Hacking the Siglent SDM3055 Bench DMM

[alexvg]

During my SDM3055 reverse engineering... I've got 2 questions to you ?

Would you agree to pay a small amount for a new SDM3055 software (and maybe for the SDM3045) ?
http://www.pixule.com/39863606081_would-you-agree-to-pay-small-amount-for-new-sdm3055-software-and-maybe-for-the-sdm3045.html

Would you like to improve your SDM3055 by adding a some options ? (The SDM3045 may be compatible with these options.)
http://www.pixule.com/398636486944_would-you-like-to-improve-your-sdm3055-by-adding-some-options.html

Thanks for your help !


Hi,

I'm working on a hack to show more digit to the Siglent SDM3055
I've got a source voltage of 2.5V (not a reference voltage)

Original : 240 000 points / 5 1/2 digits


First hack result : 2 400 000 points / 6 1/2 digits


The hack does not work correctly but give interresting result.
To be continued.

[tv84]

 

Alexia, as requested. 2 of the latest app versions attached. Good luck!

[alexvg]

New results with 7.5 digits
The calibration data are not currently loaded :

[Mr. Scram]

Nice experiment! Not to rain on your parade, but are these numbers relevant? Do we know whether the 6.5 digit versions maybe have cherry picked or simply better references? I'm asking because many meters can show more digits but it's generally noise. Don't let it discourage you though and be sure to hack on!

[alexvg]

I've noticed some mistake into the code :
A real value of 1.499995 is displayed 1.4999 instead of 1.5000.

But a value of 1.4999995 is displayed 1.5000. It's because the rounding is done with a value of 5.0x10-7

[coromonadalix]

I would love to see this hack against a AC / DC / Ohms  calibrator ??

[tautech]

Not to rain on your parade, but are these numbers relevant? Do we know whether the 6.5 digit versions maybe have cherry picked or simply better references?

Yes of course measurement stability can be questioned as SDM3065X uses the better LM399 reference.

[coromonadalix]

Could the 3065 and 3055  share the same design, minus the reference quality ?? 

Q: The hacked firmware, does it come from the 3065 ??

[tautech]

Images from SDM3000 series thread
https://www.eevblog.com/forum/testgear/siglent-new-bench-dmm-sdm3055/

SDM3045X


SDM3055


SDM3065X

[alexvg]

Could the 3065 and 3055  share the same design, minus the reference quality ?? 

The SDM3065X uses a LM399H (0.3ppm/C with 20ppm/1000h and 7uV noise) voltage reference and AD7175-2 "B" A/D (up to 24 peak-to-peak bits and 24 effective bits at 5 to 16 SPS).
The SDM3055 uses a MAX6325 "C" (0.5ppm/C with 30ppm/1000h and 1.3uV noise) voltage reference and AD7190 "B" A/D (up to 22.5 peak-to-peak bits and 24 effective bits at 4.7 SPS).
The SDM3045X uses TI REF5025K "High grade" (2.5ppm/C with 100ppm/1000h and 3uV noise ) and AD7190 "B" A/D (up to 22.5 peak-to-peak bits and 24 effective bits at 4.7 SPS).

Edit 2019-08-18
Difference vs SDM3055 :
- SDM3065X uses a separate power supply
- SDM3065X uses a LM399H voltage reference and a higher resolution ADC
- SDM3045X uses lower grade components but with a SDM3055 design (or SDM3055X-E PCB)

I suspect the SDM3055 could have a better ADC (see un populated U503 SSOP28) like ADS1256 (seems to be compatible)

Q: The hacked firmware, does it come from the 3065 ??

I'm only using the original SDM3055 firmware.

[tv84]

If anyone with a 3045X wants to test the 3055 in his DMM, I can easily prepare an .ADS.

BUT, no responsibility for my side.

[tautech]

I'm working on a hack to show more digit to the Siglent SDM3055

To be continued.....................

And she busted it ! 

PM pleading for help ...............
And returned.

[BillB]

Well, that escalated quickly!   

[alexvg]

I've done some bash correction to correctly use XML calibration file.
Now the value displayed is correct and the result is very interresting.

I'm investigate to display digits with a little lower font size.

With the input shorted to the ground :

[alexvg]

How to test the sdm-mod ? (only with the latest 1.01.01.19 firmware)

Step 1
Plug a formatted FAT32 USB Key (I use 4GB) into the SDM3055

Step 2
Connect to telnet (username:root, password:ding123)

Step 3
Backup all Siglent data into the USB key :
"cp -R /usr/bin/siglent /usr/bin/siglent/usr/mass_storage/U-disk0/"
After copying, unplug the USB key

Step 4
Plug the USB key into your computer
Copy "sdm-mod.app" into the "/siglent" of the USB key
Unplug the USB key

Step 5
Plug the USB key into the SDM3055 and go to telnet
Stop the current sdm.app application : use "ps" to list all application PID and kill the "./sdm.app" PID.
Generally it's the PID 550, you could kill it with command "kill 550"

Step 6
Prepare and run the mod
"mkdir /usr/bin/siglent/usr/mod"
"cd /usr/bin/siglent/usr/mod"
"cp -R /usr/bin/siglent/usr/mass_storage/U-disk0/siglent/ /usr/bin/siglent/usr/mod/"
"cd siglent"
"./sdm-mod.app"

Step 7
To stop sdm-mod.app, just press "Control C".
Restart the device to return to normal operation.

Step 8
To remove the mod folder
"rm -r /usr/bin/siglent/usr/mod"

Code: [Select]

sdm-mod.app 20190812 v01
SHA1 : 85ceab5b27570de364ab127d0c67a7deb17903ab
SHA256 : d26b360d6db9a89b058b4d44a2116c546b492e402df37fa1d491276a415fd5d0

Features
- Add 2 digits (7 1/2 digits)

Drawback
- On some menu the sign and/or the last digit is not display correctly


[alexvg]

I'll post later 2 other versions :
- sdm-mod.app with only one additionnal digit
- sdm-mod.app with two additionnal digits and some special text display (if it's work !!!)

I've not currently found the solution to change the font size.
It takes very long because my ARM debugger does not work...

If anyone know a solution to do a remote debugging of an ARMv7 GDBServer using Windows (or Debian)...

If you have a voltage reference, could you post your result ? (I use the BARmeter menu to display all digits)

[tv84]

If anyone know a solution to do a remote debugging of an ARMv7 GDBServer using Windows (or Debian)...

IDA?

[Zorc]

  Very cool you've got this far!
Interesting to find out how usefull this actually is!
I don't have a benchtop DMM but maybe in the future this might be a model I would consider buying (whenever I decide I really 'need' (read: want to have) one to supplement my handheld multimeters)

[coromonadalix]

Has it been compared against a precision source  ??,  nice to have more digits,  are they worth displaying ??

[alexvg]

Has it been compared against a precision source  ??,  nice to have more digits,  are they worth displaying ??

I would like to know if additionnal digits are usefull or not.
The only thing I could say, there are an internal rounding error.

IDA?

I've tested IDA but it does not support this application and/or GDBServer integrated into the SDM3055.

[tv84]

I've tested IDA but it does not support this application and/or GDBServer integrated into the SDM3055.

What about via JTAG?

[alexvg]

Here's the 6 1/2 digits version :

Code: [Select]

SDM 3055 m1-1908-01

Feature
- Add 1 digit (total of 6 1/2 digits or 2 400 000 points)

SHA1
cb89541b22959c8cdae8a7a917b29ca7042d4bcf

SHA256
73bef60988a4b4ec684132795c0f6b99bb4fcf4d4e1824f0acad414d9f9f64cb


How to test the sdm-mod ? (only with the latest 1.01.01.19 firmware)

Step 1
Plug a formatted FAT32 USB Key (I use 4GB) into the SDM3055

Step 2
Connect to telnet (username:root, password:ding123)

Step 3
Backup all Siglent data into the USB key :
"cp -R /usr/bin/siglent /usr/bin/siglent/usr/mass_storage/U-disk0/"
After copying, unplug the USB key

Step 4
Plug the USB key into your computer
Copy "sdm-mod.app" into the "/siglent" of the USB key
Unplug the USB key

Step 5
Plug the USB key into the SDM3055 and go to telnet
Stop the current sdm.app application : use "ps" to list all application PID and kill the "./sdm.app" PID.
Generally it's the PID 550, you could kill it with command "kill 550"

Step 6
Prepare and run the mod
"mkdir /usr/bin/siglent/usr/mod"
"cd /usr/bin/siglent/usr/mod"
"cp -R /usr/bin/siglent/usr/mass_storage/U-disk0/siglent/ /usr/bin/siglent/usr/mod/"
"cd siglent"
"./sdm.app"

Step 7
To stop sdm.app, just press "Control C".
Restart the device to return to normal operation.

Step 8
To remove the mod folder
"rm -r /usr/bin/siglent/usr/mod"

[alexvg]

If anyone with a 3045X wants to test the 3055 in his DMM, I can easily prepare an .ADS.

BUT, no responsibility for my side.

Never try to use SDM3045X or SDM3065X into the SDM3055
- The SDM3065X app program have an error loop
- The SDM3045X app program seems to work, but there no value on the display.

I'm currently testing a modification : I've removed the 0.5uV add/sub.

Edit: If you want to use the initial mod version I've posted (7 1/2 digits), You need to use the stat menu to display all digits and sign without any artefact.
I've noticed after a measurement of a battery (around 8V), using 20V range, there's a +/-20uV variation. This corresponding to a +/-2uV at the input of the ADC (with its 2.5V reference).
This seems to be very good.

[alexvg]

I've add 2 little heat sink on voltage reference and ADC and I removed the fan.
The result is interresting (using 200mV range, short-circuit on the input and more than 4000 samples) :
- Previous noise result was 1.75uV
- Now the noise is reduced to 1.34uV (difference between min/max) and standard deviation is 0.207uV

I've mesured a -90dB noise into the 5V line... Why Siglent integrated the analog/logic power supply on the same board ? (except for lowering cost).
I'm thinking about adding a pre-rectifier/filter/regulator, changing the 5V regulator and probably changing the voltage reference to the LTC6655B-2.5 (if I could access to an excellent reference voltage).

I've received a AD584LH voltage reference, I've measured 20K samples at 2.5V with 20V range :
- 78uV span (+/-39uV)
- 11,1uV standard deviation

I think the SDM3055 seems to be more interresting than I though first (It only need some modifications to be more better).

[alexvg]

I've tested IDA but it does not support this application and/or GDBServer integrated into the SDM3055.

What about via JTAG?

I've successfully running the remote debug using Texas Instrument Code Composer Studio...

[tv84]

I've successfully running the remote debug using Texas Instrument Code Composer Studio...

Must study that.   

[alexvg]

I've done a very long measurement to check the quantification level.
All measured values are using 22.5bits (slow mode or 4.7SPS). There's no loss during calculation.
22.5 bits corresponding to 6.77 digits. A full 24bits ADC (as seen in the SDM3065X) could provide 7.2 digits.
A long integration time with a sinc filter could provide more bits.

[alexvg]

I need your help, and I've added 2 questions on the first post :
https://eevblog.com/forum/testgear/hacking-the-siglent-sdm3055/msg2605347/#msg2605347


I've done multiple meaures with an AD584 at 2.5V with 7.5 digits.
- Statistic averaging of 500 samples give 2.496551V
- Averaging of 23 values in 5 seconds give 2.49655078V
- Sinc filter of these 23 values give 2.4955098V

measurements of a 10K 0.1% 30ppm resistor with 7.5 digits
- Statistic averaging of 500 samples give 10.005007k
- Averaging of 23 values in 5 secondes give 10.0050084k
- Sinc filter of these 23 values give 10.0050086k
- Statistic averaging of 4k samples give 10.005006k


The noise level during A/D conversion is very low in Resistance mode (much more lower than DCV).

[maginnovision]

Are you comparing what you get to something that is actually more accurate, or are you just assuming this is giving you relevant data(not noise)?

[alexvg]

Are you comparing what you get to something that is actually more accurate, or are you just assuming this is giving you relevant data(not noise)?

Actually I trying to identify all most important noise sources (ADC, Voltage ref, op-amp, relay...).
I measure the distribution of data to see if internal processing maintain all available bits (from ADC).

The SDM3055 PCB analog-part is the base of the 4.5 and 6.5 digits DMM and I think it could be go to 7.5 digits (but with another ADC and some addtitionnal important modifications).
I don't know if 7.5 digits could be really possible with less minor modification, but I'm currently working on these little hardware modifications to check that (with software modifications and/or new software).

[alexvg]

Would you agree to pay a small amount for a new SDM3055 software (and maybe for the SDM3045) ?
http://www.pixule.com/39863606081_would-you-agree-to-pay-small-amount-for-new-sdm3055-software-and-maybe-for-the-sdm3045.html

Would you like to improve your SDM3055 by adding a some options ? (The SDM3045 may be compatible with these options.)
http://www.pixule.com/398636486944_would-you-like-to-improve-your-sdm3055-by-adding-some-options.html

Thanks for your help !
Sorry for double-post.

[tv84]

That's a wrong move!

[maginnovision]

You start charging and you might be asking Siglent to challenge you. Since you're modifying their firmware it could be a legal issue.

[joeyjoejoe]

Asking for something to go wrong. Anyone needing more resolution would probably spend on the 3065 for supported 6.5 digits. Only hacker community would entertain this, and these are the folks who wouldn't pay for something like that.

[alexvg]

You start charging and you might be asking Siglent to challenge you. Since you're modifying their firmware it could be a legal issue.

That's not what I want.
My first tests was only done for evaluation only, I don't want to bother Siglent (I appreciate theirs products !).
I prefer providing my own firmware (not related to Siglent) but compatible with their product while offering additionnal, new or different features.
The SDM3055 is very good, I think it could be even better. My goal is to maximize this device.

Asking for something to go wrong. Anyone needing more resolution would probably spend on the 3065 for supported 6.5 digits. Only hacker community would entertain this, and these are the folks who wouldn't pay for something like that.

You could be right... In the Android community some people are ready to pay some OS updates and advanced features for their old device. The SDM3055 is an old device.


I really appreciate your help, yours advices and opinions. Thank you.

[maginnovision]

As far as I can see the 3055 is not old, it's a current product.

[blacksheeplogic]

You start charging and you might be asking Siglent to challenge you. Since you're modifying their firmware it could be a legal issue.

OK, just so I have this straight. It's OK to steal but no OK to charge for stolen items? I was obviously under the misunderstanding that both were wrong.

If your providing a patch to 'extend' functionality (your own work, not enabling someone else work) to a device not sure where the legal issue comes from (sold or given). If your providing FW that's been patched which includes the original FW that's quite different as you are giving/selling someone else work.

[maginnovision]

You start charging and you might be asking Siglent to challenge you. Since you're modifying their firmware it could be a legal issue.

OK, just so I have this straight. It's OK to steal but no OK to charge for stolen items? I was obviously under the misunderstanding that both were wrong.

If your providing a patch to 'extend' functionality (your own work, not enabling someone else work) to a device not sure where the legal issue comes from (sold or given). If your providing FW that's been patched which includes the original FW that's quite different as you are giving/selling someone else work.

Both are wrong, but if you're charging people are more likely to take notice. It's not open source software, modifying it could be illegal.

To add some context Keysight Daniel bogdanoff once said they don't care if you hack your scope but they do care if you try and sell hacked scopes.

[Mr. Scram]

Both are wrong, but if you're charging people are more likely to take notice. It's not open source software, modifying it could be illegal.

To add some context Keysight Daniel bogdanoff once said they don't care if you hack your scope but they do care if you try and sell hacked scopes.

He's saying he'd provide his own firmware not related to Siglent. Writing your own firmware and selling it is fully legal but not trivial to pull off.

[coromonadalix]

Is it me   or its going in the wrong direction ?   even if he provide his code,  where the code base came from ? Siglent   no/yes ?

[Mr. Scram]

Is it me   or its going in the wrong direction ?   even if he provide his code,  where the code base came from ? Siglent   no/yes ?

Apparently not. Writing your own code completely from scratch is legal. Patching existing firmware with your own code also tends to be legal as you don't copy or provide someone else's code.

[JohnG]

This is really nice work, and the related discussion interesting as well.

But, there is a practical matter. A modified meter will have no warranty and reduced odds of support. Have you checked the prices of used 34401 meters lately? Not as many modern features as the 3055A, but they use a LM399 reference and do 6.5 digits without mods. The reference has a good chance of being aged and stable.

I have both meters. The 3055A is easier to use, for sure, but not much of a track record regarding accuracy. A modified version, even less so.

Just my $0.02.

John

[alexvg]

Is it me   or its going in the wrong direction ?   even if he provide his code,  where the code base came from ? Siglent   no/yes ?

The SDM3055 is based on Linux operating system with only a binary file written (partially) by Siglent launched during startup. Used internal libraries and all external libraries come from Texas Instrument, Linux system...

As far as I can see the 3055 is not old, it's a current product.

It's an old device from 2014... Lots of units have been sold : maybe some people would be interested in updating or improving their device without buying a new more powerful and perhaps more expensive DMM.

[NoisyBoy]

Since you are trying to build something to sell, I doubt this is an appropriate discussion for this forum especially when you try to do a product interest survey. 

The 3055 is not designed as an open hardware platform for open source firmware development.  I will never pay for “enhanced” firmware that does not come from the manufacturer, it is a sure way to void the warranty plus I will never trust the reading. 

I am not saying it is a bad idea, just that it is more appropriate to be discussed in another venue.

[alexvg]

Thank you for this very interesting and informative discussion.

About software, I'm continuing working on the new firmware :
- I've successfully access to the display and keyboard.
- I'm waiting for the Lattice USB programmer.
- I'm currently investigate the protocol between digital and analog boards.

About hardware, I've received the components to build a pre-regulated power supply.
I'm waiting for the LTC6655BHLS8-2.5 voltage reference. I'll try one chip to measure ADC noise.
I'd like to build a 6-chip voltage reference : 6x LTC6655 in parallel to limit noise and deviation.
I'm waiting for a brand new calibrated voltage reference (based on the LM399AH).
I'm always searching for a calibrated LTZ1000ACH voltage reference module, I could not find it. Any ideas ?

[alexvg]

I've found some interresting things :
- Only low-23 of 24 bits are used in the software (the 24th bit is used as a polarity bit - shame)
- I've done some precision simulations using datasheet ADC and I presume the ADC used have an internal precision of 24bits (instead of probably 32bits ADC used in the SDM3065X). Using software filtering and fast ADC sampling rate, it's seems to be possible to have a some better results.
- I've found some better Op Amp at the same price (ADA4528-2 instead of AD8629 and LTC1151 instead of AD706)
- I've received a LM399AH-based voltage reference calibrated at 10.00000V (my SDM3055X displays up to 10.001V)
- Resistance measurement (2 or 4 wires) is more stable than current or voltage measurement.
- The internal +5V, +15V and -15V have some 100Hz (from 50Hz AC).

[Kleinstein]

It is kind of normal that 1 bit is used for the sign. So 23 bits + sign is kind of normal. Chances are the noise of the ADC is high enough, that there is very limited value to more resolution. With long time averaging a slightly high resolution can be possible, but this does not need the numbers from ADC chip itself.

The ADA4528-2 OP is not per se better than the AD8629 - they are different:  the ADA4528 is very low voltage noise, but higher bias and much higher current noise. The ADA4528 is good for signal sources with less than some 10 K impedance. However the AD8628 is much better for high impedance sources, like the input divider for the higher voltages.

In case one really want's to get sub 100 nV resolution, it would be more like an extra external preamplifier based on an OP like ADA4528. This could also add more gain and have a higher precision protection part (though possibly lower maximum voltage, more bias). Chances are in the lower cost meters the protection is just a resistor, maybe PTC. This could add enough resistance and noise so that it does not make sense to use the ADA4528.

Similar the LTC1151 and AD706 are good for different applications. So it depends on the use. AZ OPs can add quite some spikes, that might upset other circuit parts.

If resistance measurement is more stable than voltage and current measurements, this indicates that the reference noise is a mayor contribution. This might also be the external reference. A lower noise reference may be one of the few actually possible hardware mods, especially if the reference noise is really that bad. Usually the voltage readings near 0 V should be not effected by reference noise, but a reading near to the full scale (e.g. 2 V range, maybe 20 V range) should.

The chip SD ADCs are usually more like good for the lower end 6 digit meters, as the INL can be relatively high. The other problem is the limited range, so that there can be additional gain errors from resistors. So one may get more resolution (low enough noise), but the accuracy is still limited. These SD ADC chips are good for 5.5 digits, and they can even have a good point: the bandwidth for the input noise can be low, as they may not loose time to AZ phases.

[alexvg]

Kleinstein> Thanks for your very interresting post !!!

I've intercepted data from the FPGA to get the used protocol.
Now I could convert the raw value into voltage value using the DMM calibration XML.

I've detected the real range of the analog input (VDC mode) :
- 200mV range : -312.33mV to +312.55mV (without ADC overflow)
- 2V range : -2.4984V to +2.5110V (without ADC overflow)

In VAC mode (100Hz sine)
- 2V range : 0 to +5.0356V (without ADC overflow)

Measurements done using Rigol DG4102 (as a DC voltage generator) and UNI-T181A (as a voltage measurement)

The VDC mode uses 23bits and 1bit for the polarity (sign)
The VAC mode uses 24bits but within a range from 2^23 to 2^24-1 (like 23bits)
The resistance mode (2W and 4W) uses the same limitation as VAC mode (23bits)

I'm also working on 2 mods in parallel : power-supply low-noise preregulation (plug and play) and multiple LTC6655BHLS8-2.5 chip for the voltage reference (need soldering).

[alexvg]

It was hard to do but now I could access to ADC raw data.

Done :
- Read keyboard state
- Access to display
- Read ADC raw data

To do :
- Control capture mode
- Speaker
- External trigger
- VMC Output

[alexvg]

I've done the investigation of the 20V VDC range... It's very funny :
-31.25V to +31.25V (without calibration)

[Nornand]

Gainsetting seems to be 8 for the PGA.
Vref/8=0.3125V

[alexvg]

Yes the PGA is used... Shame.
The DMM is not an internal 240.000 points because it's possible to have up to 312500 points.

