Setting ADC clock

[Patronus]

I'm new to PIC programming.
Can someone please tell me what are the factors that I should be concerned about when setting ADC and timer clock inputs. I have given a project to complete but this is not my study area. I'm learning everything from the beginning. I've written the code but I need it to be accurate.

Thanks in advance.
_________________
"Silence is argument carried out by other means"

[PCM programmer]

Post your PIC and the oscillator frequency for the PIC.

[Ttelmah]

and (of course), the old answer, look at the data sheet for your PIC.
There are normally examples there, showing what ADC clocks can be used for each processor clock.

Best Wishes

[Patronus]

hello PCM programmer, My PIC is 16F877A and frequency is 20 MHz.
Thanks both PCM Programmer and Ttelmah.

I read the datasheet.
According to that what I finally decided was, to satisfy the minimum Tad(1.6 us) I have to use 32 as the prescaler (setup_adc(ADC_CLOCK_DIV_32)).

Please tell me whether my decision is correct and tell me if I've missed anything.

And can you please help me with another thing. What we have got to do is to control the speed of a fan(AC motor with triac). Now I'm a bit comfortable with ADC converting part. Now I have to learn how to control AC motor with triac, from beginning. I am not responsible for making the circuit and all I've got to do is to write the code. Can you please tell me how should I begin. This is my first project and first PIC programming experience.

Thank you very much.
_________________
"Silence is argument carried out by other means"

[PCM programmer]

Here are two sample programs for using the adc on a 16F877A.
http://www.ccsinfo.com/forum/viewtopic.php?t=32168&start=1
http://www.ccsinfo.com/forum/viewtopic.php?t=40007&start=1
These programs are setup for a 4 MHz crystal, so to change it to
20MHz operation, you need to change fuse from XT to HS.
Also change the #use delay() to say 20MHz instead of 4MHz.
Also as you have noted, the setup_adc() must use ADC_CLOCK_DIV_32
for 20 MHz operation.

[Ttelmah]

Good Start.

[Patronus]

Thanks again.

I had already done this part. Actually what I asked was how to begin that AC motor controlling part. Apologize for not being clear.
But I had kept following lines which were given by IDE when I created the project.

[Mike Walne]

[Ttelmah]

Yes, you need to worry about fuses....
These control a whole suite of basic configurations/settings for the PIC, that have to be right, for the chip to work at all, and vary according to _your_ hardware. For instance you have one here that is probably wrong. RC. If again you look at the data sheet, and look at the oscillator settings (under 'special features of the CPU' - section 14.2). You have a choice of setting the oscillator to use 'LP' (low power crystal), 'XT' (standard crystal up to 4MHz). 'HS' ('high speed' crystal >4MHz to 25MHz), and 'RC' (simple resistor capacitor oscillator). You have this last one selected. Figure 14-3, shows the external circuit needed to make this work, but unless timing is _very_ unimportant, this is a rarely used setting. You almost certainly need to change this to 'HS' for a 20MHz crystal.
You don't need _any_ delay 'after read_adc'. You need one _before_. There must be a minimum of 'Tacq', after selecting an ADC channel, before reading it, _and_ between successive readings. The sample code is updating a value to a display, so delays, since the display can't update very quickly anyway... The delay is done after reading, so the display can be seen, but still occurs 'before' the next reading. To give the required delay _before_ the very first reading, there is an extra 20uSec delay before the loop is started. The Tacq, for your chip, is again in the data sheet. Equation 11-1. 10uSec is too quick. The _best_ for this chip, with a 'zero' source impedance feeding the converter, is 10.57uSec. Even with a 1KR source impedance, you need just under 12uSec.

