Author Topic: TLA7xxx hacks (32M Depth on TLA7AA4CS enabled successful) :)  (Read 32000 times)

0 Members and 1 Guest are viewing this topic.

Offline KK

  • Regular Contributor
  • *
  • Posts: 100
  • Country: us
Re: TLA7xxx hacks (32M Depth on TLA7AA4CS enabled successful) :)
« Reply #75 on: February 19, 2015, 07:15:13 pm »
Verbatim is correct. Use my listed steps verbatim and it will upgrade.

Time on the instrument was close to the time stamp I generated but not exact. I don't think it's important as another user verified he used my same timestamp at a later date.

Good luck!
 

Offline Mark_O

  • Frequent Contributor
  • **
  • Posts: 937
  • Country: us
Re: TLA7xxx hacks (32M Depth on TLA7AA4CS enabled successful) :)
« Reply #76 on: February 20, 2015, 02:28:14 am »
For me, upgrading my TLA704  was a multi-year adventure. Although I do have all the linkages, it'll still require some cutting, hacking, solder'n, copying, pasting...and sweating.  But frankly it was very rewarding and it works great when run remotely via EtherNet!!  I love it now and with the 32 Mb TLA7AA4, its hot! 

Thanks, bench_knob!  I'm not sure I have the time or energy for a multi-year adventure.  But I do know there's much to be learned here.

Most of my lab gear from my latest move isn't even unpacked yet, but I just noticed I had a couple TLA-7N3 modules unboxed, one 2S and one 6S, that I can upgrade to 7S versions (200 MHz, 4 MSa), using the techniques described here recently.  That will certainly be worth doing.

One thing I would definitely like to figure out is how to do the channel stacking I read was possible with these.  I.e., let's say 136 channels width was overkill for me (it always has been, on my 7N4 module), and sometimes I could get along fine with only 17 channels wide.  That collapses the width by 8x, and pushes the depth up by a corresponding factor, so 4M to 32M sample depth. 

I never was able to track that one down, but it would come in handy.  I think it was described as a "software option", but I never found it.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 17631
  • Country: nl
    • NCT Developments
Re: TLA7xxx hacks (32M Depth on TLA7AA4CS enabled successful) :)
« Reply #77 on: February 20, 2015, 12:06:48 pm »
AFAIK Tektronix calls it channel grouping. For the TLA7AA you can reduce the number of channels by 4 and get 2Gs/s with a 128M depth. But I don't recall being able to set lower sampling frequencies while grouping the channels.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Sailor

  • Regular Contributor
  • *
  • Posts: 170
Re: TLA7xxx hacks (32M Depth on TLA7AA4CS enabled successful) :)
« Reply #78 on: February 22, 2015, 09:57:36 am »
First of all: - Thank you very much to everyone who has made this possible :-+

The short version:
Last night I upgraded my TLA7AA2 from 120MHz state / 32M depth to 450MHz state / 32M depth. Today I upgraded my TLA7AA4 from 450MHz state / 8M depth to 450MHz state / 32M depth.

Terrific, worked as advertised, what more can you want?

Now the longer version:
When upgrading the TLA7AA2 I did everything absolutely exactly as in KK's post. All upper-case where he had it, and I included those silly Set/Save Parameter and Code: [Select] lines, etc. And, as I said, it all worked perfectly. But come this morning and those 'useless' lines were still annoying my subconcious, so I decided to have a quick look through the TLA7AAx firmware for any strings that looked like they might be associated. The bottom line is that there is a lot of stuff to look through, and I haven't yet made any connections, but in my perusals I noticed that there were virtually no all-upper-case symbols/table-entries etc. The password is there with its mixed case, but that was it. So I decided to try upgrading the second module with just lower-case command entries, and to also only enter the values that needed to be updated. In this case, the only change that needed to be made to the module was to increase the sample depth to 32M, so after establishing communication (remotely) using the Talker/Listener app, all I entered was

        syst:pass:inst:pass "PitBull5.0.002"
        syst:conf:depth 33554432
        syst:pass:inst:upd 1404025200

and that was it.

So now I have 204 channels of nice high-speed capture. I hope I don't need it all to fix this $@#&*! Intel 8039 in my 3468A :palm:

I think there's an option for Ida Pro to look at Wind River stuff (can anyone confirm?) It might be interesting to have a bit of a dig through the code. And, of course, I could have the TLA get all incestuous with itself and capture what the 68360 is doing... ;D

Ahhh...I just noticed an earlier post by dxl

Quote
I dissassembled the Firmware in 2009 to upgrade the Module.

