Author Topic: digital signal acquisition device  (Read 1506 times)

0 Members and 1 Guest are viewing this topic.

Offline metebalciTopic starter

  • Frequent Contributor
  • **
  • Posts: 460
  • Country: ch
digital signal acquisition device
« on: September 09, 2023, 06:30:53 am »
I am looking for a device with features below, there are different types of units I see but I didnt find one yet at reasonable price or supporting all these, looking for suggestions:

- only for digital signal acquisition/capturing (<3.3V)
- capturing max 8 (ddr) digital signals synchronously with a provided clock at both transitions
- max clock speed around 100MHz, higher speed up to 200MHz is appreciated but not a must, I may live with slower speeds depending on price and other features
- preferably streaming to USB but I can live with capturing if streaming not possible
- I do not need any particular on-board feature (decoding etc.)

I am describing something like Digital Discovery with synchronous clock input and capturing at both clock transitions.
 

Online RoGeorge

  • Super Contributor
  • ***
  • Posts: 7888
  • Country: ro
Re: digital signal acquisition device
« Reply #1 on: September 09, 2023, 08:29:26 am »
If you need that, you are probably doing something wrong.
Closest in functionality would be a Logic Analyzer (LA).

Though attaching extra probes on a DDR bus would probably alter the signals too much to be of any relevance in debugging.  However, if you still want to try that with a logic analyzer, there are some models that can sample at 400MSa/s on 4 bits, or at 200MSa/s on 8 bits, e.g. the DSLogic Plus, but no matter wich logic analyzer would be used, 100MB/s means more than 800Mbit/s, while USB2.0 max speed is 480Mbit/s, so there will be no USB2.0 streaming possible at 100MB or 200MB.

Might be possible for an LA to stream over USB3.0, doubt there would be any affordable USB3.0 LA, because in practice there is not much need of such an LA.  At high speed it is not trivial to guess the delays through the wires of a parallel bus, so connecting 8 parallel channels of such an LA to a DDR bus would be quite a stunt.

Offline metebalciTopic starter

  • Frequent Contributor
  • **
  • Posts: 460
  • Country: ch
Re: digital signal acquisition device
« Reply #2 on: September 09, 2023, 10:23:55 am »
If you need that, you are probably doing something wrong.
Closest in functionality would be a Logic Analyzer (LA).

Though attaching extra probes on a DDR bus would probably alter the signals too much to be of any relevance in debugging.  However, if you still want to try that with a logic analyzer, there are some models that can sample at 400MSa/s on 4 bits, or at 200MSa/s on 8 bits, e.g. the DSLogic Plus, but no matter wich logic analyzer would be used, 100MB/s means more than 800Mbit/s, while USB2.0 max speed is 480Mbit/s, so there will be no USB2.0 streaming possible at 100MB or 200MB.

Might be possible for an LA to stream over USB3.0, doubt there would be any affordable USB3.0 LA, because in practice there is not much need of such an LA.  At high speed it is not trivial to guess the delays through the wires of a parallel bus, so connecting 8 parallel channels of such an LA to a DDR bus would be quite a stunt.

OK. I was thinking naturally USB 3.0 but if that goes out of budget, I guess I can decrease the clock, not sure but maybe quite a lot also to minimize probing issues. Lets say clock is 1MHz.

"In the manual of DSLogic View it writes:
 Using External Clock:
Selecting this option indicates that the sampling clock is provided externally (via the CK channel of the cable)
 Using Clock Negedge:
By default, the system samples on the rising edge of the sampling clock. Selecting this option indicates that the system will sample on the falling edge of the sampling clock."

So it seems to me it is not possible to sample at both edges.

One example signal is ARM ETM trace (4 bit ddr data, separate clock). I am not looking for a high quality solution but a reasonably low cost way to sample such signals (USB 2 and lower clock speeds are OK). It sounds relatively simple to make such a product at low speeds but I dont see any covering all features needed.
 

Offline DavidAlfa

  • Super Contributor
  • ***
  • Posts: 6819
  • Country: es
Hantek DSO2x1x            Drive        FAQ          DON'T BUY HANTEK! (Aka HALF-MADE)
Stm32 Soldering FW      Forum      Github      Donate
 

Offline metebalciTopic starter

  • Frequent Contributor
  • **
  • Posts: 460
  • Country: ch
Re: digital signal acquisition device
« Reply #4 on: September 09, 2023, 03:03:23 pm »
Maybe a EZ-USB FX3 Explorer kit (About $40) and this:

https://community.infineon.com/t5/Knowledge-Base-Articles/EZ-USB-FX3-Explorer-kit-as-16-channel-100-MHz-logic-analyzer-with-sigrok/ta-p/283993?profile.language=en

I do have FX3 kit already and this looks pretty cool. It seems I should look into the sigrok project.
 

Offline Performa01

  • Super Contributor
  • ***
  • Posts: 1766
  • Country: at
Re: digital signal acquisition device
« Reply #5 on: September 10, 2023, 07:29:39 am »
So it seems to me it is not possible to sample at both edges.
It should not be too hard to use a handful of logic gates plus a few passives to build a simple edge detector that spits out a constant pulse for every edge of the input signal, regardless of its polarity.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf