DG4000 - a firmware investigation

[Wim13]

I upgraded my DG4102 from fw 1.08 to fw 1.14, but the upgrade was different from the attached upgrade document.

I had, fw 1.08 and bootloader 0.4,
(for new readers, the bootloader version is on the system info page, under keyboard version !?!?)

First i had to find a working USB stick i have several, a 256Mb did not work a 2 GB older type did work with FAT32.

I tried to change fw first without loading a new bootloader, to see waht happens,(nothing)
sometimes it did nothing and utility and store button stay on.
and sometimes all kind of buttons began to flikker, nice, but nothing worse happens.
Later i discovered that it had to do someting with the setting save last powerOn.

So i updated the bootloader, the file is only 166Kb, it took several minutes of flikkering utility button,
then suddenly the waveform buttons went on, and in a second later the unit rebooted by it self, but ..
no changes in the system info menu ??

So i turned the unit off and on again,...black screen !??, nothing, rebooted again
still nothing, then i noticed that the mod, utility and store button were on again.....

Loooks it was waiting for something, and could not load fw 1.08 with the new bootloader !!
Then i put the USB stick with the fw 1.14 in the unit and watched wath would happen.

The utitlity button began to flikker for lots of minutes...not much activity on the stick..
and then suddenly the ramp button light up, and the flikkering of utility stoped.

Well, slowly all waveform buttons lights went on....and finally it rebooted by it self,
and a happy end, all worked oke.

So the attached file is not correct, in my case, between loading bootloader and fw.









 

[ralphrmartin]

Thank you for everyone's efforts on this thread, especially TV84. Having now gone through the steps here on my DG4062, and performed calibration as explained, on doing a sweep from 1MHz to 200MHz, I'm getting output which jumps up and down by about 5dB from 1MHz to about 30 MHz, is then flat ± 0.5dB to about 100MHz, then falls of pretty much linearly to about 4dB down at 200MHz.
Worse, for some combinations of voltage and frequency (e.g. 10MHz 200mV, 400mV) I get no output at all.

Have I missed something? Any suggestions for how to fix at least the issues up to 30MHz?

[ralphrmartin]

Here's a plot of the output versus frequency

[PA0PBZ]

This is the output of my liberated DG4062, I did not touch the calibration.
1dB down at about 160MHz and almost 3dB at 200MHz.

[TurboTom]

I guess I got lucky with mine... It started live as a DG4102. 

[bgm370]

I had the same calibration problem with my 4062 “enhanced” to a 4202 a while ago. Both 1.12 and 1.14 were not calibrating properly. I downgraded to 1.08 (it might have been an even earlier fw version) and it calibrated just fine. Then I upgraded it back to 1.14 and the proper calibration was preserved.

[ralphrmartin]

BGM, your suggestion certainly helped, although there's still something not quite right.
I now get a more-or-less flattish trace out to 100 MHz, and while I dont get so many weird jumps up and down now below 30MHz, I still get one jump down of about 0.8dB at 12 MHz or so.
There is then a linear-ish ramp down from about 100MHz to 200MHz, dropping a further 5dB.

I'm not too worried about the latter - perhaps the device just has some default values for this frequency range which wouldn't have needed factory calibration anyway.

However, the step at 12 MHz is rather more puzzling as I would have thought that would have been calibrated out.

On the other hand, the device is 7 or 8 years old now, so perhaps it has just drifted away from the original calibration.

PA0PBZ, I see if you look carefully, your plot also seems to show a bit of a drop at around 12MHz.

Perhaps I'll try to run through the full calibration procedure - but for both channels that seems to involve taking perhaps 300 power and voltage readings, and typing them in, which will be no fun when the DG's keyboard suffers from key bounce.

Thanks again for all the help.

[ralphrmartin]

Well, having gone back to 4.08, done the calibration with defaults, and got a not too bad trace, and come back to 4.12, I then spent a morning doing a proper calibration (with multimeter and spectrum analyzer). The net result is back to like that shown in my original sweep: several serious jumps of a few dB up to about 12MHz, flat-ish out to about 100MHz, and then a linear-ish slope losing 4dB by 200MHz. Where it's flat and linear are probably flatter and more linear than before - but the large artefacts, and slope, still exist.

I suspect the slope might be due to different circuitry in the (newer?) higher frequency models.

However, if I do a sweep not from 1Mhz to 200MHz, but from 1MHz to 12MHz, all the jumps go away. So, they are not a calibration issue, but some sort of artefact produced by the sweep function,. Different ranges of sweeps make jumps in different places. Ugly.

[tv84]

Let's call it a 30-100MHz model...

[ralphrmartin]

Haha. Gained 40Mhz, lost 30Mhz, for an overall win by 10Mhz! 

But even stranger - if I set everything back to the factory calibration - all of the jumps disappear....
This sounds to me awfully like there's a "differencing two big numbers to get a small number" kind of problem somewhere in there.

So I now have a 0 MHz to 110MHz model ... 
So apart from wasting a morning on a useless calibration,  I'm definitely winning!
And as long as I only want a narrow band at even higher frequencies, that's good too, as long as I check the exact output level separately.

[RoGeorge]

I see no reason why it should jump like that.  Maybe the calibration instruments are not calibrated, or maybe the measuring setup has something wrong in it, maybe a cable is faulty, IDK, but my unlocked DG4102 doesn't show any amplitude jumps in the first 30MHz or so (also never touched its calibration - I suspect the altered calibration might be the most probable cause for seeing jumps).

Does the amplitudes and frequencies of the jumps preserve when a narrower frequency sweep is performed?

