DG4000 - a firmware investigation

[KedasProbe]

I just got an accurate 10MHz source (Leo Bodnar, Mini) and I thought about calibrating the freq of my DG4102  (not hacked, yet, FW1.07 HW1.01)

In the manual they mention:

Warm DG4000 up for at least 30 minutes. Press Utility Test/Cal
Secure Code, use the knob and direction keys to enter the correct password, press
Secure and the calibration safety protection turns off.

Do we know what this secure code is? (I didn't find it in this thread?)

The internal frequency is only off  by 1.2Hz too low though.

[Orange]

I just got an accurate 10MHz source (Leo Bodnar, Mini) and I thought about calibrating the freq of my DG4102  (not hacked, yet, FW1.07 HW1.01)

In the manual they mention:

Warm DG4000 up for at least 30 minutes. Press Utility Test/Cal
Secure Code, use the knob and direction keys to enter the correct password, press
Secure and the calibration safety protection turns off.

Do we know what this secure code is? (I didn't find it in this thread?)

The internal frequency is only off  by 1.2Hz too low though.

Secure code is 2010 - this will enable the Test/Cal Submenu on the DG4062/FW 00.01.04

[KedasProbe]

I got access to the calibration menu and tried to update the frequency but something isn't right.
When I press the up or down button it adds a lot of jitter so I didn't press save.
Pressing the other button doesn't restore it to what is was. (1 up + 1 down should have changed nothing)

It stayed in this bad situation until I power cycled,  I didn't press save.

I'm just going to keep my 10Mhz attached to it.
I only find it annoying that it changes the 10Mhz ref input to output without asking my OK, if it doesn't see a signal.
It would be ok if it also switch back to external when it sees a signal. (power on delay)
So it's very easy to end up with two signal outputs connected and not be aware of it. (not good for my reference output)
(I used my DHO914S Gen to see what it did since I didn't want to risk my 10Mhz ref)

Yellow is my stable reference, triggered on blue = Out CH1 of DG4102

[KedasProbe]

Anyone know where I can find the odler firmwares for the DG4000?

I need an older one to be able to upgrade starting from 1.07
On the Rigol website they say contact Rigol in case of an older firmware but they just ignore my email.

I understand that I need to upgrade the bootloader first before I can update to the latest versions.

[RoGeorge]

Have you tried the modified .GEL, from TV84 in this post:  https://www.eevblog.com/forum/testgear/dg4000-a-firmware-investigation/msg2828866/#msg2828866

[Darkover]

I have V1.04, V1.06, V1.07, V1.08, V1.09, V1.10, V1.12.
Give me your mail and I can send it to you.

Olaf

[KedasProbe]

I didn't try the hacked version, I prefer to stay official for now, I see I can always do it later if needed.

It has the advantage if something is wrong I can blame Rigol instead of myself

[RoGeorge]

AFAIK, firmware downgrades are not possible when using original Rigol firmware.

You must use the hacked .GEL attached by TV84 in order to downgrade from your current version, then run the 'cengen' on a Windows machine.  Once the instrument is unlocked (the *IDN? name and the model reported by your AWG will become DG4202), the instrument will remain as DG4202 for any further firmware updates.  Then, update the instrument to the latest firmware from Rigol.

Read a few posts before and after the TV84 post linked above, for the exact instructions.  The hacked .GEL is inside that 'DG4000 ... .zip' file attached by TV84 at the end in that post.

[KedasProbe]

But I need boot loader version 6 first anyway, not included in the zip file. (currently I have v4, firmware 1.07)

[RoGeorge]

1. Did you run 'cengen' on yor computer already?  Did you manually input the generated licenses in your instrument (input the generated numbers only, no hack on the instrument is needed for now)?  What message do you get?

2.  What versions do you have now inside the instrument?

-To see FW version press 'Utility' -> 'System' -> 'System Info'
-To see detailed version press 'Utility' -> 'System' -> 'System Info' -> 'G1' ->'G3' -> 'G5'
       where G1...G5 are the grey buttons on the right of the screen

[tv84]

Bootloader 00.06.02 attached.

[KedasProbe]

I haven't updated it yet, I got this answer from Rigol:

Generally speaking, we do not recommend the upgrade of the DG4000 firmware with series before 00.01.08.002 or the keyboard version before 05.01.
After upgrading the DG4000 using the bootloader program, the calibration data will be cleared, and the unit needs to be recalibrated. This cannot be completed on the customer side, as bootloaders are usually not directly developed for the customer's use.

However, if customers emphasize on upgrading the firmware, the unit will have to be returned for replacing the DG4000 mainboard. This will imply a charge of the service.

So I would lose calibration data when I load the bootloader, but not sure why the mainboard would need replacing.

[Aleks Alekseev]

I can't update the firmware from version 1.09 to the modified 1.08.  Can anyone give me an exactly working modified version?

[tv84]

I can't update the firmware from version 1.09 to the modified 1.08.  Can anyone give me an exactly working modified version?

Is your bootloader v06? If not, update the bootloader before trying the patched version.

[Aleks Alekseev]

software version  00.01.09
fpga version        00.01.09
hard version        01.03
keyboard version  06.01

[tv84]

Help yourself:

1. Try flash downgrade FW again.
2. If OK, license it and upgrade FW again.
    If NOK, change USB disk and go to step 1.

[Aleks Alekseev]

Everything worked out for me. I took a flash drive from friends and everything was flashed from it. My 5 flash drives unfortunately did not work

[Blanc]

This is my first time writing.
This is a question regarding a problem with RIGOL DG4062.
I bought a used DG4062. Firmware is 1.04. (I'd like to update this to 1.08 somehow, but I can't get it.)

I posted this because I wanted to know if the following problem can be improved by updating the firmware. RIGOL's official website doesn't seem to have a history of bug fixes, so it's impossible to determine whether this is a hardware failure or a software bug.

When testing, I found that for CH1, the output was largely discontinuous at output impedance settings other than High-Z (for example, 50 ohm). If you gradually increase the output from the minimum output (-56dBm), the output will be abnormally large in the 4 steps before the relay switches. When the relay switches, it returns to the continuous increasing state before the abnormal value. Also, in the High-Z output, the output is slightly discontinuous between 3.5mV-4.0mV and 80mV-110mV.

If it's a hardware failure, I'd like to try to repair it, but unfortunately I don't have a circuit diagram, so I think it will be quite difficult. If anyone has any clues, I would appreciate it if you could let me know.

[TurboTom]

Welcome to the forum!

Changing the "output impedance" doesn't change the AWG Impedance in reality (it always stays at 50 Ohms). What changes is the calculated output level / voltage to represent the correct figure when you attach a load as specified in the "output impedance" setting. The menu item should rather be called "load impedance". If you find the "real" output voltage of your AWG to be discontinuous, this may well be a problem with calculating the output levels for "non-standard" load impedances, but it may as well be a problem of inaccurate calibration of the instument.

Regarding this issue, you are lucky since the DG4000 is one of the few instruments with a "real" calibration guide available (attached to this message).

I uploaded the firmware package 00.01.12.00.02 that includes a bootlaoder, here's the link. After this, it should be possible to install the latest F/W 00.01.14.00.01.

All the best and good luck! 

P.S.: ...and it can be "improved" to full DG4202 specs... 

[Blanc]

Thank you for your reply.
Reading your explanation, it seems like the DG4000's discontinuity is coming from the bug of calculation software and/or incorrect calibration. A kind person gave me an archive of old FWs containing the bug fix history. Although the history was not detailed, it seems likely that updating the FW and performing appropriate calibration will resolve the issue.

Unfortunately, I don't have a USB memory that the update process of DG4000 can recognize, so I haven't updated the FW yet.

Thank you again giving  me a manual for the calibration.
Fortunately, I already have a six and a half digit DMM, but no high frequency power meter.
However, I don't think the accuracy of high frequency output is much of a problem.

It's good for me that my troubles are not due to hardware.
I have already opened the cabinet and checked the PCB.
Seeing the small surface-mounted resistors and relays seemed a bit cumbersome.

[RoGeorge]

By the high Z question, this is probably your first AWG generators.  Most probably the generator is calibrated and working fine.  The amplitude is constant only when the cable is also 50 ohms, and properly terminated in 50 ohms, otherwise the output can be very different than expected.

You may want to read about terminators, reflections, transmission lines and standing waves.

[Blanc]

Thank you for your advice.

As you know, DG4000 has 2ch outputs.
I compared output voltages of ch1 and ch2.
These settings were the same, set to 50 ohm output.
I then connected them to a 2 channel 100Mhz oscilloscope.
Its each channel setting was the same and connected through 50 ohm terminators.

Then, let's try increasing the output power of DG4000.

I found that the voltage on ch2 rises smoothly,
but the voltage on ch1 rises discontinuously.
If the cause of the discontinuity is due to impedance mismatch,
I believe that ch1 and ch2 should have the same discontinuity.

This is the reason I believe Ch1 has some defect.

[TurboTom]

You may be onto something here that's actually not a bug but just a performance limitiation. As it appears, you find these inaccuracies only at low output levels. Please verify your findings vs. Rigol's "Performance Verification Guide", pg. 2-4 ff. The PDF is attached.

AWGs aren't meant to be precise low-level sources (like RF signal generators) but rather the "Jack of all Trades". Already the ground-reference (PE) of the outputs on most of the models shows that real low level signals will have to fight ground-coupled noise interference as soon as several instruments are interconnected. There are some AWGs available with outputs that are totally isolated from ground, but those aren't entry-level instruments (which one may argue the DG4000 isn't either, but pricing-wise, there's still a big margin to "professional" grade instruments...). A good signal tansformer to isolate the output may be helpful but probably won't solve the observed inaccuracies.

I just compared my own DG4000 regarding your mentioned inaccuracies, but i don't find anything out of the ordinary. Both channels match pretty well around the "switchover points" of the attenuator chain, though there may be a tiny amount of discontinuity. What worries me more is that my scope has some bad discontinuity at its gain switchover point on its first channel ... more than 4%! And the "autocal" function won't iron it out...  The other three channels are okay. The scope is a DHO1000 btw, but that's OT.

[Blanc]

Hello everyone.

Thank you very much for your supports and advices.
Last night, I found a USB memory that can refresh the FW of DG4000 among a lot of old USB memories.
After perusing this thread, I managed to get the latest "well tuned" DG4000 :-).
During the FW update, I was frightened by the lights of buttons turning on and off many times.
I made two mistakes along the way.

This morning, I calibrated the DG4000 again.
The attached photos are the conclusion.I set the AWG 50 ohm output, 1kHz.
Then I increased the amplitude in steps starting from -56dBm.
You can see abnormal rises and falls in the amplitude.
This symptom only occurs on ch1. I think this is evidence of a hardware failure in ch1.

I understand that the DG4000 is a hobbyist device, not a professional device,
and the amplitude of the AC frequency is sensitive and affected by various environmental conditions.
I think the photo shows the discontinuous of amplitude exceeds what would be the case in such conditions.

[RoGeorge]

DG4000 is a professional AWG, not a hobby one.  Its specifications are not a lie.  The instrument is expected to have the performance told in its specification/manual.  You probably have a defective DG4000.

Please open a new topic about repairing/calibrating your DG4000.  People here avoid answering in a hijacked topic.  This topic is about the firmware of DG4000, in general, not about repairing a particular DG4000 unit with a defective channel.