RIGOL DP832-trial is flawed?

[broderp]

I'm loving my new supply (from what little I can use it until I get some good leads  ) but there are a few quirks that I'm disappointed with.

RIGOL was gracious enough to give us trials of two features: Analyzer and Monitor.  I was even more happy to see in Dave's video the trial was good for 35 hours.    More than enough right?  Nope.

As it turns out, the clock is ticking on those 35 hours ANY time the supply is on!     Maybe for those with a lot of experience with supplies and supply features, this is no problem.  You either know right away if you want or do not want it.  But some of us (me!  ) my last supply I used for projects was literally THIS:

http://www.amazon.com/Philmore-Multi-Voltage-Regulated-Power-Supply/dp/B004M45MHQ/ref=sr_1_107?ie=UTF8&qid=1429923033&sr=8-107&keywords=12v+power+supply

This supply has TONS of stuff for me to learn and it will be quite a while before I even consider or know if the two trials are of any benefit to me.  Based on the current model, always timing down when powered up, I will have used up this trial offer way before I'm ready to use them. 

Would it not have made sense to allow the owner to start the trial when they want to?  Or even better only time down each trial when they are being actively used?  I would even settle for 10 hours with this model.  As it is I may never know if I want to buy the upgrade as I will never get to try it.  I won't buy it if I can't return it if I don't like it or it doesn't work out for me.  So RIGOL has just potentially lost this future sale from me.

Just ranting...ignore me.

Anyone else have any peezs on this supply? 

[dadler]

The device does not have a real-time clock. This is the best they can do with the hardware-it writes to the I2C FRAM chip at a specified interval as a function of the main clock frequency. It does not know the true time.

Just use riglol and unlock the features.

[kjdotts]

Valid points of contention, however, I never worried about this because I purchased my Rigol DP832 knowing I could easily have all the options for no additional costs (RigLOL). Naturally, I waited to make the options "Official" until I knew I was 100% keeping my unit and everything was good to go (cannot be reverted, could affect warranty coverage). With Rigol being a Chinese manufacturer, I have absolutely no problem justifying this decision.

-Kris

[broderp]

The device does not have a real-time clock. This is the best they can do with the hardware-it writes to the I2C FRAM chip at a specified interval as a function of the main clock frequency. It does not know the true time.

Just use riglol and unlock the features.

Just learned what this was....

Valid points of contention, however, I never worried about this because I purchased my Rigol DP832 knowing I could easily have all the options for no additional costs (RigLOL). Naturally, I waited to make the options "Official" until I knew I was 100% keeping my unit and everything was good to go (cannot be reverted, could affect warranty coverage). With Rigol being a Chinese manufacturer, I have absolutely no problem justifying this decision.

-Kris

Good points, I've had mine for a few days.  I'm 99.999% certain I'm keeping it.  So I gamble that I wont need warranty work or that I can uninstall it with SCPI later.  Seeing how this feature works and is activated, not sure even RIGOL can tell if the codes are from RigLOL or them. 

[kjdotts]

Good points, I've had mine for a few days.  I'm 99.999% certain I'm keeping it.  So I gamble that I wont need warranty work or that I can uninstall it with SCPI later.

After successfully installing all the options on my unit, I tested the SCPI method to uninstall option licenses (just as a test) and it failed (exactly as others have reported). I believe a firmware update at some point in the past broke the SCPI license uninstall option (maybe intentionally, maybe not).

Seeing how this feature works and is activated, not sure even RIGOL can tell if the codes are from RigLOL or them.

When a legitimate option license key is obtained, you visit THIS FORM and enter your unit S/N, so, they would know. However, whether a Rigol repair technician would bother checking is the unknown.

With that being said, I too was willing to accept the risk so I installed all the option license keys. It's a little tedious entering all those keys, so be careful, if you make too many mistakes, I believe you get locked out from entering keys for 12 runtime hours. Good luck!

