Temperatrue Sensor Circuit

[mihirshah100]

Hello

I am trying to design a temperature sensor circuit.
Below is a schematic of what I have done so far:


I am using an ATMEGA32 and in this design I cannot seem to find pins 10, 11, 31.
Could someone please explain to me why this is the case?

Thank you

 

[dr pepper]

I dont know atmel, but looking at the schem I'd say they were power and reset lines and not shown on the drawing.

 

[mihirshah100]

ohh ok - its because I need to connect components to it.
The image below shows what i mean:


I am not sure how i would connect them. (the pins are located near the top right of the
image)

 

[Dean Huster]

Whatever happened to electronics that it takes a microprocessor to do what we used to do in 1990 and earlier with a few simple components that didn't require software and a development system? Next, everyone will want a microprocessor buried in every light switch in their house and another in their hankie for help in picking their noses.

 

[misterT]

Pins 11 and 31 are connected to net "GND" and pin 10 is connected to net "VCC" by default. If you name Power Terminal "VCC" it will be connected to pin 10, and if you name Ground Terminal "GND" it will be connected to pins 11 and 31. Take a look at the schematics you posted (the blue schematics).. you'll see that pin 10 is connected to VCC and pins 11 and 31 are connected to ground.

Whatever happened to electronics that it takes a microprocessor to do what we used to do in 1990 and earlier with a few simple components that didn't require software and a development system? Next, everyone will want a microprocessor buried in every light switch in their house and another in their hankie for help in picking their noses.

The motivation is to learn to program and use microcontrollers and later do more complex things (like robotics) and maybe someday to get a job in embedded systems. I teach embedded systems and in the first class we make a LED blink using a 64 pin microcontroller running at 20 Mhz. You can't imagine how much they learn and how much fun it is.

 

[cowboybob]

Whatever happened to electronics that it takes a microprocessor to do what we used to do in 1990 and earlier with a few simple components that didn't require software and a development system? Next, everyone will want a microprocessor buried in every light switch in their house and another in their hankie for help in picking their noses.

Amen to that, Dean...

I've got an indoor "Digital Environmental Status" device (my name for it) that, for one, tells me the indoor temp to the tenth of a degree... Woohoo! Utterly wasted accuracy.

And it presumes to know the impending weather (without an outdoor sensor of any sort) by dispalying various images of the Sun, or partial clouds, or all clouds, etc.; apparently based on the INDOOR humidity .

Plus a whole bunch of other useless crap.

More often than not, if I'm even remotely curious, I check the outdoor bi-metal analog dial temperature reader outside my kitchen window and while I'm at it, look up at the sky for the weather pattern(s) of the moment.

If I'm really, deeply concerned, I'll even go outside and wander around for a bit...

I must admit, however, that the digital "Snot-O-Meter" hankie might have legs.

<EDIT> Seeing misterT's response, I realize that he's probably correct: the OP is looking for experience, and that's laudable.
My apologies to the OP for my flippant response (however honestly felt).

 

[Dean Huster]

As the first five steps (making an LED blink) of a 4583-step microcontroller course, that' s one thing. You learn step-by-step rather than starting out and using a µC to control all the traffic lights for two intersecting 4-lane streets complete with left- and right-turn lanes and sensors to vary light timing depending upon traffic load. I'm referring to the magazine articles I see that use controllers and their requisite programming hardware/software to make circuits that can be had with a 555 timer or a couple of counters. The end result of all of this is that we, as consumers, end up with absolutely EVERYTHING electronic that we own being totally non-repairable. It's come to that already with laptop computers, televisions, telephones, radios, CD and DVD players. Pocket scientific calculators led the way in one of the philosophies: They're all going to end up costing around $20. To sell John's version vs. Bob's, John programs 575 functions into the model he manufactures while Bob only programs in 157 functions into his top seller. Trouble is, no human being on earth is going to use more than 36 of those functions in his entire lifetime and of those, only 10 will be used on a recurring basis, and those only if it's being used beyond "4-banger" operation.

 

[misterT]

Agreed.. You have good points there Dean. Even with microcontrollers people are tempted to go with Arduino because "it is easy and fast and you don't need to learn C or Assembly". All this internet and media are making people impatient and lazy.. I feel it affecting me sometimes also.