@dxl - Did you use Ida, or some other 68000 dis-assembler? How far did you get into it?

Sailor
 

Offline gslick

  • Frequent Contributor
  • **
  • Posts: 287
  • Country: us
Re: TLA7xxx hacks (32M Depth on TLA7AA4CS enabled successful) :)
« Reply #79 on: February 22, 2015, 07:23:06 pm »
When upgrading the TLA7AA2 I did everything absolutely exactly as in KK's post. All upper-case where he had it, and I included those silly Set/Save Parameter and Code: [Select] lines, etc. And, as I said, it all worked perfectly. But come this morning and those 'useless' lines were still annoying my subconcious, so I decided to have a quick look through the TLA7AAx firmware for any strings that looked like they might be associated. The bottom line is that there is a lot of stuff to look through, and I haven't yet made any connections, but in my perusals I noticed that there were virtually no all-upper-case symbols/table-entries etc. The password is there with its mixed case, but that was it.
Sailor

Most likely the only string that is case sensitive is the password string, "PITBULL" on the TLA7Lx and TLA7Mx and "PitBull" on the newer analyzer modules, and all command strings comparisons are case insensitive.

If you dump the strings from the firmware the command text strings are all in mixed case, in the usual convention where the upper case part is the abbreviated version of the string and the lower case part is the remainder of the full length string, and text inside square brackets is an optional qualifier.

I wonder how the channel qualifier would be used with depth and speed and not apply to all channels at the same time?
:SYSTem:CONFig[:CHANnel]:DEPTh
:SYSTem:CONFig[:CHANnel]:SPEed

For the curious attached is the command string text dumped from version 5.0.002 of the TLA7Axx firmware. (The TLA7Axx.lod firmware file is a Motorola S-record file which can be converted to binary and then searched for strings).

It would be interesting to sniff the VXI bus and observe how these commands are used during the operation of the analyzer.
 

Offline Sailor

  • Regular Contributor
  • *
  • Posts: 170
Re: TLA7xxx hacks (32M Depth on TLA7AA4CS enabled successful) :)
« Reply #80 on: February 22, 2015, 11:43:11 pm »
(The TLA7Axx.lod firmware file is a Motorola S-record file which can be converted to binary and then searched for strings).


Yes, that is what I did, but your list is a lot cleaner than I recovered. What did you use to scan the binary? Is your list the raw output from the scan, or have you done some manual cleanup?

It still nags me that it simply ignores set/save and code: commands. Considering that there is no apparent recognition of those strings (i.e. not in the command list), yet they are ignored so quietly while the slightest error in any other command can provoke anything from a warning message to a 'hang'. It makes me wonder if they never actually reach the module... which raises the question of who would be filtering them?

All very low on the list of priorities, I know, but that's just how I'm wired :palm:

 

Offline gslick

  • Frequent Contributor
  • **
  • Posts: 287
  • Country: us
Re: TLA7xxx hacks (32M Depth on TLA7AA4CS enabled successful) :)
« Reply #81 on: February 23, 2015, 12:22:13 am »
I just used a simple strings utility on the firmware binary.  I was using a Windows PC at the time and this is the one I happened to use (which originally came from www.sysinternals.com).
https://technet.microsoft.com/en-us/sysinternals/bb897439.aspx

As far as I could see all of the command string text was located in one contiguous table. I just manually selected and copied that text.

After seeing some of the other text in the firmware I'm curious what you can do if you attach a serial connection to the analyzer module CPU board. Maybe you can't do anything useful other than learn more about how the analyzer module firmware works. I haven't tried to do any disassembly of the firmware myself yet.
 

Offline rtekal

  • Contributor
  • Posts: 25
  • Country: in
Re: TLA7xxx hacks (32M Depth on TLA7AA4CS enabled successful) :)
« Reply #82 on: September 11, 2017, 05:52:59 am »
I have a TLA614 which I have been using for past 10 years. Last month it started failing the power on diagnostics with the NVRAM check fail. Have gone through the other replies. The question is If I replace the NVRAM  IC DS1225AD-85, will the error be eleminated after self-calibration? How to backup the contents of the NVRAM?
 

Offline jasonbrent

  • Regular Contributor
  • *
  • Posts: 174
Re: TLA7xxx hacks (32M Depth on TLA7AA4CS enabled successful) :)
« Reply #83 on: February 28, 2018, 05:32:45 pm »
For posterity, this works on the 5000 series also.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf