Great stuff, Perry! I will be building up a protoboard just to play with this! [8^)
btw, an idea I'm going to try is to modify the ADC circuit a tiny bit:
* use a resistor on both the sense and feedback pin
* different values (e.g. X and 2X)
* that way I can switch their roles and have an instant change in the ADC range
* assuming the sense pin's input impedance is high enough, the extra resistance in series shouldn't mess with the ADC's accuracy
Great stuff, Perry! I will be building up a protoboard just to play with this! [8^)
btw, an idea I'm going to try is to modify the ADC circuit a tiny bit:
* use a resistor on both the sense and feedback pin
* different values (e.g. X and 2X)
* that way I can switch their roles and have an instant change in the ADC range
* assuming the sense pin's input impedance is high enough, the extra resistance in series shouldn't mess with the ADC's accuracy
Jonathan
This is good news. I really need someone to debug the VGA interface.
on another note I am making progress with dumping the screen to the PC via serial port
I always used B/W but have now has to get my head around color schemes
Here's an example of a TV capture
I use tinycore linux for OS
brad's spin tool for development
display BMP is uploaded to PC in binary form to CuteCom
Thanks RAY, but the only decent Windows system I have is used as a router. I will have to go offline to try it .
but I have been successful in using the BMP dump you wrote to upload display to linux. need to get used to the wacky color schemes though. I have been using 8bit color or just B/W mostly.
As soon as you connect the propeller to a PC you can use ViewPort from Hannoware as an oscilloscope.
Not only for ADC but for ALL variables you would like to.
Another possability is propterminal from insonix which can be used as a TV-emulator on the PC.
Hello Perry,
I've tested but displayed nothing. The changes I've made were
' gr : "TV_graphics"
gr : "VGA_graphics" 'Use for VGA screen.
and
'basepin = 12 'Setting for Demo Board TV.
'basepin = 24 'Setting for Hydra TV.
basepin = 16 'Setting for Demo Board & Hydra VGA
Are there anything I need to change as well? On the hardware side, I'm using prop proto-board with mounted VGA/PS2 KB/Mouse connector and base 16 was tested OK.
Thanks a lot.
I don't see where you have modified the A/D pins.
I am convinced that the VGA should work. I did try the VGA settings with the new dump screen to PC code. I could not see anything 'cause I had no VGA hardware, but was able to dump and save a display image by hitting the correct keys. There is only on thing we overlooked that being the selection of the A/D pins.
I will publish this new version when I am satisfied with the color choices for the dumped output.
' start ADC
SampleTime := 127
InPinA := 18 <---- change this
oPinA := 17 <----- and this
nAverageA := 10
nPeakA := 1
InPinB := 8 <---- and maybe this
opinB := 10<---- and this
nAverageB := 10
nPeakB := 1
inPinA and oPinA would conflict with the VGA setting. please choose pins other than those used already for the A/D converters and try again
I am convinced that the VGA should work. I did try the VGA settings with the new dump screen to PC code. I could not see anything 'cause I had no VGA hardware, but was able to dump and save a display image by hitting the correct keys. There is only on thing we overlooked that being the selection of the A/D pins.
I will publish this new version when I am satisfied with the color choices for the dumped output.
' start ADC
SampleTime := 127
InPinA := 18 <---- change this
oPinA := 17 <----- and this
nAverageA := 10
nPeakA := 1
InPinB := 8 <---- and maybe this
opinB := 10<---- and this
nAverageB := 10
nPeakB := 1
inPinA and oPinA would conflict with the VGA setting. please choose pins other than those used already for the A/D converters and try again
It worked!!! OK, I've checked the schematic of proto-board & yes, those pin 17 & 18 were the culprits. Changed & it worked.
I guess the maximum-sampling rate and precision is limited. But for a lot of purposes it will be enough.
As we are a userforum and have no pressure to sell as much units as possible, how about doing some tests that show the limits of this oscilloscope?
I mean some screenshots with a description of parameters. (rectangle-signal with a freq of x kHz/MHz
measuring the rising-time of some digital signals shown on the PMO (poor man oscilloscope) compared to a screenshot of a high-end 5Gigasamples /second
slow optocoupler, RC-circuit, fast IO (prop-pin itself)
and as a medium perfomance oscilloscope two ADCs that were used in the IO dreamkit.
again showing the limits of this kind of scope.
