to the rescue…

Unfortunately I currently have a not so nice Sensor, again:20160930_095806I tried to get this sensor replaced by Abbott, but failed because of a to little Deviation.. I could push my sugar well over 180 stable to get the required deviation… But anyway, that’s another topic.

The sensor had worked very smoothly for the first 8 days, very accurate and stable. Now, thanks to the correction functions in xDrip plus and the blueReader it now works smoothly again:screen-shot-2016-09-30-at-09-59-05The values are fitting very well with my other Glucose meters. Multiple times a day I look for discrepancies with my Aviva Combo or Libre trough bloody measurements.

Please support the further development of the blueReader with the campaign.

and the Band plays on…

As I already announced: the startNext campaign  goes on! Until the 17th October everybody has still the opportunity to support the campaign!

Please share and like the project! According to the number of visits only half all visitors need to contribute € 10 and the blueReader would already be funded!

What else is new? The prototype, with alternate firmware, runs and runs. After the first 9 hours of continuous operation it worked yesterday 16 hours without any problems. As I have yet to implement the monitoring of the battery, the battery was discharged very deeply after the 9 hours and the prototype had to spend the last night on a proper charger.

Now the blue reader is running again for over 15 hours and I’m curious how long it keeps sending me values …


Two common questions have arisen:

  • Will the sensor not be damaged it is repeatedly scanned?
    No, after all our knowledge and understanding of the built-chip, the sensor draws the transmissionpower from the nfc-reader. Simply put, you can say it like this: The chip consists of two parts, one measures every minute the Glucose value and does all the pre-evaluation. This current value is written in a memory where all the past values from the last 16 minutes are. In addition, it writes every 15 minutes a historical value into the memory. These are the values which the readers retrieves and from which it generates the curve of the last 8 hours. The other part of the chip for communication with the reader, or in our case the blueReader. This part is “activated” by the electric field, which is build up by the reader. This electric field is not only for the detection of the sensor, it also provides it with power and by “disturbances” of the field the actual data is transmitted.
  • How do you attach the blueReader at the sensor?
    I don’t attach anything at the sensor. The chance is much to high to accidentally rip off the sensor! I still have to pay each sensor out of my pocket, because my health insurance does not consider this system to be necessary…
  • But how are you attaching the blueReader now?
    I’ve secured the first prototypes mainly with Velcro (an industrial type, which consists of Velcro and fabric with adhesive surfaces).I put a plaster or wound dressing over the sensor, one side of the Velcro on it, the other side onto the antenna of the blueReader. The disadvantage of this is that is stick really firm and I’ve ripped off a lot of platters, while trying to put off the prototypes.
    Currently I just use a tape strip, like that stuff that is usually used after the quarterly blood analysis to hold the swab on the spot. You can see Pictures of this in the previous post.

On the technical side, I prepare currently the first version of the “real” blueReader firmware. This will have to do without a few features:

  • Encryption
  • Changeable name
  • Changeable connection pin
  • Softwareupdate via Bluetooth

However, until the final version is needed, there is still some time. I am now focusing on primary and important features to move forward more quickly and to be able to provide access to the sources as quickly as possible to all interested parties.

Incidentally, I also want to draw your attention to a great project: openAPS

What is it now? One of the many ways how you may want to use a blueReader!

Shortly after I began to occupy myself with the continuous monitoring of my blood sugar using the Libre sensor, I came across a bunch of diabetics, very willing to experiment. This international group of diabetics uses data from all possible available CGM systems with only one goal: to be able to use a pump as an artificial pancreas!

Of course not like our real pancreas, because of sugar, especially the interstitial glucose, which increases only delayed, especially if we have eaten well. However, most of the time of the day we eat nothing and yet our pump can not always respond to our values. We plunge into hypoglycemia because we forgot to reduce the basal rate or forgot to take extra carbohydrates to us. We go high because we are upset, stressed out, having an illness, or simply did not properly assessed carbohydrates.