[ralphrmartin]

It's the voltage settings (not RF power settings) which caused the jumps, and that was done with a 6.5 digit Agilent meter I've no reason to suspect to be out of spec. BNC coax lead to a BNC-banana plug splitter. Same coax lead to BNC-to-N adapter for spectrum analyzer. No problems with these leads etc in other work.

I've tried calibrating twice now, and both times ended up with jumps, once before going back to firmware 08 and redoing things there, and once after coming back to the latest fw.

Note that my machine doesn't show any jumps if you reset to the factory calibration.

Maybe the calibration procedure itself has some issues. It would be interesting to know if anyone has successfully recalibrated their DG4000 with the procedure given, with the current fw.

[RoGeorge]

Does the same Agilent voltmeter shows the generated voltage (after calibration) as correct (constant) when tested at various fixed frequencies?  If not, then the calibration procedure went wrong.  If yes, then it's either the voltmeter or the spectrum analyzer.

Voltage jumps of 3 or 6 dB should be easy to spot on any instrument.  Does the jumps show on other instruments, too, e.g. an oscilloscope or some other less precise voltmeter?

[ralphrmartin]

Thanks for your comment, RoGeorge.

Does the same Agilent voltmeter shows the generated voltage (after calibration) as correct (constant) when tested at various fixed frequencies? If not, then the calibration procedure went wrong.

True (as long as I use frequencies within the meter's AC range), but that doesn't tell us where it went wrong:
- voltmeter not working correctly (e.g. giving inconsistent readings at different frequencies)
- I misapplied the calibration procedure (failed to write some values, entered incorrect values by mistyping for example)
- there's a bug in the user calibration procedure code e.g. it
-- tries to measure voltages at frequencies outside the DMMs capabilities
-- does something wrong with the entered values (stores them in the wrong place, wrong format, etc)

If yes, then it's either the voltmeter or the spectrum analyzer.

Not necessarily. The software in the DG4000 in general, and the sweep function in particular, may make some assumptions about the calibration. It is certainly doing something strange when sweeping, as the jumps may or may not be present depending on exactly what frequency range you sweep. Intuitively, miscalibration should always lead to the same voltage at the same frequency, just the wrong voltage, and that is not happens. The voltage you get at a particular frequency depends on the frequency range of the sweep.

Unfortunately, because of the jumps, I reset it to the factory calibration (which doesn't have them), so I have now way of telling which of these possibilities it is, without wasting another 1/2 day to redo the calibration. Note that lack of jumps observed with the factory calibration indicates that the jumps are real in my calibration (you said to see if they exist using a different instrument).

Furthermore, the factory calibration does give readings on the voltmeter (using the same leads) from 1kHz to 100kHz (in steps of 1kHz) which are within 0.2% or better of the stated output of the DG4000, so I have no reason to think the voltmeter is playing up.

Perhaps the software makes some assumptions about the entered calibration values, and if the calibration values entered are too far from the expected ones, it breaks these assumptions

[ZETOR1]

Hello. I have dg4062 software version: 00.01.01. I want to update to 00.01.06. Where I can find 01.01.06. or 00.01.07?

[nive]

does someone sees this strange little peak on about 79.719 khz in a Sinus sweep from 1 hz til 100 hz kHz (0 dBM output) signal?

[nive]

i found out that it was a spurious signal inside my Siglent SA. 

[nive]

I confirm that the calibration on firmware 00.01.14 not works when you has updated your frequency with a license key in firmware 00.01.08.

Work around for me was going back to firmware 00.01.08 (with licence key installed), then applying a calibration for LFLAT en HFLAT. After saving this calibration and updating again to firmware 00.01.14 the DG4202 is flat within a 0,5 dBm.

[Yeson]

I hack the DG4062 to DG4202.The Amplitude is lower more than -4dbm from 60Mhz to 200Mhz. I havn't a power meter to do the calibration.  I record the factory default value. Can I edit some calpiont value of High Freq Flat to make the Amp more flatter.

[Yeson]

How do you calibration? Need Powermeter?

[TurboTom]

How do you calibration? Need Powermeter?

Rigol clearly specifies in their calibration guide (attched) the required instruments. For high frequency flatness, a power meter is required, and I don't see much chance for a replacement if you need the amplitude accuracy.

A spectrum analyzer may be a makeshift solution to get rid of the major errors you observe, but you'll rarely get better accuracy than +-1dB. Moreover, using an SA requires interconnection cables and probably adapters that all need to be of prime quality, and the input impedance of the SA may also not be accurate enough (though in the relevant frequency range, that's probably no real problem).

I got lucky with my DG4102 since apparently it's been calibrated all the way up to 200MHz in the factory already.

[Yeson]

From 20Mhz to 200Mhz。

[PA0PBZ]

My 'upgraded' DG4062, trace max hold. Not sure about the few peaks, maybe interference. I did not touch the calibration at all.

[TurboTom]

And here's my DG4000's channel 1 that started life as a DG4102 at three signal levels. Ch2 performs equally well.

[Kujo]

Thanks to all for their efforts and patience on this topic.

So noting that this thread has been quiet for some time i just wanted to add some USB drive combinations that ended up working out for me.

I initially tried SanDisk Cruzer 16GB usb 2.0 drives. They were recognised by the device in normal operation but would not load a new firmware file. I noticed that someone mentioned that they had used a microSD adapter. Now also remembering that someone else had mentioned that they felt the USB controller implementation at boot was maybe a bit off i grabbed the dodgy microSD adapted that came with my Ender 3D printer. I dropped a 16GB Strontium Nitro sd card (came with a phone i think) formatted to FAT32 with default allocation size and lo and behold. It worked a treat. Interestingly (well to me anyway) when updating the license file i used the original Cruzer stick and that worked fine. So this seems to support the fact that the USB implementation at boot is different from that once up and running. Anyway in case this is helpful to anyone in the future.

Kujo