Analog-to-Digital Converter Code

[Kevin Watson]

For those who might be interested, I've just posted new analog-to-digital converter interface code for IFIs robot controllers. A few advantages of using this software include parallel, non-blocking operation of the ADC hardware, precise control over the sampling period, which is important for signal processing and control applications and the ability to make measurements with greater precision using a technique called oversampling. The code can be found here: http://kevin.org/frc. As always, if you find a bug in the code or a problem with the documentation, please let me know.

-Kevin

Here's the readme file:

 
The source code in adc.c/.h contains software to setup and run
your robot controllers analog-to-digital converter in the 
background. A few advantages of using this software include 
parallel, non-blocking operation of the ADC hardware, precise 
control over the sampling period, which is important for signal 
processing and control applications and the ability to make more 
precise measurements using oversampling. Number of channels, 
sample rate and oversampling parameters can be changed by 
editing adc.h and re-compiling.
 
This software makes the assumption that when it's running, it 
solely controls the analog to digital conversion hardware.
 
WARNING: The higher sampling rates can place quite a load on
the robot controllers CPU. If you get the red-light-of-death
or notice any general wackiness in the operation of your
robot, first try a lower sampling rate to determine if this
is the problem. In general, you should select the lowest
sampling rate that meets your performance criteria.
 
This source code will work with the Robovation (A/K/A EDU-RC) 
robot controller and the FIRST Robotics robot controller.
 
The included project files were built with MPLAB version 7.20.
If your version of MPLAB complains about the project version, 
the best thing to do is just create a new project with MPLAB's 
project wizard. Include every file except: FRC_alltimers.lib 
and ifi_alltimers.lib and you should be able to build the code.
 
****************************************************************
 
Eight things must be done before this software will work 
correctly with your robot controller:
 
1) A #include statement for the adc.h header file must be 
included at the beginning of each source file that calls the 
functions in adc.c. The statement should look like this: 
#include "adc.h".
 
2) Initialize_ADC() must be called from user_routines.c/
User_Initialization().
 
3) On the EDU-RC, all analog inputs must be configured as 
INPUTs in user_routines.c/User_Initialization().
 
4) The call to Set_Number_of_Analog_Channels() must be removed
or commented out in user_routines.c/User_Initialization().
 
5) The timer 2 interrupt handler, Timer_2_Int_Handler(), and
ADC interrupt handler, ADC_Int_Handler(), must be installed in
User_Routines_Fast.c/InterruptHandlerLow(). See the included 
copy of User_Routines_Fast.c to see how this is done.
 
6) Define the number of analog channels that you'd like this
software to track by opening adc.h and following the embedded
instructions above #define NUM_ADC_CHANNELS.
 
7) For advanced users, analog channels can be oversampled to
decrease noise and gain resolution in your analog measurements.
The oversampling ratio can be changed from the default x16
by commenting-out the line #define ADC_SAMPLES_PER_UPDATE_16
found in adc.h and then removing the // from in front of one
of the other options. Measurement range and resolution can be 
determined from this table:
 
ADC Samples Effective
Averaged	 Bits of	Measurement Voltage
Per Update Resolution	 Range	 Per Bit
___________ __________ ___________ _________
	 1		 10		0-1023	 4.88 mV
	 2		 10		0-1023	 4.88 mV
	 4		 11		0-2047	 2.44 mV
	 8		 11		0-2047	 2.44 mV
	16		 12		0-4095	 1.22 mV
	32		 12		0-4095	 1.22 mV
	64		 13		0-8191	 610 uV
128		 13		0-8191	 610 uV
256		 14		0-16383	 305 uV
 
8) Finally, pick the master sample rate by selecting one
of the available rates found in adc.h. The update rate can
be determined using this formula:
Update Rate = 
	Sample Rate / (Samples Per Update * Number Of Channels)
 
****************************************************************
 
Here's a description of the functions in adc.c:
 
Get_ADC_Result_Count()
This function returns an unsigned char containing the number 
of ADC updates generated since Reset_ADC_Result_Count() or 
Initialize_ADC() were called. Use this function to determine
when fresh analog to digital conversion data is ready.
 
Reset_ADC_Result_Count()
This function can be called to reset the ADC update counter
to zero.
 
Get_ADC_Result()
This function can be called to retreive analog to digital
conversion results. Call this function with an unsigned char
containing the ADC channel you'd like data from. This function
will return an unsigned int with the last conversion result.
 
