| Author |
Message |
|
|
 Posted: Nov 01, 2007 - 11:21 AM |
|
Joined: Jun 27, 2005
Posts: 2691
Location: St.Petersburg, Russia
|
|
| Johan, did you buy that Spartan-3E board yet (or Altera DE1 for that matter)? I don't know much about logic analyzers in general but it would seem to be a natural platform for developing one. A logical choice ;) |
_________________
The Dark Boxes are coming.
|
| |
|
|
|
|
|
|
Posted: Nov 01, 2007 - 01:01 PM |
|
Joined: Mar 27, 2002
Posts: 9264
Location: Lund, Sweden
|
|
Well, Svo... One never should talk openly about ones plans. They tend to not materialize...
Still, you can all put two and two together at this point and find that I am interested in getting a potent logic analyzer going, and that I have an aspiring interest in FPGAs. So yes, I came to the "natural platform" conclusion quite a few moths back. It helped that I had four FPGA/VHDL programmers in the cubicle next to me at work.
No, I haven't bought the Spartan-3E board yet. Future financial uncertainty combined with my natural caution-ness has stopped me (although I am very tempted right now, with the USD at such a favourable rate). BUT thanks to people in our vincibity (yes, a 'freak) I have another Spartan-3 board heading my way as we speak, and it will immediately be used as my first platform for learning FPGA programming. (And should the doner actually stumble upon this thread, he should know that I am forever greatful!)
Anyhow, after mention here I dod what I should have done months ago and took a stroll down Google Avenue looking for Logical Analyzer Street and the FPGA Alleys. There's a few designs that actually look quite good!
I am reading up on this as we speak: http://www.sump.org/projects/analyzer/
Edit: This is written exactly for the board that is heading my way! |
Last edited by JohanEkdahl on Nov 01, 2007 - 06:18 PM; edited 1 time in total
|
| |
|
|
|
|
|
|
Posted: Nov 01, 2007 - 01:53 PM |
|
Joined: Jun 27, 2005
Posts: 2691
Location: St.Petersburg, Russia
|
|
I'm sure all of it will sort out soon and you'll end up having a lot of Spartan-3E's around and, with luck, maybe even some free time to play with them 
That design (sump.org) looks very nice, almost polished. I think I know a couple of occasions where it could save me a couple days of hard beating my head against a wall. I also wonder why it samples only at 100MHz where 200 should be feasible.
Sorry for hijacking the thread! |
_________________
The Dark Boxes are coming.
|
| |
|
|
|
|
|
|
Posted: Nov 01, 2007 - 05:42 PM |
|
Joined: Aug 26, 2003
Posts: 214
Location: Trondheim, Norway
|
|
|
|
|
|
|
|
Posted: Nov 01, 2007 - 06:42 PM |
|
Joined: Mar 27, 2002
Posts: 9264
Location: Lund, Sweden
|
|
OK - Full disclosure: A couple of years back I built a logic analyzer based on FIFO memories. This was a kit described in the Elektor magazine, and supplied by a german firm whose name eludes me at the moment. It was a daunting task to get a hold on the expensive FIFO memories in an adequate size and the right package - a problem that was finally solved by a fellow freak (what else - the people here are wonderful!). The clock, and trigger logic was all in "74 logic" and it soon became apparent that this logic lacked some crucial functionality (that would have been easy to implement had the designer just though a wee bit longer than he did). The most annoying parts was the really lame abilities to specify trigger conditions, and the total lack of pre-trigger sampling (you really want to see the event that led to the trig just as often as you want to see wgat happened afterwards!).
In hindsight, the only good thing about the board was that it has an AVR on it...
I worked for a while with the idea to patch (in hardware) the board to get some of the functionality. Then there was the issue of (PC) host software. THe one that was released for the board was really crappy to say the least - so I had to rewrite that also. So now not much was left of the original thing and I soon more or less abandoned it alltogether thinking that I could just as well build my own with cheaper SRAM or some such. I had a vague theoretical notion on how to utilize programmable logic (PALs etc) and spent some time looking for cheap/free PAL programmers hoping I could use PALs to implement a better trigger, a better memory unit etc.