-Kris

[broderp]

After successfully installing all the options on my unit, I tested the SCPI method to uninstall option licenses (just as a test) and it failed (exactly as others have reported). I believe a firmware update at some point in the past broke the SCPI license uninstall option (maybe intentionally, maybe not).

Seeing how this feature works and is activated, not sure even RIGOL can tell if the codes are from RigLOL or them.

When a legitimate option license key is obtained, you visit THIS FORM and enter your unit S/N, so, they would know. However, whether a Rigol repair technician would bother checking is the unknown.

With that being said, I too was willing to accept the risk so I installed all the option license keys. It's a little tedious entering all those keys, so be careful, if you make too many mistakes, I believe you get locked out from entering keys for 12 runtime hours. Good luck!

-Kris

Oh crap... So well I guess for now I can't uninstall.  I took the word of another forum user that it could be done.  I got all excited about the ability to do this that I rushed thru it without thinking it thru and doing some more research. 

I used SCPI to install the keys one at a time, using copy and paste.  Was pretty quick and painless.  It was a matter of seconds and the display went from "trial" to "Official".  I cycled power and confirmed everything was still good.

I guess RIGOL could technically know if their servers do not have record of the serial number.  Never crossed my mind, I knew from Dave's video that it was web based.  Maybe all their web activation process does is generate the keys like RIGOLOL, and have no recording or data base.  I also have to wonder that even if it did record this info that this info would be linked or shared to the repair centers or part of the repair centers validation of warranty beyond a proof of purchase of the actual supply.   

But if you think about it,  why would a non-oem license key invalidate the units warranty? Its just a key to turn the feature on.  It doesn't download or install the program to run any feature.  The features are already included in the unit.  It's not like installing a program on a computer that trashes the registry or causes operational errors.

That would be the same as Toyota denying my claim for a faulty motor sensor on my car because I have aftermarket wheels or installed a Kenwood radio instead of using the OEM one.

I hope..

[dadler]

Hmm, you used to be able to uninstall via SCPI--at least I was able to do it back originally (older firmware, though).

Can you just re-install the trial codes? I think all of this is discussed in the mega thread. Either way, someone has sent their device back to Rigol for warranty repair, generated codes and all, and not a word was said. Some people have even received "new" devices that came unlocked (probably customer returns re-sold as new). Voiding a warranty for typing a code into the front of the device seems extremely questionable, but imagine the PR nightmare if a warranty claim was denied--I just can't see that happening. This topic has been debated over and over again in the uber-thread.

Sorry if I steered you wrong on the uninstall, I gave you my best understanding. However it's unlikely you'll need to return your device if you are happy with it and it has been burned in/running for a while. I have not heard many complaints about the new DP832 hardware (with the better LM317 heat sink). New design seems pretty bulletproof for what it is.

Edit: should probably link to the mega-thread: https://www.eevblog.com/forum/testgear/sniffing-the-rigol%27s-internal-i2c-bus/

[kjdotts]

Oh crap... So well I guess for now I can't uninstall.  I took the word of another forum user that it could be done.  I got all excited about the ability to do this that I rushed thru it without thinking it thru and doing some more research.

Prior to actually trying the SCPI uninstall method and having it fail, I was like you, no big deal because I can just uninstall. Then, after installing the option license keys, I thought "lets just try that uninstall" and CRAP! The RigLOL keygen has command line options for generating "trial" licenses ("B" instead of "F" for the device option code first letter). However, others have reported that fails also. I actually did not try this method to revert back to trial licenses (I was too busy enjoying my new power supply ).

I used SCPI to install the keys one at a time, using copy and paste.  Was pretty quick and painless.  It was a matter of seconds and the display went from "trial" to "Official".  I cycled power and confirmed everything was still good.

Excellent! I didn't have my workstation setup for SCPI access to the Rigol at the time I installed the option license keys so I just did it manually on the unit. Definitely much easier installing via SCPI. Thankfully, installing option licenses is not broken.

I guess RIGOL could technically know if their servers do not have record of the serial number.  Never crossed my mind, I knew from Dave's video that it was web based.  Maybe all their web activation process does is generate the keys like RIGOLOL, and have no recording or data base.  I also have to wonder that even if it did record this info that this info would be linked or shared to the repair centers or part of the repair centers validation of warranty beyond a proof of purchase of the actual supply.

But if you think about it,  why would a non-oem license key invalidate the units warranty? Its just a key to turn the feature on.  It doesn't download or install the program to run any feature.  The features are already included in the unit.  It's not like installing a program on a computer that trashes the registry or causes operational errors.

That would be the same as Toyota denying my claim for a faulty motor sensor on my car because I have aftermarket wheels or installed a Kenwood radio instead of using the OEM one.

I hope..

Yes, valid points. I moved on and stopped worrying about it knowing there are many unknowns and my unit requiring warranty service is an unlikely event. Should I ever need warranty service, I will simply prepare for the worst case scenario and hope for the best case scenario.

-Kris

[kjdotts]

...Either way, someone has sent their device back to Rigol for warranty repair, generated codes and all, and not a word was said. Some people have even received "new" devices that came unlocked (probably customer returns re-sold as new). Voiding a warranty for typing a code into the front of the device seems extremely questionable, but imagine the PR nightmare if a warranty claim was denied--I just can't see that happening...

Good to know, I thought I remembered reading about someone sending in their unit for warranty service with "generated" keys and nothing was said. I totally agree, very unlikely to have issues with Rigol warranty service and "generated" keys.

-Kris

[broderp]

Sorry if I steered you wrong on the uninstall, I gave you my best understanding. However it's unlikely you'll need to return your device if you are happy with it and it has been burned in/running for a while. I have not heard many complaints about the new DP832 hardware (with the better LM317 heat sink). New design seems pretty bulletproof for what it is.

No worries!    I didn't intend to throw you under the bus nor did I say who had told me this.   
It was my decision to do.  If I had bricked my supply, it would not be anyone's fault but my own.

Even in my haste to get functionality in the name of not loosing the ability to use the trial stuff, I realized this was pretty mush a safe bet.  The one thing I wanted the most, the precision, I did last.

[HiassofT]

FYI: you can uninstall all licenses on the DP832 with this SCPI command:

Code: [Select]

:LIC:CLEARYou won't see any confirmation on the screen, but also no error message like incorrect remote command :)

so long,

Hias

[kjdotts]

FYI: you can uninstall all licenses on the DP832 with this SCPI command:

Code: [Select]

:LIC:CLEARYou won't see any confirmation on the screen, but also no error message like incorrect remote command

so long,

Hias

Oh really? Excellent! This is the first I've heard of this command. I had only tried...

Code: [Select]

:SYSTem:OPTion:UNINSTall
-Kris

[dadler]

Oh wow-was this just discovered? Very cool!

[pascal_sweden]

The device does not have a real-time clock. This is the best they can do with the hardware-it writes to the I2C FRAM chip at a specified interval as a function of the main clock frequency. It does not know the true time.

Even in that implementation, it should be possible to implement it slightly more intelligent, in such a way that the clock event is only committed and registered when you are actually using the specific trial feature of the scope.

They could implement this with a Finite State Machine, that keeps track in which mode of the scope you are. They can do the actual registration at the lower level in hardware, but should put some more intelligence at a higher level in software, to mask the clock event when you are not using the actual trial option.

Ideally they should split this for the different trial options, so that you can wear out one trial option faster than another trial option, and use the respective trial options individually at your own pace.

[dadler]

The device does not have a real-time clock. This is the best they can do with the hardware-it writes to the I2C FRAM chip at a specified interval as a function of the main clock frequency. It does not know the true time.

Even in that implementation, it should be possible to implement it slightly more intelligent, in such a way that the clock event is only committed and registered when you are actually using the specific trial feature of the scope.

They could implement this with a Finite State Machine, that keeps track in which mode of the scope you are. They can do the actual registration at the lower level in hardware, but should put some more intelligence at a higher level in software, to mask the clock event when you are not using the actual trial option.