I've done a continuous decimation with a sinc function over 4096 samples captured with speed=FALSE. The result is similar to the speed=SLOW with 30sec statistics.
I confirm the fact that it's possible to have a super fast (with a low resolution) bar meter and a high precision (slow) display number simultaneously.

I'll try to enable a more faster tranfert between ADC, FPGA and ARM (4800SPS). I'll need to check if the current onboard Linux is compatible with critical realtime processing (or change to another Linux).

I've found the LT6657 voltage reference... Better than the LT6655...
Probably a stupid idea (or not) : If I'll build a LTZ1000AH voltage reference to supply multiple LT6657A-2.5. Does the final output voltage have a lower drift than a standard low-noise 5V voltage regulator powering multiple LT6657A-2.5 ?

[Kleinstein]

A standard 5 V regulator (if not getting too much heat and variable current) is good enough to supply a LT6655 or similar reference. The next step up would be a bootstrapped supply - though it should not be needed here.

In many aspects the LT6657 is not better than LT6655 (especially the LS8 version). Both are lower noise than the max6325, but the TC may be higher.
The big weakness of all these reference is not so much the noise, but the drift, hysteresis and sensitivity to humidity. So the difference between the MSOP and LS8 case may be significant.  Anyway for a 5.5 digit meter they should be good enough.
As the ADC needs the 2.5 V reference level, there are not that many better references to choose from. Several in parallel help with noise, but only little with long term drift rate.

The other limitation of the DMM design is that it is only +- 2.5 V (may be a little more, but not much) with high impedance.
So there are limitations to the design, that can not be circumvented.

[alexvg]

The LTC6655 (LTC6655BHLS8-2.5) and LTC6657 (LT6657AHMS8-2.5) have a better long term drift than MAX6425 after more than 1000Hr and the noise is lower : the ADC used does not like the noise into its reference.
It's possible to use a LTZ1000A into the SDM3055 with a programmable voltage divisor (based on Vishay high precision low drift resistances). It's simple to do.
I study all possibility : After some little modifications on the device, I would like to know what level of precision it is possible to reach.

[alexvg]

A standard 5 V regulator (if not getting too much heat and variable current) is good enough to supply a LT6655 or similar reference. The next step up would be a bootstrapped supply - though it should not be needed here.

In many aspects the LT6657 is not better than LT6655 (especially the LS8 version). Both are lower noise than the max6325, but the TC may be higher.
The big weakness of all these reference is not so much the noise, but the drift, hysteresis and sensitivity to humidity. So the difference between the MSOP and LS8 case may be significant.  Anyway for a 5.5 digit meter they should be good enough.
As the ADC needs the 2.5 V reference level, there are not that many better references to choose from. Several in parallel help with noise, but only little with long term drift rate.

The other limitation of the DMM design is that it is only +- 2.5 V (may be a little more, but not much) with high impedance.
So there are limitations to the design, that can not be circumvented.

You're right ! Using the LT6657 will provide lower performance than the LT6655 (LT6655BHLS) with the AD7190.
So what to do ?
1. I could try to use up to 6x LT6655 ? (with a thermal isolation)
2. I could try to use up to 4x LM399A ? (but with a voltage conversion adding drift and low pass filter)
3. I could try to use one LTZ1000A ? (like LM399 with a similar voltage conversion adding drift and low pass filter)

Don't know what to do...

[Kleinstein]

The rest of the DMM (like gain stage and input divider - already used for the 20 V range) are still more suitable for 5.5 digits. So there is limited use in replacing just a a few points and hope to get much better performance.
It starts with having high impedance up to some 2.5 V only.
At some point is would be more like building your own DMM from scratch.

So I would consider anything more like a single LT6655 (with maybe some simple thermal regulation) overkill. 
The main advantage of an LM399 would be better long term stability - still needs a very stable divider stage. It would also mainly help the 2 V range. The divider (+ buffer)  from a 7 V level down to a 2.5 V (maybe 3.5 V ?) level is probably more tricky than the LTC6655.
One also has to keep in mind that the power supply and thermal design may not allow much more power.

A change in the HW would mainly make sense if there is a simple, easy upgrade possible because of something like a design fault (could be a cost driven decision)  causing an isolated weak spot. Not sure if there is one such weak point -  chances are multiple source of similar size add up.

So a first point would be an in depth analysis of the performance of the original hardware. So how much of the noise is due to the input amplifier, possibly the input buffer, the ADC and the reference ?
For the drift this is more tricky as the result is valid for the one unit only.

[alexvg]

The rest of the DMM (like gain stage and input divider - already used for the 20 V range) are still more suitable for 5.5 digits. So there is limited use in replacing just a a few points and hope to get much better performance.
It starts with having high impedance up to some 2.5 V only.
At some point is would be more like building your own DMM from scratch.

So I would consider anything more like a single LT6655 (with maybe some simple thermal regulation) overkill. 
The main advantage of an LM399 would be better long term stability - still needs a very stable divider stage. It would also mainly help the 2 V range. The divider (+ buffer)  from a 7 V level down to a 2.5 V (maybe 3.5 V ?) level is probably more tricky than the LTC6655.
One also has to keep in mind that the power supply and thermal design may not allow much more power.

A change in the HW would mainly make sense if there is a simple, easy upgrade possible because of something like a design fault (could be a cost driven decision)  causing an isolated weak spot. Not sure if there is one such weak point -  chances are multiple source of similar size add up.

So a first point would be an in depth analysis of the performance of the original hardware. So how much of the noise is due to the input amplifier, possibly the input buffer, the ADC and the reference ?
For the drift this is more tricky as the result is valid for the one unit only.

You're right, it's possible to use a 5V voltage reference but I need to check if this voltage is compatible with the analog input (AD7190 is compatible).
It's also possible to change the ADC... But I don't want to do this.
I'll need to check the noise from the input to the ADC (I'll need to build a LF amplifier probe).

I've also successfully set to 4800Hz the ADC and done a sinc filter to enhance the final resolution, the result is better than the original "Speed - Slow" (or 5 SPS)

[Kleinstein]

I don't think a 5 V reference would be practica, as it's too far of the original 2.5 Vl. However a 3 V or 3.5 V (7 V / 2) ref. level could be working.

For checking the noise source, much can be done with the meter itself, reading data and look at the noise under different conditions. With s short, there should be no reference noise (at least not the normal one). With high gain in front chances are the amplifier noise would be dominant.
With the SD ADC there is however the ADC internal "gain"  (Usually actually only more frequent input sampling), that can make the analysis more complicated.

Is there an analog gain stage, or are they only using the ADC internal PGA stage for the 0.2 V range ?

[alexvg]

It's possible to use a +5V voltage reference with the AD7190, but with the design of the SDM3055, I don't know...

Using VDC mode :
- 200mV - PGA = x8
- 2V - PGA = x1
- 20V - PGA = x8
- 200V - PGA = x1
- 1000V - PGA = x1

When I finish the logic analysis, I think I'll do an in-deep inspection of the analog part, and do lots of measurements.

At the moment I'm having trouble understanding the measurement of resistance of very large values (10M to 120M).

[Kleinstein]

For very large resistors a common method is to add some resistance like 10 M in parallel. This allow to measure very large resistors, though with reduced accuracy and a nonlinear function to convert from the raw reading to resistance.
As the SDM3055 is more like a budget instrument, I would expect this method.

[alexvg]

For very large resistors a common method is to add some resistance like 10 M in parallel. This allow to measure very large resistors, though with reduced accuracy and a nonlinear function to convert from the raw reading to resistance.
As the SDM3055 is more like a budget instrument, I would expect this method.

Yes, I've tested multiple methods without the correct results (I need to find the exact value displayed by the original DMM software). I continue my investigations.

[Nornand]

Datasheet says
Range:  100 MΩ              200 nA || 10 MΩ

[alexvg]

Datasheet says
Range:  100 MΩ              200 nA || 10 MΩ

Yes it's a current generator.
But I could not find the exact result.

Edit: I think I've found it... I need to do multiple checks.

[alexvg]

I've finally found 2-wire and 4-wire measurement method.
I'm working to found how calibration work...

[alexvg]

I'ts ok !!!
I've successfully found the method used for 2-wire and 4-wire resistance measurement.

I've discovered 2 problems with 2-wire and 4-wire 100M measurements :
- I've measured 144mV p-p 50Hz at the input of the ADC (with open input but with LO/Losense shorted)
- The XML calibration is not fully used : 0-50M segment is used for 0-120M and 50-120M is not used at all !!!

A interresting information :
- I've success fully generate a 0V ADC input with 0x800000 output data (with +/-1LSB precision during more than a minute- 7FFFFF to 800001)

I've done my measurements with :
- 10K to 130M with 0,01% 0,1% or 1% resistance value (for 2-wire and 4-wire)
- I use a Rigol DG4102 with filtered output and 20dB or 40dB attenuation for simulate resistance (only 4-wire)

[Kleinstein]

It is not a surprise to find some hum for the high resistance values. Normally the high Ohms ranges should use relatively high voltage ranges. Here this seem to be the 2 V range. It may take some shielding to get good readings. Ideally the DMM would detect if there is too much AC background (e.g. to cause clipping in peaks).

144 mV_pp is not too bad, less than some 5% for the full range.  For a good measurement one may need to use a shield to keep hum away and maybe avoid using long unshielded cables.

For the 10 M range 4 wire Ohms measurement does not make much sense.

[alexvg]

It is not a surprise to find some hum for the high resistance values. Normally the high Ohms ranges should use relatively high voltage ranges. Here this seem to be the 2 V range. It may take some shielding to get good readings.

The voltage range could go up to 2V, it's current generator of (around) 200nA

For the 10 M range 4 wire Ohms measurement does not make much sense.

I think you're right :
- Right because it's a nonsense to mesure high resistance value.

But I need to add something I'm thinking :
- Using a 4-wire cables it's possible to use shielded cables and differential input (for sense wires) for a more precise measurement.
But I could be wrong with this, I'm not sure (never tested, never simulated...)


I've also notice a strange math simplification inside the original software :

You could write :
f1(x) = 1/(1/x)
f2(x) = x
It's okay to say f1(x) = f2(x)
But it's false for a CPU !

Original Siglent Software uses a function like f1(x) that could cause a "divide by zero". But without simplifying the function, it's possible to avoid the "divide by zero" like f2(x). I don't understand why Siglent have done this...

[Kleinstein]

The capacitive coupled mains hum during the high ohms measurement is present at the DUT and the hum is thus directly added to the voltage to measure.  So a differential measurement does not help here. The main things that helps is the usually good suppression of mains hum by the ADC, because it is integrating over a multiple of mains periods.
Shielding for the high ohms measurement can be tricky, as there may be more leakage. Some high end DMMs offer a special driven shield for ohms shielding, that reduced the voltage over the isolation and also help a little with settling.

I sometimes makes sense to measure high resistors, but in the MOhms range one does not really care about cable resistance and could thus use a 2 wire measurement.  Some meters only offer 4 wire Ohms for the lower ranges.

