changing the rigol DS1052E to DS1102E using USB , the dummy guide

[willfarnaby]

+1 successul (everything appears to work fine) in applying the mod, including final upgrade to the latest firmware. Thanks all.

[Polossatik]

Just received my Rigol today, it too had firmware 2.04 SP1. I took the plunge and followed the guide with no issues.

glad it went ok and thank you for letting it know !
I have updated the guide.

System info lists DS1102E and software version 00.02.04. I'm guessing I'll have to wait for the 2.04 SP1 firmware to popup? It doesn't seem available.

no, so far the 02.04 SP1 firmware has not popped up yet as a file set.

[Skappy]

Hi,
please apologize my poor english but i'm french...
I'm sorry to disturb you but i'm about to buy a RIGOL DS1052E oscillo with a 02.04.00.03 firmware version.
Will the upgrade to the DS1102E will be possible ?
I can't no find information concerning this firmware version ...
Thnak you very much for your help.
Have a nice day
Fabrice

[Polossatik]

Hi,
please apologize my poor english but i'm french...
I'm sorry to disturb you but i'm about to buy a RIGOL DS1052E oscillo with a 02.04.00.03 firmware version.
Will the upgrade to the DS1102E will be possible ?
I can't no find information concerning this firmware version ...
Thnak you very much for your help.
Have a nice day
Fabrice

simply follow the guide on page 1 and choose in point 3 the "3)c) your firmware is HIGHER than 00.02.02 SP2." section , it mentions explicit 00.02.04.00.03

[Skappy]

Hi,
please apologize my poor english but i'm french...
I'm sorry to disturb you but i'm about to buy a RIGOL DS1052E oscillo with a 02.04.00.03 firmware version.
Will the upgrade to the DS1102E will be possible ?
I can't no find information concerning this firmware version ...
Thnak you very much for your help.
Have a nice day
Fabrice

simply follow the guide on page 1 and choose in point 3 the "3)c) your firmware is HIGHER than 00.02.02 SP2." section , it mentions explicit 00.02.04.00.03

Thank you very much for that, I will read it carefully now. Many thanks again

[yyao]

Hey guys, I just got the DS1052E scope in the mail 2 days ago. When I go to the model screen  my software version seems to be 2.04 SP1.

Is this a new version or did I just not read the instructions well? Has anyone tried downgrading using this version ?

Yep same here. I just got my scope 2 days ago with the same firmware. Couldn't downgrade it to 2.02 SP2 though. The scope refuses to downgrade to any revision that has a lower version number, but this is easy to get around. All you have to do is load the 2.02 firmware into a hex editor and change the apparent version number to something higher than 2.04.01. Once you downgrade to 2.02, you can easily apply the fix that has been mentioned in Dave's video and upgrade to 2.04. Ingenious, eh?

Like Professor Collins said, the root word of engineering is ingenuity. Come to think of it, he was my civil engineering professor! And he is an Aussie too! He hated to be called professor though, because he thought it was too academic and engineers needed to consider real life problems. Oh boy, its funny, the things that you remember...

The problem now is, after scope self calibration, the scope is showing incredible amounts of ADC mismatch which manifests itself as noise on a trace. The mismatch is quite significant too, almost an entire vertical div.  I am convinced that this is ADC mismatch because:

• It doesn't happen my friend's oscilloscope and he hasn't performed self recalibration yet
• It only happens when the combined sampling frequency is higher than 100Msps
• The noise is periodic and repeats every 10 samples in single channel mode and every 5 samples in dual channel mode
• The noise takes up about 1 div regardless of the vertical scale - e.g. it shows up as 1.6mV noise on 2mV/div scale and 1.6V noise on a 2V/div scale

Has anyone else noticed this problem? Is there any way to undo the calibration? I'm going to try to revert my changes to see if there's something more to it.

