[Previous Months][Date Index][Thread Index][Join - Register][Login]
[Message Prev][Message Next][Thread Prev][Thread Next]

Re: [IP] Re: why pick a certain pump?

> >>  >>  What is the most
> >>  >>  ACCURATE and SMALLEST pump?
> >>  >>
> >>  >
> >>  >All of the pumps are accurate, they just have different features,
> >>
> >>  Is there a third party which has done precise measurements of the
> >>  accuracy the pumps?  Perhaps the FDA does tests before they approve
> >>  them?  Do we really know that one is not substantially more accurate
> >>  than another?  It might even be that one is more accurate at low
> >>  basal rates and another at higher ones.  Is there reliable and
> >>  impartial information on this somewhere?
> >
> >The mechanics of how they work dictates the ultimate accuracy. The
> >only way they would not be accurate is if the actually missed a
> >stroke or increment or if their is an EXTERNAL leak in the insulin
> >delivery tubing or syringe.
> Any mechanical device has a certain amount of error, both in
> accuracy and in precision.  It seems likely to me that there are
> significant differences between the pumps, even in the absence of
> leaks, tubing problems, or user errors.  I would not expect any pump
> to be perfect.

Unless the drive mechanisms actually slip, which is nearly 
impossible, accuracy is difficult to screw up.  All these things are 
incrementally screw driven. Slipage is practically impossible just 
because of the nature of the mechanical design. It takes a gross 
failure like stripped teeth on a drive screw, the motor acutally 
failing to turn, a completely missed actuation cycle, a defect 
in the syringe seal or syringe itself (a leak), etc.... to produce 
any kind of error. There could be some small short term errors due to 
pressure in the system "blowing up" the tube like a balloon (partial 
occlusion) or slack in the drive mechanism, but these can not persist 
with continuous operation. Mechanical failures are probably the least 
likely. More likely failures are in the electronics -- usually pretty 
go-nogo, or a software error of some kind. In the case of electronics 
failure, again you would be looking for missing or extra incremental 
cycles. A software error is probably the most insidious, but would 
manifest itself in the same manner.

I'm writing generally as an engineer making an observation about 
systems I did not design or work on but can examine from a design 
standpoint based on the published information and looking at working 
pumps. These things are pretty bullet proof. The metering of insulin 
is not done on an analog basis, but on an incremental volumetric 
basis. The kinds of errors I would expect to see percentagewise would 
be like those in and electronic watch. Ultimately a frequency 
standard withing the pump determines the accuracy of everything it 
does. I suppose if the pump time started to wander you should get 

email @ redacted
for HELP or to subscribe/unsubscribe, contact: HELP@insulin-pumpers.org
send a DONATION http://www.Insulin-Pumpers.org/donate.shtml