So everybody can decide wich type of scope will be suitable for his needs
I'm think I'm fine in contributing images with this scope in the mist of my use.
Don't get too carried away with this version. Just get your A/D working.
The screen dump works good now.(maybe still need some color tweaking)
I have decided to change the interface to use mostly function keys, instead of number keys, ( that was a throwback from when I mostly used input from infraRED controls)
There is a new method that allows you to put in numbers to set variables maybe should also save settings to eeprom I still have over 100 longs to add functions even after loading a DDS
Also you can move the two traces individually, as well as scaling them individually
I have also decided there is enough room for one or two signal generator channels either with duty cycle on the audio outputs or with some other DDS.
Just updated the software on the first post of this thread to include dump screen to serial port and changes to allow inputs of numerical variables. Also independent controls of scaling and movment of each channel via shift for channel A and control for channel B.
There is over 400 longs still available for functions (nearly 600 with BST extensions). I am beginning to get lots of new ideas for functionality. Use audio outputs with PLL for function generators, " lonesock" had a "Prop_Video_HW_DDS" post that nobody responded to that I have been wanting to try out for a long time. Also one should be able to use the TV video hardware to generate RF FM signals (I left an outline for a stereo FM idea I will work on)
I still have to make the display accurately show timing information(Interger scaling arithmetic). There is a function that allows you to mark a set point and then move a pointer to tell you the time difference between points.
Oh! and F12 dumps the video screen binary BMP data serially at 57600 baud thanks to" Rayman"
F2,F3,F4 have preset times of 1,10,100, with F2 you can see the equalization pulses in the TV vertical sync. at the maximum recommended resolution of 1M samples/second.
I thing I will make a minimum version just to show the basics of this codes usage for other possible uses.(... that shows about 1000 longs available)
Noticing that this scope seems to miss a lot, is it possible that the capacitance of my breadboard is enough to smooth some of what I am trying to measure? It does not seem as though the small amount of capacitance of the breadboard conductors should have this effect, and if I compare what I get here with my main scope, after attaching the sample to the breadboard, and reading from the opposite end of the breadboard conductor, it does not seem to have such an effect. The distortions that are not showing up are 25% of the signal and slower than 500Hz. Any ideas why this could be?
Noticing that this scope seems to miss a lot, is it possible that the capacitance of my breadboard is enough to smooth some of what I am trying to measure? It does not seem as though the small amount of capacitance of the breadboard conductors should have this effect, and if I compare what I get here with my main scope, after attaching the sample to the breadboard, and reading from the opposite end of the breadboard conductor, it does not seem to have such an effect. The distortions that are not showing up are 25% of the signal and slower than 500Hz. Any ideas why this could be?
Haven't a clue, a schematic would be helpful, as well as data rate your using there are 4 presets F1,F2,F3,F5 that you should try
The delta-sigma ADC with the propeller works only good if you
1.) are using a prop in a QFN-package on a PCB
2.) wires from prop-IO-pins to capacitors are less than one inch long
3.) you don't use neighbourhooded prop-IO-pins f.e. prop-pin no 7 and 8 (to avoid cross-talking use prop-pin 7 and 12)
A bread-board is definitely not good for this application
Edit: anyway I'm highly interested what perfomance you can achieve if a DIP40 with capacitors soldered next to the Pro-IO-pins
If I understand right the bit-resolution and sampletime depends on the frequency with which the IO-pins are switched
StefanL38:
Thank you. I always like to test a circuit on breadboard before making a PCB, I guess this is one of those I have to risk (by going directly to PCB).
SetBuffer actually isn't needed with EasyADC. I replaced it with my latest version (had to add in a parameter to specify if the data is filtered). I also had to remove a line about PrintScreen_serial (IIRC), and I had to modify my pin definitions in config.spin. After that, it worked fine.
Saved archive is set for VGA could someone try it and report results?
I wimped out and used FloatMath and FloatString to show period information.
added "f" key to use function generator
Right now it's a simple square wave generator you can toggle on/off .
Here is a scenario of usage.
connection one channel to the function generator output pin
Start PMDO
hit down arrow(or up) key to get display near center
hit "f" key to turn on function generator
you should see a square wave
hit "space" key to stop sampling
hit "i" key to turn on Information display
hit "F10" to move pointer to beginning of signal to measure
hit "s" to set position
hit "F11" to move pointer to end of period to measure
hit "F12" key to dump screen to SD card
screen will go blank for 8-10 seconds
you will get a PMDO_100.BMP
use whatever graphic tools you have to make the image more useful
There is crosstalk in the test image only one channel should not see the square wave.
