For those involved in the discussion of bubbles in the Omnipod... I took apart a pod and examined it, and found the following. Before you read on, my conclusion is: There is no reason to worry about bubbles in the pod as long as you keep the pod in any orientation other than pointing "straight down". Referring to photos below: 1) The insulin port on the bottom goes has a rubber membrane where you inject the insulin. After that the insulin enters the pod chamber directly (see label "Fill Port" in the photo). There is nothing to prevent you from injecting bubbles into the pod's insulin chamber as far as I can see. So you'd want to be careful to keep bubbles out of the syringe when filling the pod. 2) During priming or actual use, a bubble could theoretically exit the insulin chamber (see "outlet" in the photo). For this to happen, the bubble would have to be present exactly at the "outlet", meaning that the pod would have to be pointing "straight down." 3) I tried this, by inducing an air bubble, then orienting the pod so that the bubble was exactly at the "outlet". Then I spun the gear, and saw the bubble move through the clear tube, through the "coupling", through the metal tube, and out the cannula. After the bubble was fully compressed out of the chamber, insulin filled the clear tube and continued out through the cannula. This means that there is nothing in the path from the insulin chamber to the cannula that would prevent air from making its way through. 4) There is no evidence of any kind of bubble detector in the path to the cannula. Just a clear tube, a coupling, and metal tube that lies within the plastic cannula (the metal tube also serves as the insertion needle). None of these components had any electrical connection to the circuit board. So what does this mean? There is nothing I can see that inherently prevents a bubble from entering the chamber, and nothing could happen during priming as far as I can see that would froce out a bubble. If there was a bubble, and if it was flloating precisely at the outlet of the insulin chamber, it could make it's way to the cannula. However, the risk of this happening seems remote. For a bubble to exit the chamber, the pod's cannula would have to be pointing straight down, and held at a very specific angle in two dimensions. Even if the pod was in this position for a short time, only the "clicks" that occur during that time would expel air from the chamber. In any position other than "straight down", a bubble would rise to other parts of the chamber where it could not be expelled. My recommendation: First, prime the pod in any position other than straight down, and don't install the pod with the cannula pointing down. Either sideways or pointing up is best (We've only applied them sideways or facing up, dd prefers it that way for comfort anyway). Secondly, I would change a pod with <20u of insulin. If there was a bubble, the risk of the bubble entering the cannula would become significant if the bubble's volume approached the volume of the remaining insulin. Our experience in almost 3 years of pod use has been that we've not seen anything that would indicate bubbles are getting into the cannula. If anyone from Insulet is reading this, any comments or corrections would be appreciated. Fig. 1: Whole OmniPod Figure 2: Closeup of fill port, outlet, compression disk, and coupling. Note small bubble under "fill port" and how it floats to the top of the chamber.