[yyao]

I would like to correct myself. I said:

• It only happens when the combined sampling frequency is higher than 100Msps

I meant to say:

• It only happens when the sampling rate on both channel are both higher than 100Msps

OK, so I went back and downgraded the firmware and model number. After a few scary moments with blank screens, I rebooted the machine and redid a self calibration. The same problem happens even on the old firmware and with the machine functioning as a DS1052E.

Perhaps there is something special with the 02.04 SP1 firmware with self calibration and maybe Rigol made a slight hardware modification in these new units. Perhaps I just have a unit with a bad self calibration function. Either way, I'm too scared to run self calibration my friend's unit.

Until I figure this out, I'll have a 50MHz 100Msps/ch dual channel scope. Oh well...

[Mechatrommer]

@yyao: are we in the same fate here? can you post pictures of the problem you are talking about? mine here http://www.rcgroups.com/forums/showthread.php?t=1222045

[dimlow]

Sounds like is the same problem i have too, lets hope Drieg comes out with his solution soon.

[Mechatrommer]

Sounds like is the same problem i have too, lets hope Drieg comes out with his solution soon.

Should we pay him when he got the fix? o i think we should! but i hope it will much quicker as buying a new brand DSO is already emerging in my mind. When i do, i think i want to burn this DS1052E to ashes or doing a full scale autopsy on it!

[yyao]

Shafri,

It looks like we have the identical (or very similar) problem. Do you remember what the firmware version of your oscilloscope was prior to changing the model number? Did you try self-calibration prior to the model changes? I am quite sure that it is not an issue of you trying to convert your scope to have the logic analyzer function, because I didn't do that and got the same result.

I'm uploading some pictures to demonstrate my theory about ADC mismatch. You can see that the frequency of the repetitive noise is dependent on the sampling rate, specifically, f_noise = f_sampling / 10. You can also see that the same pattern repeats regardless of the vertical scale. I couldn't take a clear picture of this, but if you put the display mode to draw dots, you will see that there are 10 samples for each period of the noise waveform.

I'm kind of stumped on this. You could easily calibrate this sort of stuff out if you wrote your own firmware, but this does not seem to be an alternative right now. You could do some clever signal processing if you captured the data to a computer and compensated for the mismatch, but this will not work for all signals and defeats the purpose of a benchtop oscilloscope.

[Mechatrommer]

pls try to change to 2V/div setting and measure something, what will be the reading? and then measure something else, reading again? pls post picture at open probes (0V, and again, remember..... 2V/div setting)
opps: my firmware was 2.02 IIRC, during the purchase 2.03 or 2.04 is not in existence yet and i bought it just days after Dave publish the softmod.

[Mechatrommer]

i just tried your setting, about the same noise, welcome to the brotherhood, dimlow is our elder brother and remember to do my setup above (2V/div)
so far from info: my speculation is... the FPGA store some table (calibration), but when the sh*t happens, the Blackfin will shift or modify or erase or whatever wacky on those table. but dont expand a discussion, its pointless!
since i got a pointer from drieg indicating my problem is not as simple as re-flashing the spansion where the FW lies.

[yyao]

OK, so here's the pictures that you requested. The signal is some sine wave approximately 1MHz straight from a function generator. I captured one screen shot at 250Msps (lots of noise) and another one at 100Msps (no visible noise).

I am still quite sure that self-calibration is doing something. One time, after downgrading the model number, the vertical scale was all messed up - huge DC offset, vertical gain was non-responsive. After a self calibration, the vertical scale worked properly - which is what you see now.

Shafri, do you remember the firmware version when you bought your machine?

Can someone who bought a new DS1052E recently perform a self calibration and tell us what you see when you sample @ 1Gsps? Performing self calibration on an unmodified machine is totally legitimate and is recommended by the manual. It will not void your warranty.

[dimlow]

Sounds like is the same problem i have too, lets hope Drieg comes out with his solution soon.

Should we pay him when he got the fix? o i think we should! but i hope it will much quicker as buying a new brand DSO is already emerging in my mind. When i do, i think i want to burn this DS1052E to ashes or doing a full scale autopsy on it!