I think I have a problem near the earphone connector I use to connect my propeller to the outside world
Saved archive is set for VGA could someone try it and report results?
Hi Perry,
first of all: very nice project and good work !
I could't compile this version, because object exceeds memory limit by 205 longs for vga-version, 230 longs for tv ?!?!
Tharkun
P.S. I'm working on a small case with 7"-Display for the PMDO using some buttons and encoder instead of keyboard,will post pictures when done.
I could't compile this version, because object exceeds memory limit by 205 longs for vga-version, 230 longs for tv ?!?!
.
I was afraid that
I use "Brad's Spin Tool" which has a compiler option of "eliminate unused SPIN methods"
you can get it here http://www.fnarfbargle.com/bst.html
Just my 2 cents ... maybe a specialized video / VGA driver could be created for OScope applications?! The grid can be generated programmatically, a top and a bottom line could be rendered by using the propeller font for displaying scales and other text info and for the signals a buffer which can be much smaller than the full frame-buffer could be used.
This way the resolution could be increased to 800x600 or maybe 1024x768 while using less memory - giving more room for other functions.
Comments
Perry, I think I did this already... There should be a thing called "Prop Monitor" (or something like that) on my website...
You made the correct changes, I had hoped it would just work, need to someone with more experience to check it.
or I will have to breakdown an put a VGA connector on something.
Could not find that, but the YModem page looks helpful.
You should update your links ther to reference paralixinc for thos old threads though.
btw, an idea I'm going to try is to modify the ADC circuit a tiny bit:
* use a resistor on both the sense and feedback pin
* different values (e.g. X and 2X)
* that way I can switch their roles and have an instant change in the ADC range
* assuming the sense pin's input impedance is high enough, the extra resistance in series shouldn't mess with the ADC's accuracy
Jonathan
Perry, Here's the link for PropMonitor: http://www.rayslogic.com/propeller/Programming/PropMonitor/PropMonitor.htm
This is good news. I really need someone to debug the VGA interface.
on another note I am making progress with dumping the screen to the PC via serial port
I always used B/W but have now has to get my head around color schemes
Here's an example of a TV capture
I use tinycore linux for OS
brad's spin tool for development
display BMP is uploaded to PC in binary form to CuteCom
then I use gnupaint to convert to PNG format
Thanks RAY, but the only decent Windows system I have is used as a router. I will have to go offline to try it .
but I have been successful in using the BMP dump you wrote to upload display to linux. need to get used to the wacky color schemes though. I have been using 8bit color or just B/W mostly.
Not only for ADC but for ALL variables you would like to.
Another possability is propterminal from insonix which can be used as a TV-emulator on the PC.
best regards
Stefan
I don't see where you have modified the A/D pins.
I am convinced that the VGA should work. I did try the VGA settings with the new dump screen to PC code. I could not see anything 'cause I had no VGA hardware, but was able to dump and save a display image by hitting the correct keys. There is only on thing we overlooked that being the selection of the A/D pins.
I will publish this new version when I am satisfied with the color choices for the dumped output.
inPinA and oPinA would conflict with the VGA setting. please choose pins other than those used already for the A/D converters and try again
It worked!!! OK, I've checked the schematic of proto-board & yes, those pin 17 & 18 were the culprits. Changed & it worked.
Cool in the extreme!!!! :cool: Thanks a lot!!!
As we are a userforum and have no pressure to sell as much units as possible, how about doing some tests that show the limits of this oscilloscope?
I mean some screenshots with a description of parameters. (rectangle-signal with a freq of x kHz/MHz
measuring the rising-time of some digital signals shown on the PMO (poor man oscilloscope) compared to a screenshot of a high-end 5Gigasamples /second
slow optocoupler, RC-circuit, fast IO (prop-pin itself)
and as a medium perfomance oscilloscope two ADCs that were used in the IO dreamkit.
again showing the limits of this kind of scope.
So everybody can decide wich type of scope will be suitable for his needs
best regards
Stefan
Don't get too carried away with this version. Just get your A/D working.