Depends what you want to 'do' with the motor... Generally, with an AC motor, you can't control the speed. Depends on the motor design, but most 'true' AC motors, are effectively like stepper motors, with a series of coils round the motor, which are energised in turn by three phase electricity, and the magnet in the core 'chases' the electric field this produces. Their speed is dependant on how fast the field is moving, rather than on voltage, with the rotor slipping some distance behind the field. Then you have single phase versions of these, which basically won't turn until something makes the shaft turn - these use mechanical switch, or a capacitor, or a 'shaded pole' to introduce a little 'twist' into the magnetic field to get them started. Again though they are chasing a magnetic field, and you can't easily turn down the voltage to slow them. (they will control this way to some extent, as the slip increases as voltage drops). These are the commonest motors in existence, turning things like machine tools, fans, pumps, etc. etc.. The only really reliable way to make both these motor types adjust speed, is to actually synthesise your own AC electricity at different frequencies....

If you look at things where speed control is common, they are generally DC motors, which just happen to be being run off AC. Things like an electric drill, will have a rectifier, and feed DC to the motor. With these you can use the same type of circuit that is used to dim a light, to reduce the effective RMS voltage being delivered, and this then gives a lower DC voltage to the motor, adjusting the speed. For these, yes you can use a TRIAC dimmer. A search online will give you lots of details for simple circuits using these. These have the advantage of retaining torque as the speed drops (this is the big problem with slowing single phase motors).

CAVEAT (big caveat....). Make very sure you know what you are doing, if you are starting to try to control AC mains. This is potentially _dangerous_.

TRIAC's are really quite nasty devices to work with, if you are not going to make RF interference at massive levels (early light dimmers were great at this). You really need to be using 'phase angle' control, rather than PWM. Here you synchronise your code to the zero crossing point of the AC, and adjust 'when' after this you turn on the power. Remember the hardware CCP module, as well as offering 'PWM', offers precision timing functions, and these are what is needed here.

There are several 'application notes' on the Microchip site about using the PIC to control TRIAC's, and these may help.

Best Wishes

[Mike Walne]

[Ttelmah]

[Patronus]

Thank you Mike Walne and Ttelmah.

Ttelmah, Our project is to count the number of people entering a room, display it in a lcd display. And control the fan speed according the temperature inside the room(fan should be turned off if the room is empty). The temperature should also be displayed. We are counting people using interrupts.

Circuit making is done by two other friends in the group. Still I don't know how things happen in the circuit. All I know is these these inputs are given to the pic and these should be done by my code. Simply, I don't know hardware.. Well.. not a good way at all but due to some reasons we've got to do it this way. I've coded for people counting, displaying it and displaying temperature which is calculated according to the input at A0 pin. I simulated this code using Oshonsoft simulator. Unfortunately with all those problems in the code, I got the expected result... after having 3,4 days in the hell, learning word by word from the beginning.

Now the problem is controlling the fan. My question is, do I need to learn the whole motor controlling circuit to write the code ? If I just learn what's the pic's role here will I be able to do it?
This has to be finished within less than two weeks but at the moment there's no way of meeting my mates who make the circuit. And we've got no time to work only on this.. May be this is so funny sir but it's like that.. Thank you very much for your help..
_________________
"Silence is argument carried out by other means"

[Ttelmah]

You need to know what the fan control board 'does'. Hopefully if someone else has/is building it who understands what a processor is capable of, the control input will be something like a voltage input from 0 to 5v (in which case you can drive it via an integrated PWM output), or some form of other control input, in which case they can tell you 'what voltage it needs', 'what frequency it needs', etc. etc.. You need _data_. Data is always your friend in electronics. Without knowing what sort of control input is involved, you can go nowhere. If is is (say) the gate of a TRIAC, then you will need to know how it is isolated, or provide isolation yourself, and realistically will need a zero crossing detection 'output' from the supply feeding the fan.

Best Wishes

[Patronus]

Ok sir. Thank you very much.. Yes they know it. I'll get data from them and try to write the code. After writing that part and simulating, if it causes problems I hope you'll help me.
_________________
"Silence is argument carried out by other means"