I would willingly give him a payment if it fixed my scope Im not in any rush as i have not used the scope at all for the last few months.

[Mechatrommer]

Sounds like is the same problem i have too, lets hope Drieg comes out with his solution soon.

Should we pay him when he got the fix? o i think we should! but i hope it will much quicker as buying a new brand DSO is already emerging in my mind. When i do, i think i want to burn this DS1052E to ashes or doing a full scale autopsy on it!

I would willingly give him a payment if it fixed my scope Im not in any rush as i have not used the scope at all for the last few months.

me too. as the DSO is not mission critical for me to get the vertical reading correct, i'm mostly doing digital. but recently when i checked my brother car sound system, the vertical reading kind of pissed me off! still too busy with my daily life.

@yyao: you are still good at 2V/div where for my case its totally useless! so we just half brother haha! i mentioned my FW was and is 2.02. never have a taste to upgrade to 2.04 since people only say its only for "downgrade crippled" version.

[Fraser]

Picture attached of my new Rigol DS1052E running firmware 2.04 SP1 unmodified.

Settings as per previous photographs to provide direct comparison to faulty unit. No issues evident after self calibration.

[yyao]

Settings as per previous photographs to provide direct comparison to faulty unit. No issues evident after self calibration.

Neat, thanks Aurora Comms. Are you going to modifying the model number of your unit? I'm not saying you should, but if you do, could you post the results of your modification?

[d0ss]

Well, after discovering the same problem I've deleted my post confirming the hack works with 2.04 SP1 ok.

My unit was a 2.04 SP1 which had no issues after self calibration (unmodified).

I'm fairly sure this issue did not appear after modification but only after I performed a self calibration, attached a picture below.

[Mechatrommer]

thats a weirdo (2nd pic). i've never seen that on my rigol.

[Fraser]

I can almost hear the sales of DS1052E's plummeting... are the good times over for the affordable 100MHz hobbyist DSO I wonder ?

If, and I mean IF, Rigol have resorted to what is effectively a manufacturer induced 'self-destruct' on the units calibration as a result of the firmware ID change, they are not the company I thought they were. It isn't like you were hacking the OS or core firmware data is it.

I will leave my 1052E in it's 50MHz state for the moment but if I get hungy for 100MHz bandwidth I will just hardware mod the sucker to overcome that pesky varicap diode that hobbles the bandwidth. Rigol... put that in your pipe and smoke it !

[yyao]

thats a weirdo (2nd pic). i've never seen that on my rigol.

Take a buffer's worth of data. Stop the sampling. Turn on the Delayed option in the horizontal menu. Zoom in vertically. Useful for zooming in on trace details.

[Mechatrommer]

but i wonder why i cant go down to 10mV where Aurora can

[Fraser]

Please can an owner of a DS1052E who has upgraded to 100MHZ B/W on a unit supplied with Firmware 2.04 (NOT SP1)  check to see if they have any issues at the settings that are shown in the above pictues. The issue may not have been obvious to some owners and this could potentially be a wider problem than just 2.04 SP1.

Another thought. When owners upgraded the DS1052E fitted with 2.04 SP1 firmware it was not possible to re-install 2.04 SP1 as that has not (to my knowledge) been released. We should consider the possibilty that 2.04 SP1 contains amendments for a slightly different motherboard chipset that requires different calibration algorithms or settings. I have often seen new Firmware versions that were released for compatability with a new version of an IC that the manufacturer had switched to. Such updates are often backwardly compatable but older versions cannot correctly support the new IC. Just food for thought.

I am aware that some users of units that ran F/W 2.02 also have issues but I understand that these issues may stem from a corrupted upgrade process that may have caused issues in the firmware itself. Such failures may have no common denominator with the current 2.04 SP1 issue.  

[Fraser]

There would be no harm in emailing Rigol and asking whether they would be willing to supply Firmware 2.04 SP1. They may decline but nothing ventured, nothing gained.

Alternatively, when Rigol eventually release the next firmware revision that may fix the problem for you as it will contain SP1 plus new changes.