Using only 2 wire ohms reduces the cables to 2 and thus less leakage and less surface to pick up . So there is a point at high resistors when 2 wire ohms is more accurate than 4 wire ohms. It is not just the external leakage, but also leakage at the input protection.

[alexvg]

Measuring using 100M range with this device is a joke (using 4-w and 2-w).
The precision is awful (even with 50 and 60Hz filter)

[Kleinstein]

The accuracy gets worse with high resistors already due to leakage and bias currents (e.g. at the protection). With only a 2 V range (compared to often 5-10 V for other 6 digit DMMs) this also makes things more tricky. 4 Wire resistance is likely worse than 2 wire. The 100 M range is about 10 times less accurate than the 10 M range as there is the 10 M in parallel and thus only a relatively small change used.  Anyway getting 1% or so accuracy is not that bad with such high resistors.

A really accurate measurement would need a special instrument, more like an electrometer. This usually is with applying a fixed (e.g. 20 V or 100 V) voltage and than measure the current. This also need shielding and special, short cables. Also touching the resistors with the hand can change things from contamination.

[alexvg]

My Applent AT2816 is more precise for high value... But it's a specific device.

I ve done 2 simple tests (with 2-wire and 4-wire)
- a 110M resistance measurement with 4-wire display 111,7M to 112,5M over a minute
- a 110M resistance measurement with 2-wire display 111,84M to 111,87M over a minute



A funny thing I see each time...

Code: [Select]

===============================================
|SIGLENT SDG800 project        /dev/pts/0
===============================================
SIGLENT Project http://         (none)
(none) login: root
Password:

Processing /etc/profile... Done(SDG800 app and data are present inside the SDM3055 firmware)

[alexvg]

I'm currently working on the frequency counter... It works up to 2,2MHz... Very limited ! My UNI-T181A (half the price) works up to 60MHz.

Now I could use the frequency counter.

I've measured a maximum frequency of :
- 1Vpp : 3,1MHz square and 2,9MHz sine (Range 200mV)
- 5Vpp : 4,1MHz square and 3,9MHz sine (Range 200mV)
- 10Vpp : 5,0MHz square and 3,7MHz sine (Range 2V)
- 20Vpp : 5,0MHz square and 5,0MHz sine (Range 2V)

[alexvg]

Now I've got the Capacitance measurement process.
It's possible to display a value with 5, 6, 7 or 8 digits... But the accuracy is very low...

To check capacitors, I use a LCR Applent AT2816 at 100Hz / 2V


I've done 34 measures with a list of 18 capacitors from 100pF to 10000uF.
All capacitors have been checked before and after the measure of the SDM3055 with the LCR.

Minimal error : 0,04%
Maximal error : 10,2%
Average error : 2,2%
Std Deviation : 0,0228%

The capacitance measurement of the SDM3055 is not very useful than a real LCR.
I've also checked all capacitors using a low-cost LCR (Peaktech 2170). The results are always better than the SDM3055 !

[alexvg]

I've selected some components to build a pre-regulated power supply.
- A couple of LM2991/LM2941 is used to create the +/-18V (regulated later for +/-15V)
- A single BA80BC0WT is used to to create the +8V (regulated later for +5V)
- A 100V/2A/0.7V Schottky bridge rectifier (+/-15V line)
- A 40V/2A/0.5V Schottky bridge rectifier (+5V live)

The main goal is to decrease the noise of these power lines and to reduce the heat of the main regulators.
The selected LDO regulators have a switch pin which will be connected to the internal switch signal.

I also need an additionnal power supply because I'll use a LTZ1000ACH instead of the MAX6325 for the ADC voltage reference.

[CDaniel]

Now I've got the Capacitance measurement process.
It's possible to display a value with 5, 6, 7 or 8 digits... But the accuracy is very low...

To check capacitors, I use a LCR Applent AT2816 at 100Hz / 2V


I've done 34 measures with a list of 18 capacitors from 100pF to 10000uF.
All capacitors have been checked before and after the measure of the SDM3055 with the LCR.

Minimal error : 0,04%
Maximal error : 10,2%
Average error : 2,2%
Std Deviation : 0,0228%

The capacitance measurement of the SDM3055 is not very useful than a real LCR.
I've also checked all capacitors using a low-cost LCR (Peaktech 2170). The results are always better than the SDM3055 !

If the big error is for high value electrolytic caps it is fairly normal for a multimeter ... even two different Fluke don't display the same value and could be easily 10% off . For the simple method  of charging/discharging the cap the time is dependent on losses and current used ...

[alexvg]

If the big error is for high value electrolytic caps it is fairly normal for a multimeter ... even two different Fluke don't display the same value and could be easily 10% off . For the simple method  of charging/discharging the cap the time is dependent on losses and current used ...

The error is spread over all the measures (but above the Siglent specs).
The current used is specified into the SDM3055 datasheet (200nA to 1mA).
I don't understand why Siglent doesn't incorporate a little frequency generator to do a more precise and FAST capacitor (and inductance) measurement using the 2 or 4 wires

[alexvg]

I've done DCI and ACI. It was very funny because DCI 200uA range have an internal inverted polarity (-11uA measure and +11uA display).

[alexvg]

Here is a list of reverse engineering features done :
- DCV (all range, filter mode and 10M/10G for 200mV and 2V)
- DCI (all range and filter)
- ACV (all range)
- ACI (all range)
- Resistance 2-wire (all range including 100M !!!)
- Resistance 4-wire (all range including 100M !!!)
- Frequency (all ACV range)
- Capacitance (all range)

Remaining functions :
- Diode
- Continuity
- Temperature

Edit: All functions done !

[tv84]

Edit: All functions done !

  Where there's a will there's a way!

[Kleinstein]

I've done DCI and ACI. It was very funny because DCI 200uA range have an internal inverted polarity (-11uA measure and +11uA display).

An inverted polarity could happen if they use a trans-impedance amplifier for the low range. In this case I would expect to have at least 2 such ranges, possibly the option to add a 20 µA range.

[alexvg]

It's possible to have more range, but I need to do a noise measurement (FFT analysis of the ADC running at 5KHz).
It's also possible to easily replace the used ADC with a better one (like the AD7175-2 used in SDM3065X with some additionnal electronic).
The SDM3055 board is more interresting than the SDM3045 and SDM3065, modifications are easy but available space inside is very limited.

[alexvg]

This is just a first test for the new interface

[tautech]

Puke green ?   

Otherwise nice improvements. 

[e0ne199]

dang you should use your ability to hack 121GW instead of this DMM...

[alexvg]

I'm currently writing the software using a Windows emulation to have an easier development.

I'm thinking about 2 hardware mods :
1. pre-regulated power supply (I already have components but haven't got time to assemble them)
2. remove ADC and ref and add an additionnal board with a better reference (LTZ1000A) and a better converter (AD7175-2 or AD7177-2)
But i don't know if the pcb and current components provide a noise floor enough lower to enable a good enhancement.

I haven't got the tools to measure the 0.1-10Hz noise at very low level.

[Kleinstein]

Measuring the noise if the DMM is not that difficult. The main tool it the DMM itself.  The first test is measuring the noise with a short at the input and than calculate the noise from the data. The second test needs a low noise external reference to measure a voltage at a significant part of the full scale. This gives the second noise case that includes the noise of the DMM internal reference and maybe some extra ADC parts.

The 0.1 to 10 Hz noise would be approximated by getting data with 20 SPS and calculate the RMS value over some 10 seconds. Chances are one could use a slightly different sampling rate (e.g. 25 SPS) and thus a slightly different frequency range throughout the tests.

For the reference a 7 V reference needs an extra stage to divide down the reference to the ADC level (e.g. 2.5 V maybe 3.5 V). This can add quite some uncertainty / drift. So it is questionable if it is worth using an highly stable LTZ1000 with a divider that adds quite some drift. The 20 V range would still use the front end divider that adds quite some drift and also some noise. For a 2.5 V reference there may be a simpler replacement with another 2.5 V ref chip.

For the ADC, lower noise ADCs may need a lower impedance input signal. As far as I understood it, they use the internal gain for the low ranges (200 mV, current). So the new ADC also needs this option.  So the change is not that simple.  One may have to build a small separate PCB for the Reference and ADC - so first step towards building your own DMM.
The input amplifier (buffer) may be limiting the noise performance. I would expect some AZ OP as a buffer -  a compromise between noise and input bias. At least an AZ buffer should be sufficiently linear.

[alexvg]

Some news :
I'm working on an interface using a HTML/CSS rendering for a more scalable and flexible possibilities/features.
It takes a very long time because the power of the main CPU is very limited: I need to do a high level of optimizations and remove some CSS features (who need too much process time/power).
I don't know when the first alpha/beta version will be release, "probably" in 2-3 months.

Hope you'll enjoy my work.

[supperman]

Quick Questions.. As I'm considering a Siglent DMM:

Why don't they integrate current flow over time to get Amp/H. Seems super easy.. to get them to measure total charge on a battery for instance. (all the data is there). Seems that I would need to log all the readings and then run it in Excel on a PC? Does the Siglent log both Voltage AND current over time? (or only one or the other using the PC software).

@alexvg any thoughts on expanding the math to do integration math? Also, given the time you are spending on this.. why not just design a DMM? Siglent will never let you sell software for their device..

[BillB]

Quick Questions.. As I'm considering a Siglent DMM:

Why don't they integrate current flow over time to get Amp/H. Seems super easy.. to get them to measure total charge on a battery for instance. (all the data is there). Seems that I would need to log all the readings and then run it in Excel on a PC? Does the Siglent log both Voltage AND current over time? (or only one or the other using the PC software).

Good question regarding integrating current.  I don't believe the EasyDMM software supports dual measurement mode at all.  For meters without a channel card option, it only performs a single measurement function.

[alexvg]

Hi, sorry for the delay...
This Siglent DMM only do a single a measurement at a time.
Current/Voltage uses time slicing with a relay.

I'm working on a solution to provide multiple length and simultaneous math integrations.

[klausES]

Hi there,

I think I read somewhere that the temperature (at which the factory was calibrated) was 23.5 degrees Celsius?!?
(I'm not sure anymore, can't find the text anymore).

If I remember correctly, it would be the temperature at which the voltage reference and the relevant part of the board later
most closely correspond to the factory calibration in measurement mode.

Are there measurements or empirical values by how much these values in practice, e.g. 30 or 35 degrees C device (inside) temperature could differ
(to what extent the temperature compensation corrects this or not)?

The question has a background (a thought), but I'd rather wait and see if someone answers.

[Kleinstein]

How much the readings can change with temperature is sometimes noted in the instructions. Usually some +-5 K is included in the base accuracy  for more I would expect something on the order of 2-15 ppm/K as an upper limit. The lower ranges (200 mV and 2 V) that don't use the input divider can be better, as the reference used and the ADC have a relatively low TC. The input divider for the higher ranges (20 V, 200 V,...) is expected to have lower accuracy, due to the TC of the divider that is more in the 5-15 ppm/K range.

I would not expect much of an extra correction in software, as this would need an extra step at the calibration.

[klausES]

Thank you for your answer.

Had the following considerations.

