Skip to main content
Topic: Logic Shrimp variation (Read 13520 times) previous topic - next topic

Logic Shrimp variation

I've made variation of logicshrimp, using only through-hope components found in my junk box.
More details and pictures can be found here ****
There is no schematics drawn, I built it just using datasheets and original schematics - and I'm too lazy to draw it once it is working.
Device is really great, I got it working flawlessly in two or three hours. OSL software is quite good. Cost of all components was under 10EUR.


***hmm, forum system doesn't allow to post me url of my flickr photoset. It thinks I am spammer :-)

Re: Logic Shrimp variation

Reply #1
Great project, this is also on my to-do list.

Sorry about the spam filter. You can exclude http://www. and .com and it should work. You can also shoot it through the contact form and I'll post it here and on the blog.
http://dangerousprototypes.com/contact/
Got a question? Please ask in the forum for the fastest answers.


Re: Logic Shrimp variation

Reply #3
Did you have to recompile for the 26J50?
Got a question? Please ask in the forum for the fastest answers.

Re: Logic Shrimp variation

Reply #4
I assume the crystal + 74hc04 are replacing the oscillator, nice .. but I think you reeeeeeeally need a decoupling cap on all the sram chips and on the input buffer

Re: Logic Shrimp variation

Reply #5
[quote author="ian"]Did you have to recompile for the 26J50?[/quote]

I tried to, but after struggling with include files and successful build it did not work correctly. I believe there was mismatch in descriptors/endpoints assignment. I understand problems with USB stack distribution, but document about how to adapt it for logic shrimp would be nice.
As a quick and dirty solution I FLASHed image (app + bootloader) for PIC18F24J50 into 18F26J50 and it seems to work. I didn't check bootloader functionality, however.

Re: Logic Shrimp variation

Reply #6
[quote author="arhi"]I assume the crystal + 74hc04 are replacing the oscillator, nice .. but I think you reeeeeeeally need a decoupling cap on all the sram chips and on the input buffer[/quote]

There are 100nF decoupling caps under IC sockets, as well as 22pF (for crystal) and 10uF (Vcap pin) caps under PIC socket. Seven pieces of 100nF caps, but only one visible.
74HC04 isn't oscillator replacement. I enabled HS oscillator in PIC and HC04 is just a buffer to take signal from OSC2 pin.

Re: Logic Shrimp variation

Reply #7
I didn't make any other changes than described in the source and in the wiki about the stack:
http://dangerousprototypes.com/docs/Com ... C_projects
Other than that I just include the files and compile. We use an assigned USB VID/PID, but that is just a minor change in the descriptor and you need not change from the default.
Got a question? Please ask in the forum for the fastest answers.

Re: Logic Shrimp variation

Reply #8
I don't know what wrong I've done. My device enumerated as CDC, created virtual port, but it never received any data. I tried to send SUMP commands manually from terminal, but I wasn't successful. I used TeraTerm and Tx indicator was permanently in "active" state, so it seems like data was never really sent.

I tried to build my own CDC device on the same board and it works as expected.
In fact, I have quite a bit of experience with Microchip USB stack, but I was lazy to debug it, as there was complete solution on hand :-)

Re: Logic Shrimp variation

Reply #9
Hum, maybe they've updated again and broken some compatibility in our config files. Thanks for the followup, I'll check into it.
Got a question? Please ask in the forum for the fastest answers.

Re: Logic Shrimp variation

Reply #10
[quote author="ian"]Hum, maybe they've updated again and broken some compatibility in our config files. Thanks for the followup, I'll check into it.[/quote]

That is a huge pain in the .. every time they upgrade MAL they break some applications :(

[quote author="jaromir"]
There are 100nF decoupling caps under IC sockets, as well as 22pF (for crystal) and 10uF (Vcap pin) caps under PIC socket. Seven pieces of 100nF caps, but only one visible.[/quote]

ah great :) I love to put them inside the socket too :)

[quote author="jaromir"]
74HC04 isn't oscillator replacement. I enabled HS oscillator in PIC and HC04 is just a buffer to take signal from OSC2 pin.[/quote]

You are using HS oscilator on the PIC and using clk out to drive the hc04 .. I understood you made oscillator with crystal and hc04 and then feed the signal to the both pic and memory

Re: Logic Shrimp variation

Reply #11
hi would it be posible to use different spi mem chips then the ones in the desingn, atmel's AT25DF041A has 4 Mbits adn a 70MHZ clk, and i think it is pin compatable, how dificult would be to change the code to take 4M of data insted 256k
best regards FIlip.

Re: Logic Shrimp variation

Reply #12
That looks like a flash chip, so it would have limited write cycles in theory.

You'd need bigger timers for this I think. The 16bit timers in the PIC use a /4 to get the same count as in the SRAM. I guess you could use a bigger divider, but then you start to get strange start and stop points. There's some info on the timer setup in the design overview (especially the last paragraph of the section):
http://dangerousprototypes.com/docs/Log ... ock_system
Got a question? Please ask in the forum for the fastest answers.

Re: Logic Shrimp variation

Reply #13
thatnks for a quick reply,
about the counter, couldnt a 8 bit timer0 be used to drive the Timer one thus giving 24 bit adressing or 16Mb per chanel
best regards FIlip.

Re: Logic Shrimp variation

Reply #14
*offtopic, could a paralell rom chip be used  to buffer data.
best regards FIlip.