Convert_ADC_to_mV()
This function converts the value generated by the ADC to
millivolts.
 
Initialize_ADC()
This function initializes the ADC hardware and software. It 
should be called from user_routines.c/User_Initialization().
Once called, analog to digital conversions will automatically
take place in the background.
 
Disable_ADC()
This function disables the ADC hardware and software. To
minimize the load on the microcontroller, this function should 
be called when the ADC functionality is no longer needed.
 
Initialize_Timer_2()
This function initializes and starts timer 2, which initiates
a new analog to digital conversion each time the timer 2
interrupt service routine is called. This is called by
Initialize_ADC() above and shouldn't be called directly.
 
Disable_Timer_2()
This function is called from Disable_ADC() to disable timer 2 
and shouldn't be called directly.
 
Timer_2_Int_Handler()
This function is automatically called when timer 2 causes an
interrupt and shouldn't be called directly. New analog to
digital conversions are initiated from this function.
 
ADC_Int_Handler()
This function is automatically called when new analog to
digital conversion data is ready and shouldn't be called
directly.

[mechanicalbrain]

wow thanks allot!

[Kevin Watson]

I've updated the ADC code to version 0.3. The new version uses the "special event trigger" functionality of the CCP2 module to automatically start an analog to digital conversion without software intervention. This new version generates half of the interrupts/sec of version 0.2. As always, the code can be found here: http://kevin.org/frc

-Kevin

[Kevin Watson]

Well, I finally posted the ADC and Gyro code for the 2006 robot controller. Sorry about the delay, but I ran into a nasty bug that caused me to spend a few days in the fourth and fifth levels of programmer hell. The bug was in the design of the PIC18F8722 (yes, a bug in the sillicon) and I had to figure out what was going on. Anyway, the gyro code works really, really well with the gyro in the kit of parts. The accelerometers also work really well, but I need to do more testing before I unleash that code.


-Kevin

[Joel J]

Let's suppose there are four analog signals being sampled and "buffered" by your code. The gyro is running on ADC_CH0 and the other three channels are for other devices. How do we know when the gyro specifically has new data available? I saw the ~update_count~ variable, but it seemed to just indicate when any new sample is available, not when the sample for a specific channel is. Am I going to have to modify your code (maybe add an array of update_counts, with each index correspondent with the channel's freshness of data?) to allow this functionality, or is it already there and I'm just not seeing it?

[Kevin Watson]

Originally Posted by Joel J.

Let's suppose there are four analog signals being sampled and "buffered" by your code. The gyro is running on ADC_CH0 and the other three channels are for other devices. How do we know when the gyro specifically has new data available? I saw the ~update_count~ variable, but it seemed to just indicate when any new sample is available, not when the sample for a specific channel is. Am I going to have to modify your code (maybe add an array of update_counts, with each index correspondent with the channel's freshness of data?) to allow this functionality, or is it already there and I'm just not seeing it?

The update flag only gets incremented after all enabled ADC channels have been refreshed. You can get less latency (if that's what you're after) by doing the modification you described, or by just moving the gyro to the last enabled channel and continue using Get_ADC_Result_Count().

-Kevin

[PhilBot]

Hi Kevin...

Excellent code. I'm trying to get some early Auto-Heading code testing done using an existing Robot. I was hoping I could use the team's little 2003 EDU robot as a test platform.

I'm an experienced embedded programmer, but new to the IFI hardware.

I tried deploying your gyro code on the small EDU controller. I discovered that when I included the new ADC module I got a program run error light, so I dropped back to the standard iti_utilities code for the ADC conversion. (I assumed a basic hardware incompatibility....)

My only problem is that now I'm not sure where to put the call to Process_Gyro_Data().

I'd like to put it in User_Routines_Fast.c but I don't have your ADC code installed so I don't know when the converters have got new values. D'oh.

Am I flogging a dead horse here, or is there a simple way to use the standard IFI analog utilities to drive your Gyro code? I'm only using a single analog channel (for the Gyro) so that may simplify things. I don't need fully calibrated data for my initial testing... I just need it to basically run. If I can get the calls IN, I can get the rest to work.

Thanks for your time.
Phil.

Originally Posted by Kevin Watson

Well, I finally posted the ADC and Gyro code for the 2006 robot controller. .