It is clear that if constant values ​​are to be compared, it would be advantageous to keep the instrument at the same temperature at all times.
If it is almost all about constant (much more than the best absolute values),
it would not be advantageous to maintain a constant but higher temperature than the constantly fluctuating ambient temperature.

A precisely controlled temperature that is always x degrees higher than the ambient temperature
is much easier to implement inside the device than a possible cooling to keep it the same.
In winter my room has e.g. 18 degrees, but also 30 degrees in midsummer (or sometimes just> 30 degrees, unfortunately no climate).

If the device were now "always from the start" kept at, for example, exactly 30 degrees
(controlled by a precisely temperature-controlled heating and in a control loop proportional fan speeds),
fluctuating room temperatures in the range of 18 to approx. ... 29 degrees would hardly have any changing effects on the temperature in the device.

hence the question:
How much would the absolute values ​​at a constant 30 degrees differ from those of the factory calibration at 23.5?

Edit:
corrected a text error that led to confusion (see above). 

[Kleinstein]

The TC specs for the ADC and reference are at some 1 ppm/K. The actual performance may be a little better than this at moderate / common temperature. So the 6.5 K higher temperature my change the reading by some 5-10 ppm, so hardly detectable with a 5 digit meter.
The ranges with input divider and current ranges are likely less accurate so more like 10 times the change.

For the 5 digit resolution there is not much need for a constant temperature. This would be something for a 7 digit meter. At that level a oven for the reference (e.g. LM399 or LTZ1000) is the normal solution.

[iMo]

hence the question:
How much would the absolute values ​​at a constant 30 degrees differ from those of the factory calibration at 23.5?

In case you have a temperature sensor mounted in the box, you may simply compensate the readings based on that internal temperature. I did it with my meter at 10V range (34401A, I had to mount the sensor).
The absolute accuracy of the temperature sensor is not important here. A resolution of 0.1C would be enough for your meter, imho.
Then you will calibrate against that internal temperature. The DMM starts cold, and you will be logging the external ref voltage readings against the internal temperature - as the internal temperature of the DMM rises it will sweep up the temperature for you. You will get a "Vref" voltage vs. internal temperature dependency.
And the rest is math like y=ax+b (or whatever your dependency is like) within the internal temperature range of interest.
The internal temperature of your meter (after it stabilizes, it may take hours) is always Tdmm = Tx + Tambient, where Tx is a constant (an assumption here is your meter is not internally thermostated).

[klausES]

Thanks for the explanation.

I will take a closer look at the behavior and test the suggestions when the device arrives (hope tomorrow).

At first I only came up with the idea because the fan will get a temperature-controlled control right from the start (possibly also a quieter fan type).
Then a regulated "very specific selectable and then constant" higher temperature than the fluctuating environment would have offered itself.

[Kleinstein]

A temperature regulated fan is a two sided thing. With changing air speed temperature gradients can change - this can have more effect than the overall temperature change.  It really depends on the circuit if a temperature controlled fan helps of makes things worse.

[klausES]

Thanks for the information. I can understand your objections regarding fluctuations.
I'll pay attention to that.

In any case, a linear series regulator is planned here (neither pwm on the power side,
nor something clocked in the measurement and control).

There will also be no "fluctuations" by simply switching on / off,
Instead, an adjustable minimum speed (always active as a base load), an adjustable incline, a delta t> 1 kelvin
and a limitation of the maximum speed (moderate, far below the possible full speed of the fan).

[killingtime]

Really like the new interface. It's an improvement on the standard.

Putting the measurement units after the measured value (e.g. mV) in larger typface like Keithly have done makes it easier to read.

Making the measured value digits larger would be an idea - if there is space on the display. Means you can read the meter from further away.

Well done.

[alexvg]

Hi,

I'm currently working hard on the HTML/CSS rendering engine.
I don't know when I could release a test version.


It's a long way...

Edit: You could try these mods :
6.5 digits : https://www.eevblog.com/forum/testgear/hacking-the-siglent-sdm3055/msg2614659/#msg2614659
7.5 digits : https://www.eevblog.com/forum/testgear/hacking-the-siglent-sdm3055/msg2612406/#msg2612406

[stefkpl]

I will try this week. It's seem to be a good work
Tkx

[klausES]

What options are there to explicitly save the existing calibration file
(the still unchanged original factory calibration) of an SDM3055 ?

[alexvg]

Hi,

Since the start of the health crisis in France and around the world, I have put this project on standby.
I currently have to manage 7/7 server clusters (from several websites) so that children can work during confinement.
I will resume the development of this project at the end of confinement or when I have free time again.

Take care of yourself and your loved ones.


Alexia from France.

[Electro Fan]

Nice Work alexvg!  Hope all goes well and that we see you back on this project.

tautech, maybe you could encourage Siglent to hire alexvg or retain alexvg's services.  Sure looks like a lot of creative and capable product management and product development horsepower that could benefit Siglent, their products, and their customers.

[pipe2null]

"Me Too":  Yes, great work alexvg!  I hope things calm down in the world soon so we can all get back to kewl projects like this one.

7.5 digits : https://www.eevblog.com/forum/testgear/hacking-the-siglent-sdm3055/msg2612406/#msg2612406

I have an SDM3065X-SC and want to try out UI and 7.5 digit FW mods if/when it's possible to do so.  I considered attempting the 7.5 digit hack per the instructions, but since I have a different model, I thought I'd ask first to avoid messing things up.  Is it reasonably "safe" to try out the 7.5 digit mod on a 3065?

[rwgast_lowlevellogicdesin]

I have an sdm3045x and am pretty satisfied with it for the most part but, the resistance measurement 2 and 4 wire only measure to 3 digits beyond the decimal in its lowest 600 range, how hard would it be to hack a fourth digit? I'd be willing to upgrade my voltage reference to the max one used in the 3055 if need be!

[Muessigb]

Some news :
I'm working on an interface using a HTML/CSS rendering for a more scalable and flexible possibilities/features.
It takes a very long time because the power of the main CPU is very limited: I need to do a high level of optimizations and remove some CSS features (who need too much process time/power).

Why put a full HTML/CSS rendering engine on an embedded device. That's really inefficient. I don't like when people do it on Desktop, but it's just awful on embedded.
You can always use SDL or QT embedded to do the GUI.

I've done my own fair share of aftermarket firmware and SDL usually works quite well.

[Elasia]

"Me Too":  Yes, great work alexvg!  I hope things calm down in the world soon so we can all get back to kewl projects like this one.

7.5 digits : https://www.eevblog.com/forum/testgear/hacking-the-siglent-sdm3055/msg2612406/#msg2612406

I have an SDM3065X-SC and want to try out UI and 7.5 digit FW mods if/when it's possible to do so.  I considered attempting the 7.5 digit hack per the instructions, but since I have a different model, I thought I'd ask first to avoid messing things up.  Is it reasonably "safe" to try out the 7.5 digit mod on a 3065?

It's a different main board, i wouldnt normally say this but you got the top of the line model, i'd leave it be since its the flagship less you are out of warranty or the like

You can cross application files across same mainboard revisions safely, beyond that you are on your own, its the reason you can turn the new SSA 3000X Plus into a 3.2Ghz SVA model, same main board, just a different application file

[pipe2null]

True on the hardware.  Beyond that it depends on how they architected the firmware/software if the UI code is well separated from the underneath bits.  Most of the firmware hacks I've seen posted are reasonably safe and reversible to try but only for a very specific model.  After I posted I finally saw a note from OP of the "thou shalt not" try on other models...     Was hoping it might be possible to safely kill normal UI process and try out OP's much better UI.  Odds of that actually working are low, but thought I'd ask.  I'm not up to speed enough to hack around with it myself yet, need to get through a few more learning curves with JTAG and the like.

(*Sarcastic*) And thanks a lot for bringing up the SSA3000X+ "upgrade".  It took months to convince myself to wait until next year before digging into the RF side of things and getting related "field upgradable" SA/VNA toys...  -ER-, I mean essential tools.   
...  SVA?  Sounds like I'm behind on reading.  But I really, really shouldn't catch up.  Bad idea.  Bad, bad idea.  I don't go to TEA meetings.  I don't have a problem.

[alexvg]

Hello,
Just a little message to thank you for your support.
I have been sick with Covid-19 since June. Despite all the precautions taken (mask, distance, hand washing ...), that was not enough.
It's very difficult everyday with a lot of symptoms. The healthcare team is great but clueless.
Please protect each other, take care of yourself.

[Pinkus]

  We all wish you a speedy recovery. Take care of yourself and get well!

[tv84]

Hello,
Just a little message to thank you for your support.
I have been sick with Covid-19 since June. Despite all the precautions taken (mask, distance, hand washing ...), that was not enough.
It's very difficult everyday with a lot of symptoms. The healthcare team is great but clueless.
Please protect each other, take care of yourself.

I wish you well!

[ZhuraYuk]

Hello,
Just a little message to thank you for your support.
I have been sick with Covid-19 since June. Despite all the precautions taken (mask, distance, hand washing ...), that was not enough.
It's very difficult everyday with a lot of symptoms. The healthcare team is great but clueless.
Please protect each other, take care of yourself.

I wish you a strong health and good luck.
I want to thank you for this awesome work. 3055 is very capable device. Running 6-1/2 digit mod, I checked 4 wire resistance against 10OOhm 0,002% resistance standard. Only last digit was noisy same as my freshly calibrated Keithley 2100.  Also I compensated wire resistance because it was a little 0,0136 Ohm off when 4 wires shorted. After connecting to reference resistor the reading was stable up to last digit. All measurements were dine after 2 hours warm up.

[uski]

I have been sick with Covid-19 since June. Despite all the precautions taken (mask, distance, hand washing ...), that was not enough.
It's very difficult everyday with a lot of symptoms. The healthcare team is great but clueless.
Please protect each other, take care of yourself.

Anyone has news of Alexia ? I just found this thread and the last message is months old, so I am worried now !

[alexvg]

Hi,

I was hospitalized because of covid, a "long covid".
I went back to work a few weeks ago, it's really difficult.
I hang on, it's difficult. Little by little I know it will get better, but it takes time, lot of time.
I recently received a Keysight 34470A (the new black version).
I hope to be able to resume development quickly and provide at least a "voltmeter" version to test the capabilities.

Take care of yourself and your loved ones.
Alexia Gossa.

[alexvg]

Hi,
I'm working on the software. Here is 2 pictures of the functionnal DCV/DCI/ACV/ACI/R2W/R4W version in pre-pre-pre-pre-alpha...

Hope you enjoy.



" alt="" class="bbc_img" />
" alt="" class="bbc_img" />

[coromonadalix]

Ditch the leading zeroes    i hate this kind of display ... confusing  in my case, i hated that in my 34401a meters,    but the different color help to differenciate things 

[alexvg]

Hi, I've uploaded a short video from my UI test, real measurement and fake bargraph.

https://youtu.be/x4KNsvO_Dfs

[alexvg]

Hi,

I've uploaded a new video.
https://youtu.be/JyPmyu7O6NU
Powerline is set to 60Hz for this test.
Measurement of a 3Hz 20Vpp source signal at 1PLC without PLSync

Hope you enjoy.

