any significant pressure differential in the syringe body; there’s high pressure in the pen cartridge, low pressure (Bernoulli effect) or trubulent flow in the syringe needle, ‘normal’ pressure in the syringe.

I also don’t have any problem drawing insulin from a vial without additional bubbles. It might be because my somewhat curious RE teacher tried to teach me meditation when I was in school, so I kind of relax and think of England.


After the various steps outlined by others to avoid bubbles, I take the filled cartridge and hold it upright with the white injection button at the top, tilted to make that the highest point. Then I tap on the side of the cartridge sharply with the barrel of a ball point pen. That should make any air bubbles in the cart rise to the top. I tap in a pattern of three and wait a few seconds for bubbles to rise if they are there. Then I put the cart in the pump and tap it again on the top a few times. Then I start priming. Sometimes I see bubbles go up the tubing as the pump primes. I keep tapping the top of the cartridge until at least half the priming is over, or until there are no bubbles coming out for at least a third of the tubing, continuing the prime until they are all gone. I have not detected any bubbles later on as the cart empties, but I don’t watch it all that carefully.

I keep my insulin in the fridge, take it out and warm it under running water to room temp.


I have used pens to fill my pump reservoirs “for ever”, mainly because I seem to get a lot of cannula issues and in those instances use the pen for boluses until I am sure I have a viable cannula in.
What I do is put the pump’s syringe needle into the pen’s rubber seal and use the pen’s dial to dial however much insulin I require, and push it into the syringe. Hope that makes sense.
I wonder if using less insulin makes drawing the air out of the pump reservoir easier. I usually only need to draw out about 2 ml, so I am able to pull back further on the syringe’s plunger. I pull back to the 3 ml mark and while holding it back with one hand tap the reservoir with the other hand to ensure I have expelled all the air.
Isn’t it difficult describing this stuff with just written words!?


Interesting method!

1 Like

Thank you for starting this very interesting thread.


Yes, that makes sense and it is exactly what I do when I’m using a pen.

The only issue I can see is, which way up to hold the pen? When I started doing this I “injected” down into the syringe, so the pen is above the syringe.

I’ve seen results from doing this that suggest the insulin (water) coming out of the needle is atomizing in the gas that surrounds the end of the needle; bubbles of water in the gas, rather than vice versa.

So I’ve started holding the syringe above the pen; there’s still gas at the start of the fill but the gas tends to head to the top (plunger) of the syringe and maybe stay out the way,

I also find the Novopen particularly appropriate for this method. Unlike the Kwikpen it auto-injects; just press the button and it does the rest. It also handles up to 80IU per injection while the Kwikpen is limited to 60 - that makes a difference for me because I use 150 to charge an Omnipod. The autoinjection proceeds smoothly, so the results are at least consistent - no dependency on how hard the Kwikpen button is pressed.

1 Like

An often missed source of bubbles post the connection is in the twist connector in the line near the pump. Early on as the pump is filling the infusion line hold upright and flick the connector a few tine as the insulin passes through until the bubbles are gone Usually this is where the smaller bubbles in a line will come from.