I think the SUMP analyzer firmware could be tweaked to record ADC values (limited to 8-bits?) instead of digital bit values.
The SUMP client code could be modified to display a waveform instead of logic signals, or a different scope client could be used, like the VB6 one here: http://www.semifluid.com/?p=9
He has a VB6 program that reads and interprets the data.
It would be nice to also have a buffered mode, filling BP memory at with whatever sample rate is selected, then pushing the data out the UART, and repeat. It would be like a real scope, with a screen refresh rate equal to how often data packets are sent over the UART. A ring buffer could be used for slow sample rates (continuous screen updates), and a packet mode for fast sample rates (one screen update per packet). The packet mode could be a variation of the current SUMP logic analyzer mode. I would think that the ADC samples could be limited to 8 bits and still be useful.
Of course, any programming or scripting language with COM port access and graphics capability could be used to create the scope screen.
This scope mode is similar to the sound-card scopes, except it is DC-coupled. You would have to put a 3V button battery in series with the scope probe to an bias AC signal into the 0-6v ADC range.
I also want an o-scope mode like the Seeed DSO... There are win32 and linux apps that work with a sound card, but that is AC-coupled. The BP could do DC-coupling, and more. With an SD-card or i2c SRAM, it could save more samples as well.
I have not studied the code yet. Could we do a dual-channel o-scope using Vpu as a second analog input?
Unfortunately, I normally work 100-hour weeks. I slacked off a little to play with the BP, but I need to get back to work.
It seems that I either have plenty of time to play with new toys but no money to buy them, or I can afford lots of new toys but have no time to play with them. I think that sums up the typical techno-geek's dilemma that the elusive 40-hour work week was supposed to solve.
I like the firmware having multiple modes built-in. I plan to use this firmware as just one of many applications. Eventually, I want a GUI program that lets me select "programs" to run on the BP, then flashes that application and runs it.
Also, I want to add SD card support, which can be done with one fewer lines by tying the SD MISO and MOSI together with a resistor, as seen on other PIC projects. I want to modify the boot loader to flash applications from SD card, if possible. I think flashing from SD would be *much* faster than flashing over a USB COM port. For a socket, I will solder wires directly to a microSD adapter that came with a 1 GB microSD card I plan to use for this project. For full-sized SD cards, you can use a socket like this: http://uanr.com/sdfloppy
I fixed mine. The LED was NOT bad -- it was just soldered in backwards. I unsoldered it and reversed it, and now it works.
It looks like the board layout requires that the USB LED be installed in a reverse orientation compared to the others, and on some units like mine the Mode LED is ALSO installed in reverse orientation, which makes it fail.
Ian, you need to have Seeed Studio be especially vigilant about WHICH of the LEDs needs to be installed in reverse orientation. Also, when testing units, the Mode LED needs to be tested to be sure it is not reversed. A quick unpowered LED test can be performed by applying a current-limited voltage across the LEDs and verifying which polarity lights the LEDs. In my case, I used my multimeter "diode test" function, which dimly lit all LEDs, except the Power LED was *really* bright. Having to hold the probes reversed to light the LED indicates an LED installed reversed, which is correct only for the USB LED.
In the future, it would be an easier build if all LEDs were installed with the same physical orientation, to avoid this problem.
Because of the layout causing assembly errors, I suspect there are also units where the USB LED does not light up...
I fixed mine. The LED was NOT bad -- it was just soldered in backwards. I unsoldered it and reversed it, and now it works.
When probing the LEDs with the "diode test" function of my multi-meter, the LEDs now light up for all LEDs except USB, where I have to REVERSE the probes to make it light upI
It looks like the board layout requires that the USB LED be installed in a reverse orientation compared to the others, and on some units like mine the Mode LED is ALSO installed in reverse orientation, which makes it fail.
Ian, you need to either have Seeed Studio be especially vigilant about WHICH of the LED needs to be installed in reverse orientation. In the future, it would be an easier build if all LEDs were installed with the same physical orientation, to avoid this problem.
Because of the layout causing assembly errors, I suspect there are also units where the USB LED does not light up...
I fixed mine. The LED was NOT bad -- it was just soldered in backwards. I unsoldered it and reversed it, and now it works.
