Open Source Logic Analyzers?
Rorschach1 asks: "I'm a hardware hacker, doing my part to develop useful open source projects, but my projects are getting more ambitious and my test equipment just isn't keeping up. PC based oscilloscopes have been discussed here before, but I'm particularly in need of a logic analyzer. The BitScope does offer an 8-channel analyzer with Linux support, but that hardly compares to most commercial offerings. I'm also concerned by the fact that I can't find any independent reviews of the device, and I'm hesitant to shell out $500 without knowing what I'm getting into. A logic analyzer is a pretty simple device. My current setup runs on a dedicated 486 with ancient DOS-based software. Surely there must be a more modern option available that'll work under Linux, but so far I've been able to find nothing. Perhaps someone can point me in the right direction?"
http://et.nmsu.edu/~etti/winter97/computers/logic/ logic.html
You have not defined what type of logic analyzer you need. They span the range of 10mhz and 8 inputs for less than $200 to 1ghz and hundreds of inputs for tens of thousands of dollars. You didn't say if you needed disassembly or a processor pod.
Sometimes I think that all you need to get your question in "Ask Slashdot" is to preface it with "I'm looking for an open-source blah blah blah..."
"Eve of Destruction", it's not just for old hippies anymore...
normally a company providing something of that high a cost would offer a trail version (eg, 45 days) via their website, or can be negotiated with a salesperson from that company. Sure, it might be crippled a la the old shareware days, but may contain enough features to decide whether or not the product is viable. Why not research that option should the OSS route come up short?
Take a PIC microcontroller, add a Xilinx fpga, and a few misc other items and you have a decent logic analyzer.
Check out his Carmon project if you're interested in long term GPS data tracking (stores a GPS data stream to a compact flash card)
-Adam
don't look to SCO for any Open Source Logic Analyzers.
You are being MICROattacked, from various angles, in a SOFT manner.
I know this is not what you asked and is probably out of most hobbiest budgets, but I'll mention it anyways as it could help someone. A project we did here absolutely required a scope, and we considered a logic analyzer as highly desirable but not strictly necessary. We had a budget of around $5k - significant but not really that big when you start pricing this stuff.
It had been a while since I shopped for this stuff and I happily found that Agilent (used to be HP) has what they call "mixed signal scopes". Specifically, we purchased a 54622D. It's 100MHz 2 channel oscillscope, but then it's also got a 16 channel logic analyzer built in. (See full specs here.) We purchased it used for around $2600. To get that price we had to play two of the used dealers against each other a little bit. We have been really happy with it so far. $2600 is quite a bit more than $500, but you do get both a scope and a small logic analyzer for that... It's got a built in serial interface and floppy, and it's more portable than dragging a computer around. Take care, and good luck with your quest!
DOS is what my EPROM programmer software runs on. The reason? DOS is not a multitasking OS. I'm not going to screw up an E-PROM because the programming algorithm's timing was hosed by task-switching, a hung application, accessing a swap file, etc.
I recommend that you avoid any logic analyzer which relies on software for any timing-critical elements of its operation. If that's not possible, then avoid software which runs on a multitasking OS.
that was very informative.
You'll probably only be able to get timing resolution near that of the OS tick interrupt interval, right? Sounds like the makings of a piss-poor logic analyzer.
I suppose if you were using an open source OS kernel, one where you could change the OS tick interval -- you might be able to make use of your logic analyzer idea.
I jerry rigged something "good enough" awhile back with a Xilinx 4000 series FPGA and a microcontroller. I might clean it up and release it if there's any demand, but it's pretty simple to implement a basic one.. you just need a bunch o latches synced to a clock.
Anything more than 10-20mhz you are going to have to buy something though.. although I never did tests to see how fast or reliable my hacked apparatus was.
Go ebay hunting. Lots of deals on there.
..don't panic
... when their designer announced them on the sci.electronics newsgroups. Unfortunately the control software is not open-source, but he might be persuaded to release the interface specs if someone volunteered to do a Linux port.
http://www.rockylogic.com/products/ant16.html
Dahlmann tightly grips the knife, which he may have no idea how to use, and steps out into the plain.
The oscilloscope channels have limited voltage swings, but are good enough for most of the work I have done. You can always build/purchase dividers to measure higher voltages.
If you do get a 1631 or something similar, make sure it has the pods and test clips. They get pretty expensive to purchase/replace.
Dean
With some level shifting circuitry you can turn any common RS232 serial port into a 4 channel logic analyzer. It you've got +5 volt TTL levels, you can probably get away without any extra circuitry.
Say what?!
Yep. Most anyone can do this. In my case, I went so far as to write a device driver that makes it work under Windows NT/2K/XP. It hooks the serial ports interrupt service routine and logs the state of the UART's modem status register or MSR. Four bits from the MSR tell us the state of the CTS, DCD, DST, and RI input pins. I use the RDTSC instruction which offers ridiculous accuracy (theoretically, it's the CPU's clock speed) compared to most other software based solutions - even betten than some coarse-grained hardware solutions.
The same could easily be done under Linux or anything with a serial port and Pentium compatible processor.
I never thought anyone would be interested so no website or anything. Interested persons can send me an email ...
mailto:rfmobile@swbell.net
-rick