Code: [Select]

Features done :
- Function selection : DCV, DCI, ACV, ACI, R2W, R4W, Diode and Cont
- Bargraph with min/max
- Stats
- High speed CONT-test at 250Hz with Beeper
- Power line frequency detection
- PLC selection from 1PLC to 100PLC (50Hz) or 120PLC (60Hz)
- Range selection


[coromonadalix]

Damn  It's on steroids  loll    Siglent should hire you    good work

Since there is a Siglent representative here,  he should ask for a collaboration with you ... and you could propose a new firmware update for theses meters

[purpose]

Hope you enjoy.

This did indeed bring me great joy. 

[alexvg]

Hi,


I'm working on the ADC signal data analysis to implement a more precise A/D conversion instead of the internal decimation of the AD7190.

I have gigabytes of data to analyze, but my actual server based on a Xeon E5-2640 is too slow to do efficient processing while working on the program.
I've ordered a new (old) processor, a Xeon E5-2698v4.

I hope this processor could provide me an efficient data analysis while allowing me to simultaneously working on the program source code.


I've also ordered a PDVS2mini from Ian Scott Johnston to check the new software.
https://www.ianjohnston.com/index.php/onlineshop/handheld-precision-digital-voltage-source-2-mini-detail


Have a nice day and take care of yourself.

[kcbrown]

Hi,


I'm working on the ADC signal data analysis to implement a more precise A/D conversion instead of the internal decimation of the AD7190.

I have gigabytes of data to analyze, but my actual server based on a Xeon E5-2640 is too slow to do efficient processing while working on the program.
I've ordered a new (old) processor, a Xeon E5-2698v4.

I hope this processor could provide me an efficient data analysis while allowing me to simultaneously working on the program source code.


I've also ordered a PDVS2mini from Ian Scott Johnston to check the new software.
https://www.ianjohnston.com/index.php/onlineshop/handheld-precision-digital-voltage-source-2-mini-detail


Have a nice day and take care of yourself.

I hope the processor you currently have, E5-2640, is actually an E5-2640v3 or E5-2640v4.  If it's not, the processor you ordered will not be compatible with your current motherboard.  The E5-2640 is a Sandy Bridge family processor and uses DDR3 memory.  The E5-2698v4 is a Broadwell family processor and uses DDR4 memory.  The v3 line is the first to use DDR4 memory.  All prior versions used DDR3 (but note that my understanding is that even across DDR3-compatible CPUs, you can't use v2 CPUs in motherboards that were built for v1, but can use v1 CPUs in motherboards built for v2).

I suspect you have compatible processors here and that "E5-2640" is just an incomplete name or a typo or something.   The packages tend to be designed in such a way that fundamentally incompatible processors can't be used in the same socket, so if you really do have a Sandy Bridge processor right now, you just won't be able to put the new CPU in place.

[alexvg]

You're right, I've forgot to write the correct processor name. It's a 2640v3.
Hope having a second processor E5-2698v4 as soon as possible too (dual socket motherboard).

[kcbrown]

You're right, I've forgot to write the correct processor name. It's a 2640v3.
Hope having a second processor E5-2698v4 as soon as possible too (dual socket motherboard).

Ah, good.

The motherboard will support v4 processors as well as v3 ones?  For v1 and v2, I think generally the v2 boards will support v1 processors but not vice versa.  Assessing compatibility can get annoyingly complicated but if your motherboard supports the v4 processors then this looks like an excellent choice, especially if you can pair them up.

[alexvg]

Hi,
I'm just copy-paste last news from my blog :

I've received stuffs to build the compute server needed to analyse all captured data from Siglent SDM3055 A/D converter (AD7190).
I'm also working on network/hosting/servers/emails due to massive increasing of internet attacks.
All these tasks take me long time.

About the new software for the Siglent SDM3055, the resampling implementation (power line sync) is not fully operationnal. But the first tests were very good (around 2 bits improvement over noise free bits).

[killingtime]

It's nice to see this project back on track.

As you have root access to the OS and are re-writing the GUI, a really useful feature would be a custom maths option, and the ability to display results in user definable units.

Use case: measuring the mV from a current shunt. Rarely do you see 1mV per A unless you carefully select the shunt design (just read the mV range in A otherwise). For a shunt that outputs 2.25mv/A (say) you could program the meter to divide the mV range by 2.25 and output to the display in A. Saves having to do the mental arithmetic.

We already have this ability for any meter with SCPI and a computer, but there's no reason it couldn't be done on the meter itself. Rigol have this facility on the 2000 series Oscopes, and R&S on their 1000 series Oscopes (and up), but it's hit and miss as to whether it's implemented or not. I guess manufacturers don't want us programming our own options that could be sold.

[maskedviperus]

I ordered a 3055 yesterday........are there any hw revisions over the years that could cause incompatibility?

Initially reading through this page I was skeptical when I saw the poll asking if people would pay for a firmware.  Even responded "no" to the poll  . But after seeing the youtube videos of what this thing looks like - Yea. Absolutely. On two conditions though....if you release the appropriate linux sources (keep your app layer closed), and if you provide a tool to backup/restore the original calibrations......not sure what the limitations are here, i know you couldnt provide the file but maybe even a tool that accepts an original firmware file to "go back".....Because if im paying for something - I'd be willing to pay a little bit more for it to be the complete package with the things I listed.....

Otherwise open a github and let us join in

[alexvg]

Hi.
I had a few weeks of break shared between my family and my job.
I've wanted to take this break to provide a better work now.

I've read all your private messages or commentaries.
I'll hope to provide you some answer in a near future.


Bye.

Alexia from France.

[bobxxlua]

Dear comrade, how are you? Are you alive? hope the coronavirus is defeated?
Any news? can you need help?

[alexvg]

Hi.
I'm working on the measurement processing.

Here you could find my latest results
" alt="" class="bbc_img" />

[Kleinstein]

The noise is a bit higher than for the 34470, but this is not such a surprise: the 34470 is a 7 digit meter and the SDM3055 is 5.5 digits only. In the 1000 V range the resolution is 10 mV for the display. So the noise level is well below the display resolution.

I think the higher noise is from a different (e.g. 1:1000 or maybe 1:500) divider. The divider can be the dominant noise source for the 1000 V and 100 /200 V range.

[alexvg]

I'm working hard "try-hard" to enhance my "1 PLC" measurement.

My best result is actually
std.dev = 364.9uV
span = 13547.5uV

Measurement settings :
- DMM at 1000V DC
- 2,5V with PDVS2mini
- 50HZ power-line

The Keysight 34470A at 1PLC (auto-zero during 43 mins) provides :
std.dev = 133.6uV
span = 1149uV

[Kleinstein]

The 1000 V range has a very hard stand to get low noise. Chances are the divider is 1:1000 (maybe 1:500) and the divider alone would have quite some noise. So a factor of Sqrt(10) higher noise compared to the meters that can use a divider 1:100 and 10 V range to follow is kind of normal.
There is a bit possible advantage compared to many other meters in that the AZ mode may still be able to do contineous sampling.
The fair comparison may be at a fixed reading rate of lets say some 25 readings per second. With the KS34470 (and many other meters in AZ mode) this would be 1 PLC. With the SD Chip based (and thus ADC with differential input) meters this may be already 2 PLC.

The RMS noise looks good, but the span is quite a bit larger. Is this in an non AZ mode ?

[alexvg]

Hi,
I cannot implement an « AZ mode » with the DMM Siglent SDM3055. Why ? Because I need to permanently and continuously resample source ADC data to 50 or 60Hz (the DMM doesn’t have power-line sync).
I’ve already tried to implement this AZ mode without a very good result, the integrated electronic could not provides enough feature. But I’ve tried… However the resampling always provides the better result !

Internal ADC have a clock rate variation of +/-4%, it's not very precise.
Ideally an PLL connected to the power-line could provide the perfect clock rate... But it's a dream.

I’m using the 1000V DC range because it is very hard to get good results. I try-hard, I work hard !

About the measurement of my processing. I really provide a 1PLC with 50 (or 60) measures per second. All PLC measurements provided with my processing have a settling time of 3ms.
Original Siglent software provides 800ms settling time in slow rate (5 values/sec),  80ms in middle rate (50 values/sec) and 26.7ms in fast rate (150 values/sec).

[Kleinstein]

Not having an extra AZ mode can happen with the SD ADC chips. The ADC chips internally often do some kind of chopping and have very little drift. If combined with a chopper stabilized input amplifier, there may very well be no real need for an AZ mode.
Given a 50 samples per second reading rate the RMS noise is not so bad.  The question is more the large span. So the noise seems to be not just white noise, but somewhat different, e.g. with some outliers. I may be worth looking at the noise in the time domain or look at the Allan varaiance. Resudual mains hum would cause a relatively small span compared to the RMS noise, so this is not the main ususpect.

With the usual higher order filters inside the SD ADC chips there is no real need for a PLL for the clock frequency. 4% tolerance for the clock would suggest an chip internal RC clock, which is not very good. With a higher order fitler this may still give a resonable mains hum suppression. Still a fixed crystal clock would be the best choice: good enough to get good mains hum suppression and low jitter compared to a PLL.
The higher order fitler at the ADC usually also comes with some settling time (e.g. like 60 ms). A very short settling suggest the simple averaging filter and thus not so good mains hum suppression - expecially with the only approximate frequency. It depends on the application which more is more suitable. So the 80 ms settling for the medium rate in the original SW may not be so bad.  Ideally I would like to have the choice of the filtering mode to use.

[2N3055]

Not having an extra AZ mode can happen with the SD ADC chips. The ADC chips internally often do some kind of chopping and have very little drift. If combined with a chopper stabilized input amplifier, there may very well be no real need for an AZ mode.
Given a 50 samples per second reading rate the RMS noise is not so bad.  The question is more the large span. So the noise seems to be not just white noise, but somewhat different, e.g. with some outliers. I may be worth looking at the noise in the time domain or look at the Allan varaiance. Resudual mains hum would cause a relatively small span compared to the RMS noise, so this is not the main ususpect.

With the usual higher order filters inside the SD ADC chips there is no real need for a PLL for the clock frequency. 4% tolerance for the clock would suggest an chip internal RC clock, which is not very good. With a higher order fitler this may still give a resonable mains hum suppression. Still a fixed crystal clock would be the best choice: good enough to get good mains hum suppression and low jitter compared to a PLL.
The higher order fitler at the ADC usually also comes with some settling time (e.g. like 60 ms). A very short settling suggest the simple averaging filter and thus not so good mains hum suppression - expecially with the only approximate frequency. It depends on the application which more is more suitable. So the 80 ms settling for the medium rate in the original SW may not be so bad.  Ideally I would like to have the choice of the filtering mode to use.

Maybe also an FFT of raw data to see if there are concrete frequencies present or histogram to see distribution.

[alexvg]

I have tried too many approaches without really relevant success.
The best result is the ADC running at 1200Hz or 4800Hz then I resample it at a frequency set by PLL locked to the power-line frequency.
The advantage of this method is to have a very large number of samples that can be processed in order to reject noise and get the best measurement value.
I've tried realtime multiple transforms (like FFT, MDCT / MDST, wavelet…) but the settling time is too high... CPU power needs is too high… and this reduce the final precision, a little… The processing of detecting and removing all unwanted harmonics is not very simple in realtime.
The processor in the DMM is not very powerful and I do not wish to apply a software overclock.