When probing the LEDs with the "diode test" function of my multi-meter, the LEDs now light up for all LEDs except USB, where I have to REVERSE the probes to make it light upI
It looks like the board layout requires that the USB LED be installed in a reverse orientation compared to the others, and on some units like mine the Mode LED is ALSO installed in reverse orientation, which makes it fail.
Ian, you need to either have Seeed Studio be especially vigilant about WHICH of the LED needs to be installed in reverse orientation. In the future, it would be an easier build if all LEDs were installed with the same physical orientation, to avoid this problem.
Because of the layout causing assembly errors, I suspect there are also units where the USB LED does not light up...
Regarding the request for "internal" signals for testing analyzer mode, stray power-line signals toggle the data lines fine for me by touching my fingertip to the MISO, MOSI, etc. pins. In my case I get nice 60 Hz waveforms. Even at 1 MHz sample rate, you only have to hit capture 2 or 3 times to get zero-crossings to toggle the I/O lines. At 10 KHz you get lots of toggles.
[glow=red,2,300][/glow]I received my two BPv3 units yesterday. I wanted probe kits too, but they were out-of-stock ;-(. Both BPv3 units were in a small box with no documentation or receipts. Only the boards stuck into conductive foam blocks were in the box. Luckily, I have Google. Google found BPv3 units on ebay, large quantity new [glow=red,2,300]with USB cables[/glow] for a "Buy Now" price of $30 including shipping. Mine, as ordered from Seeed Studio, did not come with USB cables, which in my case was not a problem.
The Perl test scripts did not work with my units on COM11 and COM12, so I removed old USB COM ports and made the BPv3 boards COM3 and COM4, and then the scripts worked fine. I removed the extra COM ports by setting the environment variable "devmgr_show_nonpresent_devices=1", then using Device Manager and selecting "Show hidden devices".
Both units pass all tests including the binary mode tests. I also used the Sump Client windows binary to test logic analyzer mode. Touching your finger to the MOSI, MISO, etc. pins records nice 60 hz waveforms (stray AC power-line coupling to the body). The different lines appear to switch at zero-crossings at slightly different times. I needed to set my USB COM port timeouts to 4000 in device manager as mentioned elsewhere at this site, to get the Sump Client to work with the BPv3. I used this pre-compiled Win32 Sump Client, which needed no add-ons or tweaking: http://www.gadgetfactory.net/gf/downloa ... _0.8.1.zip
My LEDs are Red-Yellow-Green-Red on one unit, and Red-Yellow-(DEAD)-Red on the other unit.
[glow=red,2,300]I have one dead Mode LED. Could it have been soldered in backwards? Is there a way to tell by visual inspection?[/glow]
I noticed another report in the forum of a dead Mode LED as well. Perhaps it has the same problem...
[glow=red,2,300][/glow]I received my two BPv3 units yesterday. I wanted probe kits too, but they were out-of-stock ;-(. Both BPv3 units were in a small box with no documentation or receipts. Only the boards stuck into conductive foam blocks were in the box. Luckily, I have Google. Google found BPv3 units on ebay, large quantity new [glow=red,2,300]with USB cables[/glow] for a "Buy Now" price of $30 including shipping. Mine, as ordered from Seeed Studio, did not come with USB cables, which in my case was not a problem.
The Perl test scripts did not work with my units on COM11 and COM12, so I removed old USB COM ports and made the BPv3 boards COM3 and COM4, and then the scripts worked fine. I removed the extra COM ports by setting the environment variable "devmgr_show_nonpresent_devices=1", then using Device Manager and selecting "Show hidden devices".
Both units pass all tests including the binary mode tests. I also used the Sump Client windows binary to test logic analyzer mode. Touching your finger to the MOSI, MISO, etc. pins records nice 60 hz waveforms (stray AC power-line coupling to the body). The different lines appear to switch at zero-crossings at slightly different times. I needed to set my USB COM port timeouts to 4000 in device manager as mentioned elsewhere at this site, to get the Sump Client to work with the BPv3. I used this pre-compiled Win32 Sump Client, which needed no add-ons or tweaking: http://www.gadgetfactory.net/gf/downloa ... _0.8.1.zip
My LEDs are Red-Yellow-Green-Red on one unit, and Red-Yellow-(DEAD)-Red on the other unit.
[glow=red,2,300]I have one dead Mode LED. Could it have been soldered in backwards? Is there a way to tell by visual inspection?[/glow]
I noticed another report in the forum of a dead Mode LED as well. Perhaps it has the same problem...