Other wish-list things. Just random stream of consciousness talk here…
Would be nice to be able to manually enter insulin that was taken by means other than a pump.
A ZB for sure. No carbs or ISF or anything like that. Simply - “I want zero insulin for X hours, Loop!”
It would be very helpful if I could limit the time it does zero basal for me based on my lows. This has happened twice on me. It saw a low, freaked out and cut basal to zero for hours. And that was followed by a horrible spike after. That’s why I don’t want to be in Closed during the day, stuff like that happening. Could I put a limit on the time it will do ZB for me when it’s trying to fix my low BG?
I think if you manually enter your insulin in Apple Health it will be reflected in Loop. I’ve done this when I used a syringe because pod had absorption issues.
Not sure, but I think that only determines the “feel” of how quickly the values are changing in response to physically moving the crown. The actual value increments are probably set elsewhere. Have not played with that part of the code at all - will take a look and get back to you later.
I used my phone for my run last night. It was not super easy, but it was manageable when at a slower pace. Turning overrides on and off was easy. Bolusing was easy. But unlocking the phone was painful. And doing it at speed will be much more of a challenge. So moving forward I will need to get the watch into the mix at some point.
No, you can’t and you should not need to do that (ever). If you had Loop zero temping for hours, something is way off. With all due respect, the custom insulin model you were showing earlier has nothing to do with reality (wish we had such insulin available ) - that model could be behind the too-long zero temping.
Great, I did this, too, but had to stop running to do it! Then decided to try the watch out!!! Yes, unlocking iPhone with fingerprint often doesn’t work w/ sweaty hands!
An insulin moving faster in and out would seem to reduce the length of zero basal.
Look at extreme cases as examples for comparison.
If my insulin duration was 15 minutes, a 30 minute ZB would mean that I had no insulin remaining. Certainly 30 minutes would be totally sufficient.
If my insulin duration was 10 hours, a 30 minute ZB would be meaningless. I would still have plenty of insulin left. So it would need to run the ZB longer.
I think it is much closer to my actual than the one that comes with loop.
Insulin does NOT last 6 hours in my body. Not even close.
On the other hand, yes, we are all different, I can’t argue with that, and I have no doubts you know what you are doing - it’s just that I think I’ve seen that statement hundreds of times from people who were just starting Loop and thought DIA = 6 hours could not possibly be correct (self included). The tail end is very thin - hard to notice.
Your quick counter-example is good, short DIA was probably not behind too long zero temping. Maybe the reason is different: since you are not entering any meals, but you are entering insulin, I’d imagine Loop thinks you are going to go low all the time. At the time when you close the loop, if there is any IOB, left, and if you are actually going low, I can see how Loop would want to keep zero temping for extended periods of time. Another possibility is that bg did remain below suspend threshold for long periods of time for whatever reason (maybe because insulin was acting longer than you thought it would?), so Loop just shut down insulin for safety reasons.
Part of the problem is that the Dex is so slow to respond. (Part of the reason for my Dexcom G5 vs G6 - Poll I posted today!)
So if Dexcom still shows me low, and the Loop stays suspended longer than it needs to, that ends up in a big spike for me.
I’d like to be able to limit the length of time Loop does the ZB. That is kind of how the Medtronic 640G (or whatever their first generation was called). They put a time limit on basal reductions.
I know from personal experience, not on Loop, but just in general pump use, if I do a very long ZB (that is not while exercising) I will have a spike later. Always seems to happen.
For what it’s worth, I always wait to have no IOB before turning Loop on at night, just to remove any mess there.
Totally agree. After long (non-exercise) ZB, our basic model is broken, probably because liver has no insulin available to convert and store glucose, and anything just shoots bg up, all bets are off. On the other hand, if bg just stubbornly remains below suspend threshold for long periods of time, what should we do other than ZB? Some arbitrary limit on ZB time, however, somehow does not look right to me. There is no way an average user could possibly set yet another obscure parameter correctly. Maybe we could include a limit on how large (negative) IOB would be allowed? Or, somehow refine the model to better predict bg rise. By the way, the concept of counting IOB against the programmed basal is also a bit strange, isn’t it? There have been long discussions about that among long-term DIA system users and developers. The original authors started with that concept, and it’s now hard to change. I am personally not sure about it - it might be better if zero IOB meant actual zero IOB.
BTW, really enjoy picking up your brain on these topics @Eric - thanks!
Yes I’ve also enjoyed the discussion. It’s been very helpful for me to better understand Loop.
I know my use is not typical.
I also know that Loop has to be useful for the general D population. And there is no sense cluttering it up for a one-off like me.
But I was hoping to be able to customize it over the coming weeks. Leave it as-is for everyone else, because they use it the correct way. But make it better for me.
Sure, that’s the real beauty of DIY open-source stuff. The last time I ran a stock Loop was maybe 2 years ago. I am doing all kinds of weird stuff with it, but all closed-loop, no manual corrections whatsoever.
First off, @Trying, I apologize for this thread hijack. We can split this off if you would like. These discussions are not completely in-line with your original thread topic! If you think we should split, let’s contact an admin and do that!
I wanted to think through this thoroughly before replying, but now after thinking about it a bit, I’d like to debate this.
One of the nice things about the normal Insulet PDM is that if you set a temp basal, you know 100% that the pod will honor that for the duration of the temp. If you set a temp basal for 2 hours and go to the hardware store and leave your PDM at home, no worries, the pod will keep that temp basal running.
Here are some scenarios I am not totally sure about, but I think that when running Loop, the Override basal would disappear.
You set a temp Override of several hours and go out for a run and leave your RileyLink (RL) at home. I think that if Loop does not see the RL, the temps are only for 30 minutes and then it returns to the normal basal, right?
The green light on the RL shows that it is connected via bluetooth. BUT, there is nothing on RL that shows how much battery life is left!! What if the RL charge wears out??? What if you are on a run and the RL battery dies?!? If the RL dies, after 30 minutes, does the pod return to the normal basal? Lemme tell you, that could be a killer…
What if the pod is connected on your back side, and the RL is in your front pocket and loses connection. Is the override lost? After 30 minutes, does it return to the normal basal? What happens if your RL loses connection?!?
What if your phone battery dies and your phone shuts off? After 30 minutes, does the pod return to the normal basal?!?
When you are exercising, you can’t check to see if the RL has the bluetooth green light on and is still connected. There is always uncertainly that the temp basal is still in play. And if the RL battery has died or it loses connection because of a bluetooth problem, the pod returns to normal basal rates after 30 minutes, and you are toast. Again, as stated above, during exercise this is a killer.
What if you accidentally turn Loop off?!? If you have set a temp basal but inadvertent shut the app off on your phone? The pod returns back to normal after 30 minutes!
Am I misunderstanding anything here?
Based on my understanding, I really think the ability to set a solid temp basal that does not need the RL and having Loop actively running, would be a huge benefit.
Lemme know if I am misunderstanding, and again, @Trying I apologize for the tangent.
I agree, lots of issues have that have little to do with the original thread. I am adding my apologies to @Trying.
Correct.
Fully charged RL battery lasts around 36 hours. A standard routine is to just keep it plugged in overnight, just like a phone. In 3 years of using RL, I have never had RL battery die. Of course, it could die. iPhone could break. Pod itself could fail.
Yes, after 30 minutes, pod returns to normal basal.
Yes it does.
You do not need to look at RL ever. Phone and Watch give a clear indication of the Loop status. But, again, anything (including the pod itself) can fail - there is no way to guarantee 100% operation of any technology.
Correct.
You are not the only one. As I mentioned in a reply to @Trying, there is on open request to include PDM-style temp basal setting from Loop, so that Loop can serve as a complete PDM replacement. I can’t say anything on behalf of the developers, but given that the demand for this feature is high, I hope this will be included in the not so distant future.
I think I should also say that omnipod support in Loop is relatively new, and is not yet included in the mainstream Loop master. What people are using now could be called a “beta” version. Pete (and other developers) are polishing and fixing stuff as we speak.
For me, it would be an extremely reassuring thing to be able to know that no matter what happens with the RL or phone or the app or bluetooth or anything else, if I set a particular basal, that thing is locked in!!!
To me, that would be a great thing to have.
I am very glad to know this idea is being considered as an enhancement!
Do those scenarios sound reasonable? I mean, nothing too crazy or out of the realm of possibly, right?
As far as charging it…yes, in an ideal world, we all get to sleep by 10 pm, get 8-9 hours of sleep, drink our 8 glasses of water a day, eat only healthy well-balanced meals, and charge our RL every night…
But in the real world, the fact remains, you may not charge it when you are supposed to. And there is no charge indicator on the RL like you have on a phone or pretty much any other electronic device that needs charging. And if the RL charge dies without you being prepared, are you SCREWED for Override basal rates? I think that is a fair question.
Yes, you are screwed if RL battery dies, or if you forget RL at home, or if RL dies, or if iPhone runs out battery, or if iPhone dies, or if you forget iPhone at home, or if comms between iPhone and RL die, or if comms between RL and pod die, or if the pod dies, and these are all fair concerns and reasonable scenarios. Hopefully, a PDM-like temp setting feature will be implemented in Loop.
Back to this question:
Below are the code changes in BolusInterfaceController.swift to get crown increments of 0.05U and +/- increments of 0.1U.
) - that model could be behind the too-long zero temping.
