I am designing a disposable baby blanket for premature babies that will plug into an Arduino.
It will have four layers:
Outside, paper/cloth microfiber
Mylar
Heating Fabric (5 cm x 10.4 cm)
Inside, paper/cloth microfiber
I am going to use (3) x Thermistor sensors diagonally, between the Heating Fabric and the inside layer. They are supper small, made any smaller and this design could go right into production if someone had the proper motivation.
My code and diagram for the thermometer are solid. I'm using "Lesson 23 Thermometer" it's came with the Mega 2560 Project kit. I have already set it up a few times to run some test.
My problem is I can't figure out exactly what code I need to add to set a temperature, and how I'm going to turn the Heating Fabric on and off. The spec sheet for the fabric indicates that it will eventually heat up to 40C/104F with a 5 volt input, at 7.4 volts it supposed to heat up to 169*F, so I think I'll probably stick with the five volts.
(Note: I can only use parts on hand at this time.)
I am sure you can/will get good guidance on the legal issues. What country might this be used in?
Meanwhile back at Ohms Law etc:
You seem to have:
a 6.7 Ohm resistor with a thermal time constant of around 2 minutes
a 5.0V DC power supply, presumably safety rated for this application
An Arduino Mega (Or UNO or Nano etc) [There seems no reason a Mega is needed]
Some type of small bead-type? thermistors
Some type of human interface
You need to talk to people who develop medical electronics and are familiar with the electrical and testing requirements. Maybe they will appear here, or you need to search on that subject. I can tell you that "Research Hospitals" do "evaluations" (Not "Experiments") on patients using devices that are not yet approved for sale for specific uses. One of my sons is an Engineer who was (quote) "A little appalled but damn careful" in positioning a protoype Intercranial Pressure Monitor" in the OR on a live patient. His employer eventually did get an instrument approved.
Added: You might start researching NeoNatal information at "Cochrane Neonatal" Here: Our Reviews | Cochrane Neonatal One of my daughters-in-law is a study researcher here in Vermont uSA.
A major consideration is control of the power to the heater. Please research "Fibre Optic Isolation" to pulse-width control of a power Field Effect Transistor.
You are at the familiar point in the design process at which you are creating questions faster than answers.
Just make sure as you progress you keep track of ALL the questions.
Maintaining proper body temperature is one only one major problem premature babies are facing in NICU's every day around the world. Most of these babies are rewarmed with infrared radiation pointing down at them. From an engineering perspective this is a horrible idea because heat rises, in-turn these babies are very hot on the side that is facing the lamp, cold on the other side, and turned by hand often to prevent burning. This also exacerbates another problem these babies face every day, they dry out easy. This is very painful for the babies as their skin becomes more susceptible to tears.
Although I have no plans of ever using this on a real baby, my prototype is only 5 cm x 10 cm, a working product would need to be 12" x 18". It is a problem that I am trying to solve for because these babies need it. In mid December I will have a few big companies looking at my idea, if some company picks it up and decides to use it in an American hospital, I'm sure it will be redesign to plug into an updated version of an incubator, running there own proprietary software and it will meet all Government Regulations. if they like it they will take it and redesign it within Government Regulations I'm sure.
My design concept is on page one of this document:
Thanks to everyone who cared enough to have a look and thanks for making me smile after no sleep in two days.
I was thinking of using PWM, because I think I want to operate between +-0.5*, and I've already done a little code using PWM to turn on LCDs.
Do you think a PWM will work on a relay? I have done some testing with 3.3V and 5V, using the Power Supply Module on the end of a 830 Tie-Points breadboard. I still have to calibrate my thermometer, but I think I can get enough amps for the heating pad to work at 5V. I'll be doing more test this week on the heating pad a 3.3, 5, and 9 volts. Note: I've already done a little testing at 9V off the end of the 9V 1A Adapter and it got very hot very quick.
Anyways I didn't expect to get so many replies to fast, I still have to knock out my Material Science and Numerical Analysis, so it'll be a few days before I can let you know what I find.
This is a nice idea but I think you should step back and research the safety requirements for medical devices before proceeding with the mindset of marketing this. You're attempting to design equipment that might be categorized as life-sustaining/supporting. A failure of this equipment may lead to burns, hypo- or hyperthermia which could result in serious injury or death of the preemie.
I designed medical equipment (minimally invasive surgical equipment) for ~16 years and can offer a few thoughts.
First, obtain and read a copy of IEC60601-1.
Understand the notion of the risk management process (hazards assessment, risk analysis, risk reduction, residual risk etc.) Risk can come from many sources, including a hardware fault (e.g. a broken thermister, connecting wire etc), software (code hangs somewhere), external (damage to the heater from something having fallen on the blanket etc) to electromagnetic compatability issues (ESD, RF fields, power magnetic fields etc.) It's a pretty arduous process to fully analyze all potential vectors of failure in a system, the risks and the ways to reduce the chances of someone being hurt as a result.
You should investigate the standards regarding software used a medical equipment (IEC 62304 for example.)
You should check to see if the manufacturer of the processor used on the Arduino platform authorizes its use in medical equipment (not all do.) I'm pretty sure the Arduino environment is not.
No. Not a regular electro-mechanical relay. They are on/off. A solid-state DC relay will work with PWM. (AC solid state relays will not.) Or, with DC you can also use a fairly-simple MOSFET driver.
But... Very-few heating/cooing systems use PWM or continuously-variable heat. Your home furnace switches on & off. Your refrigerator switches on & off. That works because temperature can't change instantly anyway. And it can be difficult to make a stable feedback system, especially with the delay. Engineers do Laplace Transform analysis to evaluate and adjust the stability of feedback systems.
Just for reference - A regular heating pad usually has no feedback. There are 2 heating elements for low & medium with both turned-on for high (exactly like a 3-way light bulb).
A regular electric blanket normally uses a "trick" of putting a small heating unit (a resistor) inside the controller which approximately follows the heat of the blanket. i.e. If the room is cold there is a longer on-cycle. Or, it might just have a timer (like very-slow PWM) with no feedback.
In this case, it might be better to monitor the baby's body temperature. I would assume they are already monitoring vital signs?
Although I have no plans of ever using this on a real baby
Good!
In mid December I will have a few big companies looking at my idea
They probably don't need your help or your prototype.
rkilpatrickii:
In mid December I will have a few big companies looking at my idea,
DVDdoug:
They probably don't need your help or your prototype.
Well, of course, that is the reality of the situation.
Absolutely nothing in the OP is in any way original or non-intuitive separately or in combination, so any company with an interest in development would have already progressed well beyond this discussion. It would not - or in a sane system should not - be patent-able.
An interesting muse on certification. On my desk (one of them ) are sitting a few Dallas DS18B20s salvaged from medical disposables. So does Dallas Semiconductor warrant these for medical use? Well, their disclaimer:
Maxim Integrated cannot assume responsibility for use of any circuitry other than circuitry entirely embodied in a Maxim Integrated product. No circuit patent licenses are implied. Maxim Integrated reserves the right to change the circuitry and specifications without notice at any time.
Does not specifically exclude "mission critical" purposes like the Atmel/ Microchip one, but would seem to be sufficiently general to do so.
A manufacturer of such equipment not only can employ components with no warranty, but necessarily must do so since they are using hundreds of components. It is clearly up to the manufacturer to take responsibility.
K++
) are sitting a few Dallas DS18B20s salvaged from medical disposables. So does Dallas Semiconductor warrant these for medical use? Well, their disclaimer: