DSOX2000 and 3000 series - licence , have anyone tried to hack that scope ?

[georges80]

pins 1 and 6 are shorted on the PCB, so the 0R1 is expected there. 1 and 6 are also the center tap of the two channel magnetics, so you will see some connection between all pins when it's mounted on the PCB. But 0R4 is rather low.

The datasheet for the magjack is here:

http://www.farnell.com/datasheets/71713.pdf

The magjack magnetics on the phy side are just transformers and I'd expect very low DC resistance. I tested a 1gig magnetics that I have on one of my boards and they basically read near zero ohms (the lead resistance of my fluke 87III has more resistance).

Also, I've noticed on the original DIY design that the center tap of the transformer (on the phy side) is going to a capacitor and then ground.

From measurements by Zucca the agilent unit has 3.3V tied to the center tap pins (via a 10 ohm resistor and a cap, likely 0.1uf, to ground). i.e. the center tap is powered to 3.3V. This is common with the magnetics (some use 0V) - all depends on the actual phy in use. In this case 3.3V to the center taps is the 'correct' voltage. Not sure if that has any bearing in this case since your unit works from your reports.

cheers,
george

Edit: Fix stupid typo - grrr...

[Sbampato12]

I've received the pcb from dunkemhigh (thank a lot!), but I haven't got the MagJack yet.
But I already did the jumper on 78-80 pins, and connected it to scope, and got a LAN options avaliable. 

Only waiting the magjack...

[iRad]

Another dunkemhigh LAN board soldered up and installed... but during boot it does give an error 'System concers detected -LAN/VGA option module fault'.

I also received two of the dunkemhigh boards and I am getting the same error on boot 'System concerns detected - LAN/VGA option module fault'. I installed the recommended magjack, pair of 220R and single 1uf components on both boards, but I get the same error on each. Anyone figure this out yet? Any suggestions?

[georges80]

As per my post #563 I don't believe the cap to ground on the center tap of the magnetics is correct. The original LAN/VGA board has 3.3V going through a 10 ohm resistor to the center taps. Also a cap to ground on the center tap side (just a bit of filtering). The key though is that the center tap is at 3.3V potential.

I'd give that a try and see if it helps. Presumably the scope is doing some checks via the Phy during boot up and sensing something isn't quite right. Given NOTHING is wired to any other pins the detected problem seems more than likely just a LAN issue.

My board (I layed out my own 'version') reports no lan/vga erros and works perfectly. I used the same magjack as the agilent lan/vga board and used the 10 ohm/0.1uf cap to get 3.3V to the two center taps.

cheers,
george.

[iRad]

Thanks George. I had read your posts, but was under the impression that others had the dunkemhigh board working fine with just the two 220R and a 1uf. It seems that way from other reports about the board anyway.

I'm not sure how to assure that the jack has 3.3 volts to the center tap without actually measuring voltages on the jack while the board is installed and running. And I'm worried about leaving it installed and running in my expensive scope given the error, and then probing around on top of that with power to the scope and board. I'm hoping someone else who has this board can give a more definitive answer about the correct component values, and if another component should be added or not. Alternatively, if someone knows what voltages should be applied to what edge connector pins on the card, I could then check the center taps with the card uninstalled and powered by my bench power supply.

I appreciate your help and comments, but I'm a bit apprehensive, especially remembering that Dave burnt his scope up once on a bad LAN card and had to return it to Agilent to be repaired.

[georges80]

The 3.3V comes from the edge connect. Go back a few pages in this thread and you'll find the pinout that Zucca posted and voltages he measured.

I soldered a few flying leads to various points on my board and plugged it in to measured and verify the 3.3V pin on the edge connector and it was correct as per Zucca's post.

Seems like quite a few folk have had similar problems to yours - so that's why I'm wondering if the incorrect center tap voltage is the issue.

I'm pretty sure (could be wrong...) that the dunkemhigh board has the center tap pins tied together and ONLY to a cap that is grounded on the other end - i.e. there is no DC bias being injected into the center tap.

cheers,
george.

[PlainName]

I wonder if the board works for me due to the cap I used, which is visible in the photo here: previous EEVblog post. Doesn't seem likely, but maybe I should try and break it...

[WVL_KsZeN]

Weird, I haven't got it work yet, wonder why dunkemhigh's board is working. Maybe I should send mine to dunkemhigh to compare in his scope?

[PlainName]

Maybe I should send mine to dunkemhigh to compare in his scope?

If you wanted to do that, I'd be happy to try it.

[WVL_KsZeN]

OK, I'll send it to you :-)

[iRad]

OK, I'll send it to you :-)

Great, I guess I will wait to see what dunkemhigh finds out first.

[iRad]

The 3.3V comes from the edge connect. Go back a few pages in this thread and you'll find the pinout that Zucca posted and voltages he measured...

I'm pretty sure (could be wrong...) that the dunkemhigh board has the center tap pins tied together and ONLY to a cap that is grounded on the other end - i.e. there is no DC bias being injected into the center tap.

Thanks, I see those posts and will try and check it out when I get a chance.

It seems several of us are having the same issue with the board. Hopefully WVL_KsZeN can get his board back to dunkemhigh quickly and the problem found.

[JimHorn]

So far, I've found how to set up version 2.35 on a USB drive - but not 2.37 which is the only version Agilent provides anymore. I'd like to try the mods described earlier on my 02.35.201... version DSO 3014A but they don't work with the 2.37 image (I'm sure the infiniiVisionCore.dll locations to change have moved, among other things).