I tell you - never even try to go down the PAL road! Specifications on how to program them are soooo proprietary, surrounded by NDAs and shit, and most puzzling there is close to no information on the Web, that you will go mad or just sit sobbing before you're half way through the mess!. I was convinced that FPGAs where much to complicated, and that the PAL mess was even worse when transferred to the FPGA domain. I also suspected that development systems would be very expensive. When I finally started to investigate, much to my surprise I found that it was just the opposite. Open standards, cheap boards, a lot of web info. By the same time I was contracted to the place where I had four FPGA/VHDL programmers in the next cubicle, and I have been slowly persuading this FPGA thingies since then.
The last few days have given me
i) links to VHDL tutorials that I wasn't aware of,
ii) the Spartan-3 Starter Kit board as a gift, and
iii) a link to a logic analyzer design intended to run on this very board. From what I see upon first read it has exactly what I am looking for: A decent "multi-stage" trigger, pre-trigger sampling and a PC host app that looks really nice.
All this has been made possible by generous fellow 'freaks. I am forever greatful for this. Thank you all!
I will report back here as soon as I have tested the setup! |
|
|
| |
|
|
|
|
|
|
Posted: Nov 01, 2007 - 10:11 PM |
|
Joined: Nov 05, 2005
Posts: 39
|
|
| Aw, no one answered my question :'( Anyways, I've received some responses from the companies I've contacted, and have some discounts offered. The links posted to the open source logic analyzers seem useful but we don't have the resources to invest in getting one of those up and running while we also try to design and build the current electrical system. Keep in mind we are all full time college students as well. Oh, also, I would really love an LA that can also generate I2C signals. |
|
|
| |
|
|
|
|
|
|
Posted: Nov 01, 2007 - 10:25 PM |
|
Joined: Mar 27, 2002
Posts: 9264
Location: Lund, Sweden
|
|
|
Quote:
Oh, also, I would really love an LA that can also generate I2C signals.
Yes, that has also seemed to me to be the next logical step. I think of it as a "signal injector". I would like to use the PC host software to edit waveforms, pump them down to the "injector" which listens to some signals for a trigger condition and then starts to pump out the waveform I have edited.
Quote:
Aw, no one answered my question :'(
Well, both yes and no. Sorry for hijacking your thread somewhat, but I hope that the discussion has been enlightening for you. It certainly has been for me. When I get back with my results I will do so in a separate thread.
Good luck, and please report back with any findings! |
|
|
| |
|
|
|
|
|
|
Posted: Nov 03, 2007 - 04:44 AM |
|
Joined: Nov 05, 2005
Posts: 39
|
|
| I've got a Logicport on the way, we'll see how this works for us, thanks guys. |
|
|
| |
|
|
|
|
|
|
Posted: Nov 17, 2007 - 05:44 AM |
|
Joined: Jun 28, 2006
Posts: 22
|
|
|
vector86 wrote:
I've got a Logicport on the way, we'll see how this works for us, thanks guys.
Hi,
I bought my LogicPort about a year back and I cannot imagine solving any digital problem WITHOUT it. On my last project I had all sorts of issues with SPI and solved them all in a matter of minutes.
Pros:
-> Compression
-> 34 Channels
-> The software is great and gets updated quite often. The triggers are very versatile and combined with the pre-trigger buffer it is possible to 'look back' at what caused the trigger event.
-> Software interpreters
Cons:
-> Longer measuring leads...can fix that though, if it starts to irritate me for long enough
Hope you find your LP as useful as I do! |
_________________
The only thing more dangerous than a hardware guy with a code patch is a programmer with a soldering iron.
|
| |
|
|
|
|
|
|
Posted: Nov 17, 2007 - 09:08 PM |
|
Joined: Nov 05, 2005
Posts: 39
|
|
| The logicport arrived a while ago but I wasn't able to hook it up until recently.. anyways, on first impression I can already tell it will prove to be useful however it seems like there will definitely be a learning curve involved. For example I tried sampling some I2C data but I'm getting inconsistent acquisitions; sometimes I see what I sent and others I see seemingly random values?? |
|
|
| |
|
|
|
|
|
|
Posted: Nov 18, 2007 - 07:55 AM |
|
Joined: Jun 28, 2006
Posts: 22
|
|
|
vector86 wrote:
For example I tried sampling some I2C data but I'm getting inconsistent acquisitions; sometimes I see what I sent and others I see seemingly random values??