Also the AI processing is not possible, yet I thought about it and I would have liked to do it ...

In my latest advances I could provide 5.5 digits in 1000V DC range at 1PLC at 50Hz (or 60Hz) - You could look at my previous post with the measurement table.

" alt="" class="bbc_img" />
" alt="" class="bbc_img" />

[Kleinstein]

There are quite a lot of glitches / far off values in the curve. This may be some spike like background from mains. Another possible cause could be a transfer problem hat the relatively high data rate. The glitches look really bad and make the otherwise good looking curve hard to use.
Most of the glitches seem to be periodic with an slightly below 50 units cycle. What are the units ?

Using the high data rate and than adjust the sampling interval to the actual mains frequency is a nice idea. Getting averaging over a suitable length interval can get the lowest effective noise bandwidth, but the mains suppression depends on a good frequency match. The SINC³ filter of the ADC directly gives better hum suppression, but more sensitivity to white noise and longer settling.

I don't think one would need to do complicated math in the DMM. The main part is some kind of digital filtering, like the simple integration with an adjusted interval. Which fitler is best depends on the application and would ideally be more a user choice.

The more complicated analysis may help to find the source of the glitches and maybe decide on how to minize the effect.

[alexvg]

It's not possible to modify precisely the ADC sampling-rate  and the ADC frequency drift is very problematic.
The curve uses Hz (X) and dB (Y).

I use several complex methods (and maths) in order to approach a better measurement value vs simple average (+ internal ADC sinc3 or sinc4).

I've found the origin of the main glitchs (230Hz, 457Hz, 557Hz...) but I can't do anything... It's a PCB/schematic problem.


About original software :
slow-mode or 5 display/sec is equivalent to 40 PLC (at 50Hz)
middle-mode or 50 display/sec is equivalent to 4 PLC (at 50Hz)
fast-mode or 150 display/sec is equivalent to 1 PLC (at 50Hz)

The ADC does a continuous conversion. At 5Hz, with a settling time of 800ms, the overlap is 600ms. The 5Hz measures displayed each second are not independents.
Using my processing the overlap is always fixed to 3ms or less and a little pause could be set (if needed) to do a fully independent measures, no overlap at all !
I've checked this at 10PLC using a square wave generator at around 2.5Hz to see 2 totaly different measures (this is impossible in the original software).


My goal is to provide the best possible measurement with this limited DMM and a better software. I'm doing my best with the little time I have.

[Kleinstein]

I had not seen the units in the lower left corner. So the data are already after FFT and thus most of the peaks from mains hum.  So mains hum seems to be a least some issue.

230 Hz is a slightly odd frequency for the extra hum. It this something like a fan or display frequency ?



The original SW seems to use the sinc4 filter with chop disabled. The window for the filtering is 40/4/ 1.3 PLC wide,  but this is not simple integration over the whole time. The points at the start and end get a lower weight. Especially the very start and end have quite low weight.
It makes the comparison between data with different fitlers a bit tricky, as the readings are no all idependent, but there is still more information than skipping all the date before getting independent.
The sliding average over data with a higher rate are closer to the integration in the more classical integrating ADCs. This can have advantages, but it can also be worse in some cases. The sinc4 case is already a bit extreme. Sinc2 may some justification.

[oz2cpu]

when you are rewriting the FW for 3055,
is it also possible to fix the rather wrong way the statistics avarage works ??
right now the average readout only display the same number of digits, as the main screen,
remember it sends out two more digits via remote logging and via usb logging.
other units the Keysight 34465A is claimed to be the design master for the siglent DMM series,
their statistics, average show at least one extra digit,
bu the way I also on a SDM3055 and is longtime comparing it with 34465A
I find the 3055 do contain at least one more good digit, when average 10 times,
and 2 extra digits, when averaging 200 times. for many things, I just use external data management and filtering,
I also work on adding temperature controlled regulated heater (very cheap and simple mod)
to the parts affecting drift inside the 3055, just for fun really :-)
I also got access to much better equipment by 10x and by 100x factor, wich is newly calibrated.

[oz2cpu]

the SDM3055 is cheap, and nice in many ways, a perfect platform for modification / upgrade experiments

did anyone map all parts, that affect voltage read out by their temperature change to affected voltage read change ?
the most obvious is the Refferance, but also the ADC , and also other parts, like opamps ? range resistors ? input selector chips ?
do we have a have a schematic to work from ?

[alexvg]

Hi,

I continue the work I've started on the SDM3055.
I'm trying to find a solution to inject my software into the SDM3055 in the simplest way possible.

I'm trying multiple hardware solution
- Ethernet via a telnet access
- Using a USB-Key (but limited to 8GB usb-key capacity)
- Using a SD-card inside the SDM-3055

I'm working on multiple solution to use my software instead of the Siglent.
- Replace all flash memory with my own data (and remove all Siglent software)
- Inject my own software as a default boot software (with an exit button to access to the original Siglent software)

Note
Siglent uses TI and GPL code. After working hard, I could use some parts already inside the Siglent SDM3055 and re-use it in my own firmware (Linux, Linaro, gcc, telnet...).
All bash scripts (from Bookwore Peng) and softwares from Siglent will never be used.
Inside the SDM3055, I've could find surprise softwares : SVN, SDG800 bin, arb, cal, fpga and other data.
I don't know if the scripts from Bookwore Peng are licenced to Siglent or not, but I've already found his code inside some webpages about the AM335x evaluation boards.

I'm working hard.
Thanks for your support.

Alexia.

[tautech]

I continue the work I've started on the SDM3055.
I'm trying to find a solution to inject my software into the SDM3055 in the simplest way possible.

Alexia.

Great to see you back ! 

Details of Alexia's work is here:
https://www.eevblog.com/forum/testgear/hacking-the-siglent-sdm3055/

[alexvg]

Hi,

I've created 2 diagrams because it seems I misexplained my current work on the device.
I'm not working on a new GUI to replace the original Siglent software.
I'm working on another way to operate the device, to process the data and to provide better measurements.

Let's look at how the device works with the original Siglent Software.


Now let's look at my current work



I hope that with these diagrams, my work seems to be clearer to you.

Have a nice day.
Alexia.

[alexvg]

Experimentation on the AM3352 Sitara processor and Siglent digital board of the SDM3055





1. I've done a full backup of the NAND Flash memory (a global NAND file and 12 mtd files)
2. Rename and clean the 12 mtd files and generate 6 files (MLO, u-boot, kernel, rawlogo, rootfs, datafs)
3. I've reset the NAND Flash memory.
4. The SDM3055 is totally crashed/bricked !!!

Why did I lock my device ?
I need to have a step-by-step and verified procedure to allow me to quickly unlock the device if something goes wrong.


The hard way : Restoring the MLO and u-boot
I've written the MLO and u-boot files into a SD-Card under Linux
The SD-Card is /dev/sdc under Linux

Code: [Select]

sudo dd if=/dev/zero of=/dev/sdc bs=1M count=10
sudo dd if=./MLO.img of=/dev/sdc count=1 bs=128k
sudo dd if=./u-boot.img of=/dev/sdc seek=1 bs=384k
My MLO file is 128KiB long, now I can't shorten it.
My U-boot file is 1920KiB long, now I can't shorten it.

I unplug the SDM3055, insert the SD-Card, plug power supply to the SDM3055, short J16 (if J16 exists) and press "power".
I look at the JTAG under Putty and see the process... (need around 2 minutes)
I press "power" and unplug the power supply and remove the SD-Card.

Now the MLO and U-Boot is on the device... It's done !

The easy way : Restoring the device
Under Windows or Linux, create a partition into the SD-Card and format in FAT format.
(Personally I create a 3GB partition).

Copy the 6 files (MLO, u-boot, kernel, rawlogo, rootfs, datafs) into the SD-Card.
Create a uEnv.txt file to read and nand erase/write all 6 files into the SD-Card.

Here's my uEnv.txt (uEnv.txt file only support UNIX line feed)

Code: [Select]

I've removed my uEnv.txt because my SDM3055 suddently crash after working on the filesystem.
I thing my uEnv.txt is wrong.Note : You could create your own uEnv.txt using the U-boot documentation and the mtd4 raw backup. I've seen inside the mtd4 files, it could possible to restore the device from a USB-Key... I must try this.

I unplug the SDM3055, insert the SD-Card, plug power supply to the SDM3055, short J16 (if J16 exists) and press "power".
I look at the JTAG under Putty and see the process... (need around 5 minutes)
The device will boot 2 or 3 times and... application is launched !!!
Turn off the device, remove power supply, remove SD-Card and open J16 (if exists).
Turn on the device, it works !


I've done this for the v19 firmware.
I need to work more to provide all tools, to backup NAND Flash more better and to create a v19 and a v25 compatible backup process.


Other userfull information

Code: [Select]

mtd0 - the MLO (or named SPL)
mtd1 - a copy the MLO (or named SPL.backup1)
mtd2 - a copy the MLO (or named SPL.backup2)
mtd3 - a copy the MLO (or named SPL.backup3)
mtd4 - U-boot
mtd5 - U-boot env (it's empty)
mtd6 - Manufacturedata (it's the raw image, but a u-boot command could upload and convert a BMP file into this area)
mtd7 - rootfs (warning ! not all data needs to be backed up - I'm working and a tool to autocorrect the file)
mtd8 - kernel (it's the Linux kernel)
mtd9 - firmdata0 (your calibration data !!! very important to backup)
mtd10 - firmdata1 (i'll don't backup this)
mtd11 - datafs (warning ! not all data needs to be backed up - I'm working and a tool to autocorrect the file)mtd7 or rootfs is mounted into /
mtd9 or firmdata0 is mounted info /usr/bin/siglent/firmdata0 (this directory is very important, it contains the calibration data of the device)
mtd11 or datafs is mounted into /usr/bin/siglent/usr

The NAND Flash memory mapping

Code: [Select]

Offset                    Size (hex)  Size (KiB)  Description
0x0000000-0x0020000       0x0020000   128KiB      mtd0 - MLO (SPL)
0x0020000-0x0040000       0x0020000   128KiB      mtd1 - MLO (SPL.backup 1)
0x0040000-0x0060000       0x0020000   128KiB      mtd2 - MLO (SPL.backup 2)
0x0060000-0x0080000       0x0020000   128KiB      mtd3 - MLO (SPL.backup 3)
0x0080000-0x0260000       0x01E0000   1920KiB     mtd4 - U-Boot
0x0260000-0x0280000       0x0020000   128KiB      mtd5 - U-Boot Env
0x0280000-0x0580000       0x0300000   3072KiB     mtd6 - Raw logo
0x0580000-0x3080000       0x2b00000   44032KiB    mtd7 - rootfs
0x3080000-0x3680000       0x0600000   6144KiB     mtd8 - Linux Kernel
0x3680000-0x6880000       0x3200000   51200KiB    mtd9 - firmdata0
0x6880000-0x9A80000       0x3200000   51200KiB    mtd10- firmdata1
0x9a80000-0x10000000      0x6580000   103936KiB   mtd11- datafs
Total                     0x10000000  256MiB      NAND Flash
About how to backup data of mtd0 to mtd11
I've noticed a difference of data backed up between U-boot and Linux.