Does anyone have ideas on how to either modify the 2.37 image to enable options -or- where I might find a copy of the 2.35 version?

Many thanks to all -

[ben_r_]

Oh wow, so theres a 2.37 now? When did that come out? Last I had heard of was 2.36. And I have to say, you guys that have been working on the LAN card have been doing an awesome job, but seriously, all that work has made me very glad I just bought the VGA/LAN card when I did and jumped on it when I was able to get a good deal! Keep it up though as Im sure it will help lots of people out there!

[Zucca]

Here the 2.37

http://www.home.agilent.com/agilent/software.jspx?ckey=2014548&lc=eng&cc=US&nid=-33573.970736&id=2014548

Released on the 21st May 2014

[kilobyte]

If someone needs the old 2.35 please send me a pm.
I have uploaded the cab file on my webspace.

[Sparky]

Oh wow, so theres a 2.37 now? When did that come out? Last I had heard of was 2.36....

Some info for all: firmware v2.36 was never available for download from Agilent; it only ever came pre-installed on new units or refurbs.  The previously downloadable version was v2.35.

I've attached the release notes for v2.37 --- there are a few enhancements, especially for power application measurements.

Previous attempts by members in the forum, myself included, were unsuccessful to get v2.36 working with the USB boot for patching options; other than USB boot not working, it appeared changes in "ProcessStartupFolder.exe" were involved, but this file could not be read from flash.  Anyone want to look at v2.37.  I will post back some findings after I look at the new firmware.

[Sparky]

For those interested, I have uploaded the "infiniiVisionCore.dll" from firmware releases 2.35, 2.36 and 2.37.  The file is ~50kB larger in 2.37 than the previous two versions (which are the same size, but different).

Since 2.36 did not allow USB boot it would be surprising if 2.37 allowed it again...  I have not tested installing 2.37 and seeing if a downgrade to 2.35 is possible...

From previous investigation, it seems \Windows\ProcessStartupFolder.exe in 2.36 may have been modified to prevent booting from USB drive.  I made a lot of attempts to copy ProcessStartupFolder.exe from internal flash when I had 2.36 installed, but it was not possible.

Is there any way to obtain this file from the 2.37 CAB?  I could not find it in infiniiVisionSetup.cab, so perhaps it is in another file (NK.BIN.COMP?)

[Zucca]

After getting a request from someone here what I did:

1) Upgrade from 2.35 to 2.37: OK!
2) Test SCPI on port 5024 with FW 2.37 (require LAN Module): OK!

Code: [Select]

Welcome to Agilent MSOX3054A InfiniiVision - MY5113XXXX
>>*idn?
AGILENT TECHNOLOGIES,MSO-X 3054A,MY5113XXXX,02.37.2014052001

3) downgrade from 2.37 to 2.35: FAILED!

See screenshot below, forgive me for the focus

[PlainName]

3) downgrade from 2.37 to 2.35: FAILED!

Erk! That's a bit of a one-way trip. We should all be grateful you've tested that, at the expense of not being able to muck about with your kit now

[eurofox]

3) downgrade from 2.37 to 2.35: FAILED!

Erk! That's a bit of a one-way trip. We should all be grateful you've tested that, at the expense of not being able to muck about with your kit now

Smart done from Agilent 

[iRad]

Once original Agilent Option Licenses are installed in a scope, there is no need to reinstall them each time the firmware is updated. The new firmware keeps all your existing licenses intact. With this being the case, I’m curious… Has anyone loaded the hacked 2.35 firmware internally to their scope (not just booted from USB), then updated to the 2.37 firmware. If so, were any or all the upgraded options kept, as they would be in a scope with original Agilent licenses?

If so, this would be a good scenario for anyone with 2.35 or lower, as the options could be loaded and then the scope updated as future firmware and features are released. It would not be good for newer scopes already loaded with firmware later than 2.35.

[Zucca]

were any or all the upgraded options kept, as they would be in a scope with original Agilent licenses?

In my case I paid $$$ for the options and I did not hack the scope. Of course I was not worried at all to test the upgrade to 2.37, since I was sure the options were still there no matter what.

Good Luck for all your hacked scopes, in my experience it is just a matter of time so be patience, don´t give up with the reverse engineering and don´t lose hope.

[mwilson]

I’m curious… Has anyone loaded the hacked 2.35 firmware internally to their scope (not just booted from USB), then updated to the 2.37 firmware. If so, were any or all the upgraded options kept, as they would be in a scope with original Agilent licenses?

Because the hack isn't really "installing" licenses, it's modifying the actual oscilloscope software to make it think certain licenses are enabled, the firmware upgrade will replace that oscilloscope software with a non-modified version, so the scope will no longer think the licenses are enabled. (For example, I did install the modified 2.35 to the internal memory but whenever I wanted to undo it to try a different method or experiment I simply re-installed the 2.35 firmware from Agilent and all the hacked license features were gone.)

[iRad]

Because the hack isn't really "installing" licenses, it's modifying the actual oscilloscope software to make it think certain licenses are enabled, the firmware upgrade will replace that oscilloscope software with a non-modified version, so the scope will no longer think the licenses are enabled...

OK, that make sense. I guess keeping 2.35 installed is a good idea, at least until something is able to be done with 2.37 (or later). Don't see any compelling reason to go to 2.37 anyway.