My apologies to mihirshah100 also. This has nothing to do with your original question.

 

[misterT]

Pins 11 and 31 are connected to net "GND" and pin 10 is connected to net "VCC" by default. If you name Power Terminal "VCC" it will be connected to pin 10, and if you name Ground Terminal "GND" it will be connected to pins 11 and 31. Take a look at the schematics you posted (the blue schematics).. you'll see that pin 10 is connected to VCC and pins 11 and 31 are connected to ground.

Here is an image that further explains the above:

 

[Mike odom]

using a microcontroller to measure temperature alone would be a waste. To use it to monitor temperature and control some function (say tune an automotive engine for max performance) is probably still overkill. But to integrate temperture, 02 sensor, co sensor, rpm, etc... to give full performance can be done without a micro, but the circuitry necessary would be enormous compared to the simplicity of using a microcontroller to reduce parts count.

and to the 'one' that has the "Digital Environmental Status" device that he can't stand and wants to ***** about, hey, you're the one that bought it and has it in your house... unless of course it was a gift... so deal with it... You know, they have a channel on tv that tells you exactly what the weather outside is like, when I was growing up, we called that a window...

I built a digital tachometer in the 80s that used divide by n counter to convert the incoming frequency to a 2 rpm period. The a sequencer chip to use a 2K lookup table to convert this count (it counted 10KHz pules in the 2 rpm period) to read 4 bytes to drive 3 digits and a 20 LED bar graph. It didn't use a micro. Microprocessors were expensive and needed chip sets (clocks, drivers). Embedded controllers (Microcontrollers) were super expensive. If I were to redo that project today, I would use an embedded controller because not only would the parts count be less, I could buy said controller for less than the cost of the logic chips. It all has to do with cost and parts count. Many many things that would have been done discretely back in the 90s can now be done cheaper (and more reliably) with controllers than with discretes.

Bimetal strips are still used in thermostats (at least in my house) because they work fine, last a long time, and are cheap, But they can't turn the heat/ac down when you go to work and back on 1 hour before you get home, or down at night and up before you get up in the morning, saving you lots of $$$ automatically. For that, you need a micro. The vacuum advance system on a car engine worked well enough to be used for years, but was not perfect and couldn't reduce emissions, for that you need a micro... need I go on?

I just designed a 16 channel temperature to 4-20mA board for an industrial refrigeration controller. It would have saved my customer a lot more money had I put a micro on it and sent the data to the PLC serially, but that's not what he wanted. He wanted something simple and so I used opamps to drive 16 circuits individually for which he needs 16 input channels (actually 32, as he uses two cards per system) to the PLC, but this is what he wanted. Still it saved him thousands of dollars per design, instead of having to buy individual $300 sensors, and I delivered the boards at 1/4 the cost of an off the shelf board that was available. He could've bought the off the shelf boards and saved himself the $1500 development charge, but the R&D was paid for with the first system, all future systems are at a huge cost savings. And he has control of assembly of the boards, so he doesn't have to bow down to the builders and wait for parts. Sometimes it pays to reinvent the wheel. So not every system uses a micro, even though it would've made the board smaller and cheaper, he didn't want to have to reprogram his system software to use a micro.

 

[Dean Huster]

You're correct, Mike. I agree that micros are fine when they're controlling SYSTEMS, whether industrial controllers, automobiles, home thermostats, microwave ovens, etc. I suppose it's the hobby magazine projects that hack me off -- 6 projects in the magazine and 5 or 6 of them using processors, most of them doing either simple tasks more easily and economically handled by analog or simple digital circuits or just as badly, a complex operation that is of use only to the writer of the article.

Because I've spent a lot of hobby time harvesting components from PCBs over the last 50 years, I have lots and lots of parts on hand, but very little in the SMT or micro age. If I want to build a large frequency/time standard and clock system, I'm going to do it using analog, TTL and CMOS because I'll probably have every last component I need right here in storage. Yes, the device will be larger and use a LOT more electrical power. But it won't cost me hardly a dime to build it all. The design will be a maximum amount of fun for me. If one part goes bad, more than likely only a small part of the system will be affected and the rest will remain operational.

