Dangerous Prototypes Open Source Hardware

My experiences with the OLS

June 19, 2010, 04:45:31 am

I’d like to share my experiences with the OLS on Windows
I’ve ordered the unit with probes on June 2 as soon as I’ve received the back in stock notification. The unit was delivered to my house (Vancouver,Canada) on June 15 from HK.

I’ve downloaded OpenBench_LogicSniffer_1.03.zip package from Gadget factory and the client BFP_Logic_Analyzer_1.0.zip.

After installing the driver (inf directory) I was curious whether the boot code was installed on the PIC:

The flash type was not recognized even after multiple attempts:

          P:My_DocumentsOpenBench_LogicSniffer_1.03FPGA_ROMbin>pump-loader.exe -p:COM7 -status
          PUMP loader

          Opening serial port 'COM7' @ 921600 ... OK
          Found PUMP HW: 1, FW: 0.1, Boot: 255
          Error - unknown flash type (48 01 46 00)

           P:My_DocumentsOpenBench_LogicSniffer_1.03FPGA_ROMbin>

///////////////////////////// power cycle – unplugging USB cable solved the problem

            P:My_DocumentsOpenBench_LogicSniffer_1.03FPGA_ROMbin>pump-loader.exe -p:COM7  -status
            PUMP loader

            Opening serial port 'COM7' @ 921600 ... OK
            Found PUMP HW: 1, FW: 0.1, Boot: 255
            Found flash: ATMEL AT45DB041D
            no input file specified !


///////////////////////////  Unfortunately the boot code was not installed


I’ve fired up the client and the OLS was recognized immediately on COM7. I’ve tested individually all 16 ports with my Rigol DS1052E scope’s calibration pin (1kHz) I’ve also connect the GND PIN to the scope’s GND.

I’ve used the following client settings:
Port Speed 115200bps

Sampling Clock : Internal
Sampling Rate : 1MHz
All 4 channel groups were enabled
Record size : 2K
Noise filter was enabled
RLE: was disabled

With the default ROM the following channels were configured:
OLS PIN   Client Channel
GND   -----
0/16   16
1/17   17
2/18   18
3/19   19
4/20   20
5/21   21
6/22   22
7/23   23
8/24   31
9/25   30
10/26   29
11/27   28
12/28   27
13/29   26
14/30   25
15/31   24
GND    -----

I was a bit surprised of the mapping, there was no activity on the first two client channel groups and 8/24 PIN was mapped to 31 instead of 24 on the client. 
So it is a good idea to enable all channels when trying out the unit for the first time.

I’ve changed the ROM to 16k8bit_inside   (..FPGA_ROM2.016k8bit_inside)
8 channel 16K depth is enough for now for my project.

P:My_DocumentsOpenBench_LogicSniffer_1.03FPGA_ROMbin>pump-loader.exe -p:COM7 -write -wH:Logic_Sniffer.mcs -run
PUMP loader

Opening serial port 'COM7' @ 921600 ... OK
Found PUMP HW: 1, FW: 0.1, Boot: 255
Found flash: ATMEL AT45DB041D
Reading HEX file 'Logic_Sniffer.mcs' ... OK! (binary size = 169216)
Will write 641 pages
Page 0x0000 write ... OK
Page 0x0001 write ... OK
Page 0x0002 write ... OK
Page 0x0003 write ... OK
Page 0x0004 write ... OK
Page 0x0005 write ... OK
Page 0x0006 write ... OK
Page 0x0007 write ... OK
Page 0x0008 write ... OK
Page 0x0009 write ... OK
Page 0x000a write ... OK
Page 0x000b write ... OK
Page 0x000c write ... OK
Page 0x000d write ... OK
…..
Page 0x027b write ... OK
Page 0x027c write ... OK
Page 0x027d write ... OK
Page 0x027e write ... OK
Page 0x027f write ... OK
Page 0x0280 write ... OK
PUMP switched to RUN mode

P:My_DocumentsOpenBench_LogicSniffer_1.03FPGA_ROMbin>

 ///////////////////////////  Everything was OK and after the ROM change
The mapping was like this:
OLS PIN   Client Channel
GND   -----
0/16   0
1/17   1
2/18   2
3/19   3
4/20   4
5/21   5
6/22   6
7/23   7

I’ve enabled the trigger and checked the pulse length against the calibration pulse. It was bang on.
Trigger 1kHz ->  .0005sec  (scope measured 500us)

I’m also happy to report that SPI protocol analysis so far worked like a charm.
I’m working on a small project where I’m connecting an ATMega168 to an AVR32 over a 1Mhz SPI.
With the client’s sampling rate set to 10MHz I was able to decode the packets. 


Issues:
-No Boot code
-After two days of testing I’ve noticed that sometimes the client won’t recognize the OLS (Com port is present) , reset doesn’t help , the workaround is to move the USB to a different port and install the driver again
-RLE doesn’t seem to work on the client. When RLE is enable the channel capture turns blue and pulses zoomed in doesn’t correspond to the 1KHz signal.

Questions:
What improvement would I see with SPI mode 2.04? Would it be faster?
Well first I have to figure out how to put the bootcode on the PIC.


This is a great project a big thank you for the contributors !!

Thanks,
James
http://nagycnc.com

Re: My experiences with the OLS

Reply #1 – June 19, 2010, 08:57:25 am

Hi James,

Thanks for the write-up.

I'm really sorry about the bootloader issue. There's a couple ways to get it programmed. If you have access to a PIC programmer you can program it directly. If you have access to a Bus Pirate, we've written a rescue application that can reprogram the OLS with a Bus Pirate. You can return to Seeed for a reflash, but they're pretty far away. I have an address of someone in nearby who might be able to reflash it for you, please send me a PM.

'No recognized' error should be fixed by the SPI mode on the 2.04 upgrade.

RLE is still a bit of a work in progress.

SPI mode won't really be faster yet, but some have reported a slight improvement. Eventually it will help us speed up the interface a lot. It should fix the intermittent 'no reply' problems.

Re: My experiences with the OLS

Reply #2 – June 19, 2010, 07:44:25 pm

I have been testing the SPI v2.04 and find it about 10 times faster than the UART version, that is about 1000000 baud.
Because there is no UART the baud rate is not used, the speed is set by the the speed of the  SPI  and  USB interfaces.
I have not been able to get the 16 bit by 8k mode to work. When I set groups 2 an 3 off and select 8k all I get is the first 4k repeated.
The RLE mode it seems to be limited to a max of approx 16 million clock periods. I suspect this is in the java client.
Any recording which exceeds this causes the java client to hang until the close button is pressed.

Re: My experiences with the OLS

Reply #3 – June 19, 2010, 08:50:51 pm

[quote author="DavidFrancis"]
Because there is no UART the baud rate is not used, ...
[/quote]

Careful with that ! I have checked linux kernel, and it actually does use throttling on the CDC serial port !! (but that might be just on the output direction)

Re: My experiences with the OLS

Reply #4 – June 20, 2010, 11:50:13 am

[quote author="DavidFrancis"]
 I have not been able to get the 16 bit by 8k mode to work. When I set groups 2 an 3 off and select 8k all I get is the first 4k repeated.
[/quote]
Thanks for the report.
This confirms my observation with the 16Channel/8k mode.
I have posted this bug two weeks ago, but nobody else seems to have noticed this yet.

Eberhard

Re: My experiences with the OLS

Reply #5 – June 20, 2010, 11:59:46 am

I think Jack is on top of this one, he's working like crazy to release another update to the FPGA core. We hope to have a new release this week.