Have you set the sampling rate significantly higher than the i2c's data rate?
Did you remember to ground the LP (The grey and black leads)?
Have you set the logic threshold to be compatible with the device you're measuring.
Hope it helps. |
_________________
The only thing more dangerous than a hardware guy with a code patch is a programmer with a soldering iron.
|
| |
|
|
|
|
|
|
Posted: Jan 20, 2008 - 12:27 AM |
|
Joined: Nov 05, 2005
Posts: 39
|
|
Been using the logicport for awhile now, it's definitely worth the cost! Even though I still wish it had the ability to drive lines (e.g. i2c), it has proven it's usefulness time and time again, thanks guys.
logic port in action:
testing the Ignition triggers:
Debugging the VFD display on our tach board:
Right now I'm using it to debug our telemetry system. It was definitely a good investment. |
|
|
| |
|
|
|
|
|
|
Posted: Jan 20, 2008 - 12:32 AM |
|
Joined: Apr 01, 2004
Posts: 3812
Location: New Mexico
|
|
|
vector86 wrote:
It was definitely a good investment.
I'm glad it's worked out well for you. It has certainly paid for itself in my work as well. |
_________________
Kevin Rosenberg
http://b9.com
http://kevin.hypershots.com
|
| |
|
|
|
|
|
|
Posted: Jan 20, 2008 - 01:00 PM |
|
Joined: Mar 26, 2005
Posts: 238
Location: Victoria, Australia
|
|
i'm still waiting for mine (not their fault dhl or customs's!)
should arrive monday 
Darren |
_________________
----------------------------------------------------
Those whom the gods wish to destroy
they must first teach to use c
----------------------------------------------------
|
| |
|
|
|
|
|
|
Posted: Jan 22, 2008 - 11:40 AM |
|
Joined: Mar 26, 2005
Posts: 238
Location: Victoria, Australia
|
|
It's here!!!
IMPORTANT NOTE! THE GROUND WIRES ARE GREY WITH BLACK TIPS!!
within 2 hours of having it it's saved me two days of work see below photo
the exact timing of this has been bu%&*^ering me for days.
Darren |
_________________
----------------------------------------------------
Those whom the gods wish to destroy
they must first teach to use c
----------------------------------------------------
|
| |
|
|
|
|
|
|
Posted: Feb 01, 2008 - 05:00 PM |
|
Joined: Sep 21, 2004
Posts: 25
Location: Colorado
|
|
|
|
|
|
|
|
Posted: Feb 01, 2008 - 06:24 PM |
|
Joined: Apr 01, 2004
Posts: 3812
Location: New Mexico
|
|
|
dmcinnes wrote:
within 2 hours of having it it's saved me two days of work see below photo
Yes, actually seeing the output can have time savings in that magnitude.
Another potential time savings during debugging is to output signals based on some condition. For example, in one application I was inputting serial input on a bit level simultaneously of different protocols, both interrupt based. The code to handle two simulaneous, asynchronous bit streams with interrupts is complex and processing happens to fast to send characters for debugging purposes. So, as debugging output, I'd flip bits on certain output lines to identify states which was a great help. |
_________________
Kevin Rosenberg
http://b9.com
http://kevin.hypershots.com
|
| |
|
|
|
|
|
|
Posted: Feb 02, 2008 - 02:00 PM |
|
Joined: Mar 26, 2005
Posts: 238
Location: Victoria, Australia
|
|
in the current app i have six interrupts, to debug them i'm setting pins 0-5 of port b high as i enter each routine and low as i leave.
It shows an amazing view of everything!
Darren |
_________________
----------------------------------------------------
Those whom the gods wish to destroy
they must first teach to use c
----------------------------------------------------
|
| |
|
|
|
|
|
FAQ
Search
Register
Log in to check your private messages
Log in
Maximize
RSS