It is at these points such a system can help! Currently, I asses what my own openAPS system, based on a Medtronic pump 522 and the blueReader prototypes, can do! Currently, the pump, still filled with water and not connected to me, sits next to me, and I look at what the system is doing and why, but in the next few days I will venture into the first experiments on myself 🙂 Until now, the predicted values, what I see the with the help of  blueReader and the real, by blood measurements, are looking very good!

Anyone interested in something like this should definitely go here: into the channel and ask for more information 🙂

… And just now, the blueReader has powered down, again. After 16 hours of continuous operation, where it, very stable, transferred every 5 minutes a value. Tomorrow I will continue to work on the energy-saving options and see if I do not still get it in this one well over 20 hours runtime…

A long wish list for the next prototype has already came out of all the Experiments so far. I hope I can discuss them with my electrical engineer tomorrow or early next week…

Stay tuned!



Tonight, the blueReader has wonderfully delivered a lot data. From last night 8:30pm till this morning at 5:30 almost every minute a value has been reported, only two disturbances occurred:

  • The blueReader lost twice  the sensor.
  • The software had one crash.

Actually, I think quite a solid result for the first test! (And an almost beautiful values the whole night …  Just one hypoglycemia alert at about 12 o’clock…)


Today I will change a bit the software, especially to make it more energy-efficient and I will change the readout period from every minute to every 5 minutes. Every minute makes no real sense, usually the sugar changes not that fast.

If my calculations are correct, it should bring the time from the observed 9h to well over 24h … I hope I can tell tomorrow …

at the eleventh hour…

Probably some might ask what’s going on. The campaign is almost over and no new information?!? Why?!?

The answer is quite simple: I had huge problems with the project! As I already said in the last post, I can not longer use the mbed system and had to cope with gcc and a pure open source setup. That was quite complicated and tedious, but worked at a certain point.

Unfortunately, no reason to cheer 🙁 Still nothing, of what I had previously written for the prototype, worked with the new environment…

But with a lot of research, even more cursing, coffee and many experiments, I was able to approach the very different issues and a resolve one by one!

Today it finally happened:

img_6488 img_6489 img_6492 img_6495 img_6497 img_6501The prototype could finally be packaged in a housing again and Testing started …

img_6506And the prototype provided data! The app shown is xDrip plus with a few minor modifications be able to display the blueReader data without its own app or library . The blueReader runs an adaptation of Jörn LimiTTer software and uploads the data into my personal NightScout instalment.

Now, somebody might think: Did she not promised to develop a blueReader app? Yes I did and i plan to do so, or maybe extensions to xDrip and a corresponding iOS software to view the data without NightScout and co and to generate proper reports! But for now and due the lack of time and I have chosen the fastest alternative for now, to be able to see if the prototype is now running well.

Since it was impractical to use with all the wires sticking out, I wrapped it again:

img_6510 img_6511 img_6514 img_6515And now I am curious if it is reliably and can deliver all the values of this night.

Ho will the blueReader project continue?

  1. The campaign is almost over, and unfortunately not half of the money is yet there. I am in talks with  Startnext, if we can extend the campaign, as I think now, with the working prototype, it might get more interesting for more supporters. Secondly, I am discussing options with my engineer about how we can further save money.
  2. We are preparing another prototype. This one will be smaller and more flat, because we have an opportunity to integrate the antenna for the charger into the motherboard. In addition, we may have a solution to integrate the NFC antenna as a flexible part, sticking out the motherboard. However, there are still a few experiments and information needed to decide on this.
  3. With the current, running software, I will hopefully proceed more quickly in my schedule and might have the opportunity to release quick updates to the functions and apps. The next targets are now:
    1. Integration of the reminding Hardware of the blueReaders (currently, the off button and batteries status are not yet integrated into the software).
    2. Editing and saving of Settings within the blueReader.
    3. libBlueReader for iOS.
    4. libLibre for iOS.
    5. An app with basic functionality for iOS.
    6. Android-Version of the software.

Still, a lot to do 🙂

Thanks with love to all current supporters! You are great! Please keep advertising for this project, and lets hope we can make it through!