The screen dump works good now.(maybe still need some color tweaking)
I have decided to change the interface to use mostly function keys, instead of number keys, ( that was a throwback from when I mostly used input from infraRED controls)
There is a new method that allows you to put in numbers to set variables maybe should also save settings to eeprom I still have over 100 longs to add functions even after loading a DDS
Also you can move the two traces individually, as well as scaling them individually
I have also decided there is enough room for one or two signal generator channels either with duty cycle on the audio outputs or with some other DDS.
Will update online version of software soon.
OK, just to feedback. In VGA_Graphics, I've selected the colorset:
Because this one produces an orange background:
Right, function keys would be best. I'll post some test screen shots later.
There is over 400 longs still available for functions (nearly 600 with BST extensions). I am beginning to get lots of new ideas for functionality. Use audio outputs with PLL for function generators, " lonesock" had a "Prop_Video_HW_DDS" post that nobody responded to that I have been wanting to try out for a long time. Also one should be able to use the TV video hardware to generate RF FM signals (I left an outline for a stereo FM idea I will work on)
I still have to make the display accurately show timing information(Interger scaling arithmetic). There is a function that allows you to mark a set point and then move a pointer to tell you the time difference between points.
Oh! and F12 dumps the video screen binary BMP data serially at 57600 baud thanks to" Rayman"
F2,F3,F4 have preset times of 1,10,100, with F2 you can see the equalization pulses in the TV vertical sync. at the maximum recommended resolution of 1M samples/second.
I thing I will make a minimum version just to show the basics of this codes usage for other possible uses.(... that shows about 1000 longs available)
Haven't a clue, a schematic would be helpful, as well as data rate your using there are 4 presets F1,F2,F3,F5 that you should try
1.) are using a prop in a QFN-package on a PCB
2.) wires from prop-IO-pins to capacitors are less than one inch long
3.) you don't use neighbourhooded prop-IO-pins f.e. prop-pin no 7 and 8 (to avoid cross-talking use prop-pin 7 and 12)
A bread-board is definitely not good for this application
Edit: anyway I'm highly interested what perfomance you can achieve if a DIP40 with capacitors soldered next to the Pro-IO-pins
If I understand right the bit-resolution and sampletime depends on the frequency with which the IO-pins are switched
best regards
Stefan
Thank you. I always like to test a circuit on breadboard before making a PCB, I guess this is one of those I have to risk (by going directly to PCB).
Changes:
new Config file contains all I/O information
dump screen now writes to the SD card
now using lonesock's Easy_ADC
Perry
Attached screen dumps show Chua's Chaotic Circuit in action
I cannot get you code to compile. I get Expected instruction or variable at
PUB SetBuffer(argh,aBunch)
buf_ptr := argh
I have same problems.
Saw to at it is not all needed files in archive.
Jonathan
http://forums.parallax.com/showthread.php?130779-EasyADC-call-for-feedback
Jonathan
Saved archive is set for VGA could someone try it and report results?
I wimped out and used FloatMath and FloatString to show period information.
added "f" key to use function generator
Right now it's a simple square wave generator you can toggle on/off .
Here is a scenario of usage.
connection one channel to the function generator output pin
Start PMDO
hit down arrow(or up) key to get display near center
hit "f" key to turn on function generator
you should see a square wave
hit "space" key to stop sampling
hit "i" key to turn on Information display
hit "F10" to move pointer to beginning of signal to measure
hit "s" to set position
hit "F11" to move pointer to end of period to measure
hit "F12" key to dump screen to SD card
screen will go blank for 8-10 seconds
you will get a PMDO_100.BMP
use whatever graphic tools you have to make the image more useful
There is crosstalk in the test image only one channel should not see the square wave.
I think I have a problem near the earphone connector I use to connect my propeller to the outside world
Hi Perry,
first of all: very nice project and good work !
I could't compile this version, because object exceeds memory limit by 205 longs for vga-version, 230 longs for tv ?!?!
Tharkun
P.S. I'm working on a small case with 7"-Display for the PMDO using some buttons and encoder instead of keyboard,will post pictures when done.
I was afraid that
I use "Brad's Spin Tool" which has a compiler option of "eliminate unused SPIN methods"
you can get it here http://www.fnarfbargle.com/bst.html
This way the resolution could be increased to 800x600 or maybe 1024x768 while using less memory - giving more room for other functions.
What do you think?