Hello.
I am designing my own chip, I haven't yet decided upon whether to use two parallel 18650 batteries or one 26650(I have found reviews that I actually trust since the person is open about exactly how he tests batteries and that he test all batteries in the same way, but the 26650 test is 1,5 years old now and some say they have improved. I don't know but I would like to be able to safely draw 40 or 50A) ether way I am designing a non-inverting 2-switch buck-boost converter that is going to regulate the 3,7V(4,2V fully charged - 3V fully discharged) into 5V-1V, and I will do as it is done in all regulated mods(I assume they are) and implement a DC-DC converter but with a 8-/16-bit microcontroller to close the feedback loop rather than analog components.
One of the main features that I want to accomplish better than any of the mods I've used is the temperature control(hence forth TC), to begin with I have the advantage due to the fact that I am not designing a commercial product that has to be profitable and since this is a hobby project I can do things that wouldn't be possible/suitable in a product made to be generating profits, such as using a higher speed, higher resolution ADC(Analog-Digital-Converter) and use a fancy dedicated instrumentation amplifier(or simply a differential amplifier) to measure the resistance directly at the 510 connector through two "sense" wires that are separate from the current carrying wires in order to come as close as I can to a 4-wire measurement system. And I plan to implement a thermocouple probe circuit(which will be using the systems separate digital temperature sensor to get an temperature value to base it's results on(that digital temp sensor is otherwise used to measure the temperature of the circuit board in an attempt to compensate for the large variation in temperature that the circuit will be exposed to)), the idea is that I should be able to connect a thermocouple probe to the mod and place it between the cotton and the coil in order to calibrate the TC algorithm, I also plan to implement a calibration routine for each atomizer to be used with the mod by ether shorting the posts of the atomizer or by connecting a known resistance so that I can calibrate out the atomizer+510 connector and be left with only the coil(s) resistance.
I think that would be a pretty neat setup to use to implement my own TC algorithm, but I have a problem.
I can't find any information what so ever about controlling or regulating a temperature of a heating element based upon estimations of the temperature derived from periodic resistance measurements.
I have found a little peace of text that reviled that professional mod makers may be using the resistance measurements to create a coefficient that is used with something else to derive a time constant for the coil(s) and then use that time constant to simulate the coil(s) heating/cooling cycle.
Which sounds surprisingly advanced, or am I simply uneducated, in any case I thought that I should write here to ask if anyone here knows anything about Temperature Control or knows of any resources online or has any ideas about the subject?
If I can't find any information about it I will just have to invent my own approach but it would be nice to avoid reinventing the wheel.
Regards
I am designing my own chip, I haven't yet decided upon whether to use two parallel 18650 batteries or one 26650(I have found reviews that I actually trust since the person is open about exactly how he tests batteries and that he test all batteries in the same way, but the 26650 test is 1,5 years old now and some say they have improved. I don't know but I would like to be able to safely draw 40 or 50A) ether way I am designing a non-inverting 2-switch buck-boost converter that is going to regulate the 3,7V(4,2V fully charged - 3V fully discharged) into 5V-1V, and I will do as it is done in all regulated mods(I assume they are) and implement a DC-DC converter but with a 8-/16-bit microcontroller to close the feedback loop rather than analog components.
One of the main features that I want to accomplish better than any of the mods I've used is the temperature control(hence forth TC), to begin with I have the advantage due to the fact that I am not designing a commercial product that has to be profitable and since this is a hobby project I can do things that wouldn't be possible/suitable in a product made to be generating profits, such as using a higher speed, higher resolution ADC(Analog-Digital-Converter) and use a fancy dedicated instrumentation amplifier(or simply a differential amplifier) to measure the resistance directly at the 510 connector through two "sense" wires that are separate from the current carrying wires in order to come as close as I can to a 4-wire measurement system. And I plan to implement a thermocouple probe circuit(which will be using the systems separate digital temperature sensor to get an temperature value to base it's results on(that digital temp sensor is otherwise used to measure the temperature of the circuit board in an attempt to compensate for the large variation in temperature that the circuit will be exposed to)), the idea is that I should be able to connect a thermocouple probe to the mod and place it between the cotton and the coil in order to calibrate the TC algorithm, I also plan to implement a calibration routine for each atomizer to be used with the mod by ether shorting the posts of the atomizer or by connecting a known resistance so that I can calibrate out the atomizer+510 connector and be left with only the coil(s) resistance.
I think that would be a pretty neat setup to use to implement my own TC algorithm, but I have a problem.
I can't find any information what so ever about controlling or regulating a temperature of a heating element based upon estimations of the temperature derived from periodic resistance measurements.
I have found a little peace of text that reviled that professional mod makers may be using the resistance measurements to create a coefficient that is used with something else to derive a time constant for the coil(s) and then use that time constant to simulate the coil(s) heating/cooling cycle.
Which sounds surprisingly advanced, or am I simply uneducated, in any case I thought that I should write here to ask if anyone here knows anything about Temperature Control or knows of any resources online or has any ideas about the subject?
If I can't find any information about it I will just have to invent my own approach but it would be nice to avoid reinventing the wheel.
Regards