For all interested people: Here it goes to the support page…

Do you have any questions? Do you need clarifications or do you have suggestions? Please write me! Either here, in my comments, or via Facebook, Twitter or email. I try to answer all requests as soon as possible 🙂


Today, almost half of the time for the Campaign has passed. 12 of 25 days are gone, and every day at least one person has pledged their support!

Thank you!!!

Unfortunately we are not yet at half of the goal, but perhaps all together we can bring this campaign through to the end!

What did I do in the meantime?

  • I had a lot trouble with the firmware. Originally we planned to test the current prototype this week extensivly. Especially we looked forward to see how long the battery holds out whether we can substain the desired 40 hours runtime, and if we can possibly use a smaller battery. What was so difficult about it? mied (The Plattform I wrote the Firmware with until now) changed a lot in the compiler system and it simply dös not work anymore. Then I experimented way too long before I realized that this probably won`t have a foreseeable future … Bad for the project, but at least a oppurtunity to switch to the openSource workflow with the gcc_arm compiler. Thus, I can then provide a virtual machine for any interested person to create their own firmware later. I think I will finish the Sporting to the new compiler this weekend and start with the delayed tests by next week. If this works out, then I can finally make the wisset for live video, with a sensor on my arm, a blueReader at the sensor and my phone which displays the values!
  • At least I used some time lost on the software for creating a new housing for the prototype:
    First, a draft of the prototype hardware in a 3D-Tool:
    Afterwards directly to the 3D printer, unfortunately with little success: img_0090
    The measurements did not fit quite…img_0091
    And another attempt, this time with generous amounts of material, to get the walls not to thin again:img_0094 img_0096 img_0102 img_0101
    That is the current prototype, and what you need to know about it:
  1. The board is still larger than really required because a few millimeters in width can be ommited, once we arrange the components slightly differently and replace the cable eyes for initial programming of the system with contacts.
  2. The front part with the antennas for the NFC communication and the wireless charger is much too thick, currently both antennas require together around 3mm in height. But we have prepared much thinner and flexible antennas for the next iteration.
  3. The length could also be reduced even by up to 4mm, as currently the converter for the wireless power receiver is still an extra board in front of the motherboard.
  4. The wall thickness of silicon will be significantly lower. Because of the stability in the 3D printing I had to make all the walls at least 1mm thick.
  5. The cables, which are still hanging out of the case, are there to operate the system without using the wireless charger, that’s only for the current development phase.

At least, I have now a case which protects the system rudimentary. It is not blue (as you would expect in blueReader), but my pinkReader 😛

I hope that this project can continue and you all diligently spread the word about this project!


Questions and answers…

Since there were some reoccuring questions about this project, I tried to list the most frequently asked questions with their answers: blue reader questions and answers

About 20% percent have already been raised! I hope that the remaining almost 80% are going to come, too 🙂 However, it’s insanely great to see that you want to support this project! Many thanks to all current and future supporters!

What are the news from the prototype? There are some! I am currently working deep within the software for the blueReader. Next week are more functional test scheduled and hopefully I can share some of the first images of a complete prototype. I could indeed now, but it still looks a bit wild 🙂

Additionally I can tell you already the following: The experiments with the wireless charging system were a great success! The system works as we had hoped for it 🙂

The posts are now published on Facebook on its own page for the blueReader:



Over 10% of target are already collected on within the first day.

I would never had expected such a response !

Thanks to all supporters!

Up to the end of the campaign, I will continue to try to report regularly on the progress in the development.

Yesterday, the first units of the wireless charging system have arrived. As you can see the tiny board inserts exactly into the system and the antenna is not too large.

Now some tests are needed to see how Powerful this system is and if nothing blocks the usage of these components.

The current goal is to finish a complete board, with battery, charger and NFC Antenne within the week and test it thoroughly then.

After that, it is mostly about the further development of the software and to some small details to the hardware.

If everything fits nicely, we will hopefully have the first complete device ready for embedding into the silicon next week. With that we will hopefully get a better understanding of the possible dimensions and the shape of the housing!