Ideally they should split this for the different trial options, so that you can wear out one trial option faster than another trial option, and use the respective trial options individually at your own pace.

Then they would have to track the timer for each option separately. Of course it could be implemented differently, but these units are built down to a price and most of the changes you describe will require different or more expensive hardware. The easiest change would be letting the user activate the countdown of the trial clock, but I can think of edge cases and security issues with that.

We are lucky we get trials *at all*! Most of the other manfacturers do not offer trial options like this.

[broderp]

FYI: you can uninstall all licenses on the DP832 with this SCPI command:

Code: [Select]

:LIC:CLEARYou won't see any confirmation on the screen, but also no error message like incorrect remote command

so long,

Hias

That's great!  Did you do this on the current firmware?  I'll have to give it a try when I get a chance.  Should be able to remove the license then reinstall them with no issues....right?
. 

[HiassofT]

That's great!  Did you do this on the current firmware?  I'll have to give it a try when I get a chance.

I tested this both with firmware 00.01.11 and 00.01.13 and it worked fine. Not sure about earlier FW versions though.

Should be able to remove the license then reinstall them with no issues....right?

Yes, exactly.

so long,

Hias

[BloodyCactus]

is that command not in the SCPI reference in the manual? (I have no dp832 so I have not looked at the manual)

[broderp]

is that command not in the SCPI reference in the manual? (I have no dp832 so I have not looked at the manual)

The manual has about one page dedicated to the use of Sigma.  So I have to say the answer to that question is a no.  There may be another manual for the software, but I haven't seen it.  It would be nice if RIGOL would make available a more technical manual for the supply and the software. At least list all the commands and functions available for the software in the supply manual.

[dadler]

There is a programming manual that lists "all" the SCPI commands. It's on the Rigol site, titled "DP800 Programming Guide". It is 205 pages in length, but does not contain ":LIC:CLEAR", which if it works, may be undocumented.

[kjdotts]

There is a programming manual that lists "all" the SCPI commands. It's on the Rigol site, titled "DP800 Programming Guide". It is 205 pages in length, but does not contain ":LIC:CLEAR", which if it works, may be undocumented.

HiassofT posted in another Rigol thread about dumping the "SCPI command table" (see below quote), so I believe he performed the same analysis on the DP832 and discovered the undocumented command (my theory).

Re: Rigol MSO2000 series hacking

The SCPI command table seems to be located at around 0x00F0EC00. You can read it with ":SYST:UTIL:READ? 15789057,20000"...

so long,

Hias

-Kris

[HiassofT]

HiassofT posted in another Rigol thread about dumping the "SCPI command table" (see below quote), so I believe he performed the same analysis on the DP832 and discovered the undocumented command (my theory).

In this case it was a lot simpler, I found the ":LIC:CLEAR" command by a few educated guesses. Just tried "reset", "unset", "clear" and a few other ones and noticed that the DP832 didn't beep and show "unsupported remote command" on :LIC:CLEAR.

Haven't checked yet for other undocumented SCPI commands. Looking at the firmware GEL file didn't turn up anything (maybe it's scrambled/compressed) and getting a memory dump via SCPI, like on the DS2000 series scopes, doesn't seem possible.

so long,

Hias

[dadler]

The gel files are encoded. In the I2C hacking thread, someone posted a de-gel utility for decoding the files.

They are not full firmware images, but each section contains a header with the target IC and block address, followed by the data. This is because the Gel files contain firmware for different controllers and FPGA bitstreams.

[HiassofT]

The gel files are encoded. In the I2C hacking thread, someone posted a de-gel utility for decoding the files.

Thanks for the hint, but I already tried degel-0.1 from http://www.gotroot.ca/rigol/ before without success. The DP800 GEL file format is different, most obvious thing when examining the file with a hex editor is that there's no header with the model name at the beginning.

The bits I'm missing might be buried somewhere in the mega thread, I'll have a look at it when I have more spare time (probably not too soon...)

so long,

Hias