If I desired to build this same system or even a substandard variant of it using a processor, I'd have to buy ($$$) all the hardware and software for program development, a bunch of new chips that I dont have ($$$) and then spend a hundred hours learning this new software, both for the processor and the programming system. If one little thing goes bad, the entire system is likely to go down. And there's something about a processor, instruction set and program operation in a micro that instills a tremendous amount of distrust in that genre. If I use a program step to set propagation delay compensation, micro-and nanosecond timing synchronization and such things, I just don't trust something that has to go through several instructions after a command or interrupt to make a change. I certainly don't ever see synchronous operation in a micro, unless it has clock speeds that are 1000 times faster than the system being controlled.

 

[Mike odom]

Again, all boils down to cost....
dude, I use to spend whole afternoons at Delta Electronics in Amesbury Mass going through their 3-4 bulding floors of nothing but scavenged parts! You could get lost (time wise) going through all that stuff...

 

[gary350]

Wouldn't it be much easier to buy a Fuji programmable temperature controller and thermocouple.

**broken link removed**

**broken link removed**

 

[Mike odom]

Gary, you sound like a girl in my calculus class many years ago: "Why do we have to take calculus, we have computers"... and who will program those computers of tomorrow? And who will design those Fuji Controllers?
Life would be much easier if we could all just stay in bed and be served breakfast, but most of us have to work for a living... and to work, you first need to learn...

besides, why spend $20 when you can build it for $5????

 

[dr pepper]

With regards to the microprocessor comments.

I handed an apprentice my old analogue meter the other week and said 'test that for resistance', I got a totally blank look 'how do I...'.

 

[Mike odom]

we had a vtvm (vacuum tube voltmeter) in school, but we didn't like to use it as it had to warm up when you turned it on... so we always preferred the simpson 260 (of which I still have one)...

 

[Dean Huster]

Mike, it's the teacher's answer that's important here. A friend's son (a 4.0 student) asked a similar question of his high school teacher regarding some math exercise he was forced to go through. Her absolutely most stupid answer in the world was, "so that you can do more advanced math classes." As a teacher, I would always provide practical applications for exercises in math or electronics. If you can't come up with a half-way decent reason for doing something, then that something isn't worth doing, whether it's math, illegal drugs or cliff-jumping. That put's math in the same category as polital science, history or physical education. About the only thing you can make a living at with those types of degrees is teaching other that same claptrap.

I still have my VTVM, VOM, RC bridge and slide rule. All with analog scales, some logarithmic, some linear. And I can still use them all and do. Granted, the slide rule is more for funzies since an electronic calculator is much faster and less error-prone, but warm-up time or not, I still love my VTVM. Ever see a run-of-the-mill DMM measure over 20M ohms? A VTVM can measure up to 1000M ohms! There are still uses for "antiquated" electronics, especially after an EMP hits.

 

[dr pepper]

I feel a little better not being old enough to own a vtvm.
Maybe I should as I do have a small collection of valve or should I say tube radio's.
Nearly every one of my college books was as mentioned boring, on the first page a string of maths equations, all the 'interesting books' kept maths to a minimum and were not written as though the person reading allready knew everything being explained - whats the point in that - or is it the author proving that they know what they're on about.

 

[cowboybob]

Mike, it's the teacher's answer that's important here. A friend's son (a 4.0 student) asked a similar question of his high school teacher regarding some math exercise he was forced to go through. Her absolutely most stupid answer in the world was, "so that you can do more advanced math classes." As a teacher, I would always provide practical applications for exercises in math or electronics. If you can't come up with a half-way decent reason for doing something, then that something isn't worth doing, whether it's math, illegal drugs or cliff-jumping. That put's math in the same category as polital science, history or physical education. About the only thing you can make a living at with those types of degrees is teaching other that same claptrap.

I still have my VTVM, VOM, RC bridge and slide rule. All with analog scales, some logarithmic, some linear. And I can still use them all and do. Granted, the slide rule is more for funzies since an electronic calculator is much faster and less error-prone, but warm-up time or not, I still love my VTVM. Ever see a run-of-the-mill DMM measure over 20M ohms? A VTVM can measure up to 1000M ohms! There are still uses for "antiquated" electronics, especially after an EMP hits.

With you there 100%, Dean.