-Kevin

[Kevin Watson]

Originally Posted by PhilBot

Hi Kevin...

Excellent code. I'm trying to get some early Auto-Heading code testing done using an existing Robot. I was hoping I could use the team's little 2003 EDU robot as a test platform.

I'm an experienced embedded programmer, but new to the IFI hardware.

I tried deploying your gyro code on the small EDU controller. I discovered that when I included the new ADC module I got a program run error light, so I dropped back to the standard iti_utilities code for the ADC conversion. (I assumed a basic hardware incompatibility....)

My only problem is that now I'm not sure where to put the call to Process_Gyro_Data().

I'd like to put it in User_Routines_Fast.c but I don't have your ADC code installed so I don't know when the converters have got new values. D'oh.

Am I flogging a dead horse here, or is there a simple way to use the standard IFI analog utilities to drive your Gyro code? I'm only using a single analog channel (for the Gyro) so that may simplify things. I don't need fully calibrated data for my initial testing... I just need it to basically run. If I can get the calls IN, I can get the rest to work.

Thanks for your time.
Phil.

Phil,

Gyro code for the EDU-RC can be found on my website.

-Kevin

[PhilBot]

Oops.

I'm in "drinking from a fire-hose" mode and never seem to look in the right place for information. Just when I think I've downloaded enough... I find more.. After posting, I discovered your adc_readme file which let me go through and add your stuff to our teem's prior EDU-RC code. Now I don't get an error & the gyro seems to be working as with the 2006RC.

Now with this new information I can double check against your webiste archives if need be.

Thanks.
I'll look harder next time.
Phil

Originally Posted by Kevin Watson

Phil,

Gyro code for the EDU-RC can be found on my website.

-Kevin

[Dave75]

Originally Posted by Kevin Watson

Well, I finally posted the ADC and Gyro code for the 2006 robot controller. ...

Kevin,

The adc_readme.txt says to remove the call to Set_Number_of_Analog_Channels() in user_routines.c. But the user_routines.c file included in the same zip (frc_adc.zip) still has it in. Same situation in the user_routines.c file in the frc_gyro.zip.

So, should the call remain as is, or be commented out?

Thanks

[Kevin Watson]

Originally Posted by spiff

Kevin,

The adc_readme.txt says to remove the call to Set_Number_of_Analog_Channels() in user_routines.c. But the user_routines.c file included in the same zip (frc_adc.zip) still has it in. Same situation in the user_routines.c file in the frc_gyro.zip.

So, should the call remain as is, or be commented out?

Thanks

Hmmm... I guess I should follow my own instructions <grin>. Actually, it doesn't matter as long as the call to Initialize_ADC() happens after the call to Set_Number_of_Analog_Channels() (it does in all the ADC-related code I've posted). If you're happy with my code, I'd just remove the call to Set_Number_of_Analog_Channels(), otherwise just comment it out in case you want to fall back to IFI's code.

-Kevin

[Chriszuma]

Okay, I know this is getting kind of abstract, and not truly necessary to a "black box" understanding of this code, but how exactly does oversampling work? I looked it up on wikipedia, and all i got was more confused. I understand that is is supposed to give more precision, but I really don't understand how.

Thanks.

[Alan Anderson]

Oversampling is essentially taking a number of samples and averaging them. If you average four samples, you get the equivalent of two more bits of resolution, but it takes four times as long to get the value. It trades time for precision.

[Chriszuma]

so then does it take values 4x faster? It seems like taking the average over 4 program cycles would make it less precise.

[Joel J]

Yes, it samples faster than the rate the data is being sent.

I have another question for Kevin (of course, I'm trying to optimize everything, but I don't want to make a bad move while doing this):

Is there a specific reason for switching from doing both the initiation of and saving of an ADC sample in the timer interrupt? I remember in last year's gyro code that both functions were done in the same place, whereas now you are using both the timer interrupt, and the ADC interrupt to take the samples. Is it safe to kinda shift the ADC functionality to the timer interrupt and reduce the extra overhead by removing the ADC interrupt altogether? Or do the two of them together execute faster, somehow, than the single timer?

Before you:
- Started the initial A/D conversion in Initialize_XXXX()
- Saved the value at the start of the timer2 interrupt
- Did the if/else/etc to put the value in the right place
- Started another A/D conversion by setting the GO bit.
- Exited the timer2 interrupt.

And if I get the go ahead, I'll just do this again.