Code: [Select]

U-Boot code to backup MLO into a SD-Card
nand read 0x82000000 0x00000 0x20000
fatwrite mmc 0 0x82000000 backup-MLO.img 0x20000

Linux code to backup MLO into a USB-Key
nanddump -s 0 /dev/mtd0 -f /usr/bin/siglent/usr/mass_storage/U-disk0/MLO.imgThe MLO file saved using the U-Boot command is 128KiB long and cannot be shorten with my own tool to clear backup flash file.
The MLO file saved using Linux command is 132KiB long and can be shorten to 84KiB with my own tool to clear backup flash file.
Why ? I don't understand...

Updated on 2023-02-21 02:24 GMT
I've removed some incorrect or false information.
I've added the Flash NAND memory mapping.
I've added information about how to backup data.

To be continued...
Alexia.

[alexvg]

Hello everyone,

I finally found a solution to make a complete backup of the NAND Flash memory of the SDM3055 for all firmware versions.
The problem with the solution? It's too complex!

Anyway, I'll briefly explain my method.

My first functional solution (quick explanation)
You need to create 3 text files on an uSD-CARD formatted in FAT32. These 3 files contain U-Boot commands to dump the NAND Flash.
You need to boot the device on the uSD-CARD and wait for it to start up completely.
Under Linux, you need to mount and then reduce the images of the rootfs.img and datafs.img files with mkfs.ubifs.
Finally, you need to create a uSD-CARD with the image files in ubifs format as well as the few other dumped image files. All this with a uEnv.txt file.

My second functional solution (quick explanation)
You need to create 3 text files on an uSD-CARD formatted in FAT32. These 3 files contain U-Boot commands to dump the rootfs partition of the NAND Flash.
You need to boot the device on the uSD-CARD and wait for it to start up completely.
Under Linux, you need to mount and then reduce the image of the rootfs.img file with mkfs.ubifs.
From a Siglent update file in ADS format, you need to extract the files to be added to the uSD-CARD and add the ubifs file of rootfs. You need to accompany all this with a uEnv.txt file.

My third solution but not yet functional
You need to place a special version of mkfs.ubifs on a USB key.
With Telnet or Putty, you need to mount the USB key in another path, dump the non-ubi partitions, unmount some partitions, and convert the ubi partitions with mkfs.ubifs.
You will need to copy all the resulting files to a uSD-CARD with the uEnv.txt file.
(Currently, I can extract all ubi partitions except rootfs because my version of mkfs.ubifs is blocked due to a known bug).

The future
For now, I'm putting the NAND Flash memory extraction on hold. I may incorporate an option into my software to extract all data quickly and simply (if possible).
I noted that firmware version 1.01.01.25 changed things for the management of the rootfs partition. But I also noted a bug in version 25 that locks the device (perhaps due to my tinkering with the partitions?)
In the meantime, I'll start working on the software again and I'll post versions of my binaries when they're available.


I am seriously considering distributing my software for free while offering people who wish to financially support me through donations.
So... free of charge !


Thank you for supporting me.

Alexia.

[alexvg]

Hi,

Here is the status of the project :
- I have improved the capture to gain one bit of precision at the ADC level.
- I still have issues with range/function selection (sometimes the selection goes wrong and the measurements are completely off)
- I haven't yet implemented the very high precision capture processing functions.
- I have improved the processor load with a significant decrease: going from 60% to 35% load.
- To date, I have only had one crash during operation, but I haven't yet looked for the cause.
- The DCV, DCI, ACV, ACI, R2W, R4W, and "continuity" functions work. The "diode" function is buggy. The other functions are inactive.

Here is a small capture with a reference set to 1V (1000V / DCV / 1NPLC)

[alexvg]

Hi,
I've done some little improvements but not enough to provide a public beta.
So here is a little video at 60FPS.
https://youtu.be/HVU-4-q3Zkc

Alexia from France.

[alexvg]

Hi,

I've done some little improvements since last post, here is the list :

2023-03-25
- final correction of the clamping measurement
- bug fix of function/range selection

2023-03-26
- better precision (NPLC) of measurement interval (microsecond instead of millisecond)

2023-03-30
- bugfix of random bad measument, the ADC (AD7190) lost the serial interface synchronisation. Apply a fast reset.
- add a better calibration processing from the ADC (AD7190)
- add a better CPU usage (not less, but a better distribution to all threads) 56% (NPLC=1 @ 50Hz)

2023-03-31
- reduce the global CPU usage to 42% (NPLC=1 @ 50Hz)

2023-04-01
- Add measurement of ADC input configurations : Vd (AIN3/AIN4), Vz (AIN2/AIN2), Vr (AIN3), Va (AIN4), Tc (Temperature Sensor)


The first public release will arrive in the some weeks, only if enough features are considered safe and fully functional.

https://twitter.com/AlexiaGossa/status/1643362760892227587

Alexia from France.

[alexvg]

I've done more investigations...

Before the ADC "AD7190", there's a Zero-drift chopper operationnal amplifier, the AD8629.
This op-amp runs at 15KHz and generate some artefacts.

The AD8628/AD8629/AD8630
family uses both auto-zeroing and chopping in a patented pingpong arrangement to obtain lower low frequency noise together
with lower energy at the chopping and auto-zeroing frequencies,
maximizing the signal-to-noise ratio for the majority of
applications without the need for additional filtering. The
relatively high clock frequency of 15 kHz simplifies filter
requirements for a wide, useful noise-free bandwidth.

Source : https://www.analog.com/media/en/technical-documentation/data-sheets/ad8628_8629_8630.pdf

Note :
The Siglent SDM3065 uses a better ADC (the AD7175-2) but with the same op-amp AD8629 (with the same artefacts...).




All magenta marks are power line frequency/harmonics.
All red mars are artefacts :
233Hz at -124dB
461Hz at -141dB
539Hz at -142dB
561Hz at -136dB

Frequencies of 233Hz and 561Hz are artefacts between 15KHz output of the AD8629 and internal sampling frequency of the AD7190.
Not sure for 2 others... (461 and 539Hz, a link with the 500Hz power line harmonic ?)

I need to experiment a filter to remove the 233Hz frequency.

An other solution should be to replace the AD8629 with a ADA4528-2 (better features and an internal frequency of 200KHz)... For the moment, I'll not to make any hardware modification.


Alexia from France.

[Kleinstein]

The AD8629 is a reasonable good compromise between input bias current, current noise and voltge noise.  The ADA4528 has much higher input bias current and current noise - so it would not be a good  choice. Chances are there could be similar interference with other amplifiers too. Overall the artifacts looks small and at an acceptable level.

The normal reading would be averaging over some 100 ms and this alone would attenuate the 233 Hz quite a bit, even if the frequency is not a notch of the response curve.

[alexvg]

The AD8629 is a reasonable good compromise between input bias current, current noise and voltge noise.  The ADA4528 has much higher input bias current and current noise - so it would not be a good  choice. Chances are there could be similar interference with other amplifiers too. Overall the artifacts looks small and at an acceptable level.

The normal reading would be averaging over some 100 ms and this alone would attenuate the 233 Hz quite a bit, even if the frequency is not a notch of the response curve.

With this kind of artifacts, my value search algorithm is disrupted. Perhaps I probably will add an auto-calibration to detect all artifacts on all calibrations and remove them before processing my algorithm. My algorithm uses noise to try to improve accuracy.
Actually, I've reached 7.4 digits / 24.6bits at NPLC=100 over 24 hours (with shorted ADC inputs) without my algorithm.

[alexvg]

Hello,

I've divided my software into 3 main parts: acquisition (hepha), processing (swani), and display (zunia).
Now I've finished the acquisition part and then I've discovered a problem with the Siglent SDM3055.

When there is a power outage for 20 to 40ms, at best, the measurement is distorted (compared to a Keysight 34470A) and at worst, the device reboots !
I discovered this after experiencing small power outages because I had left my UPS in line-interactive mode (my UPS has several modes: 100% online or 100% line-interactive, changing the output frequency to 50 or 60Hz...).
This means that the SDM3055's power supply is under-dimensioned. Not very good for a measurement device.

Since this week, I have been working on rewriting the processing part (swani).


Globally the rewrite is going well, and I am satisfied with the functionality, which is much simpler and more stable but it takes time.

Thank you for your support.


Alexia from France.

[Bill3745]

Hello Alexia,
I have been following your work on reverse engineering the hardware and redesigning the firmware/software of the SDM3055 and congratulate you on how far you have come and what you have been able to accomplish without any hardware modifications to this fairly low end benchtop DMM.  Per your comment regarding powerline outage distortions, do you know if the Keysight uses a linear or switching power supply? Since it costs around 10x the Siglent, I guess we shouldn't expect the same level of performance 

Thank you for your continued work on this and I would be willing to sign up as a beta tester when you are ready to enter that phase of development.

--Bill

[Kleinstein]

Most of the bench meters use a classical transformer - ideally a custom one with extra shields and isolation.  A few than use a switched mode regulator for the supply of the computer part - though less so with the more modern ones as the power needs have come down.
The point to get better tolerance to missing mains periods would be larger filter capacitors or more voltage reserve (and thus more heat).  A larger/additional filter capacitor may be an easy change if really needed. For a reboot chances are this would only be for the logic / µC parts not for the analog part.

A bench meter is normally used with a more well behaved supply in a lab environment, not so much in an more problematic indistrial environment. Still the normal switch over from an UPS should ideally be tolerated.
A better tolerance to supply glitches would still be nice, when using it for automated longer time data logging or experiments. I would not complain much about a distorted measurement as a short time effect - chances are other instruments and the signal source would also no behave than well. So one would likely have a glitch in the data anyway.

[alexvg]

I felt like thinking about something else... with a bit of painting and good weather!



https://twitter.com/AlexiaGossa/status/1662904304099893249

I will answer your questions within a few hours or a few days.
Thanks for your support.

Alexia from France.

[alexvg]

Hello,

The Keysight 34470 uses a "conventional" transformer, just like the Siglent. However, the power supply of the Siglent is undersized for voltage stability variations.

View of the transformer of the Keysight 34470:
https://www.flickr.com/photos/eevblog/16605656689/

View of the Siglent SDM3055:
https://www.eevblog.com/forum/testgear/siglent-sdm3055-first-impressions/msg541312/#msg541312


And I did a bit of painting today...



Have a nice painting day ! 

Alexia from France.

[bobxxlua]

Hello, where can I download your latest updates? I can be your beta tester. Thanks for your work

[uargo]

It´s alive or dead?

[alexvg]

Hello everyone.

I have both bad news and good news, in that order.
For several months I have been absent from the forum and from my project for a new firmware for the SDM3055.
I have been facing serious health problems for several months (https://twitter.com/AlexiaGossa/status/1680524683999944705 and https://twitter.com/AlexiaGossa/status/1714078292603490674).

But the good news are here! My health is gradually improving, but it takes time. Patience is required.
Little by little I will be able to resume my work on the SDM3055.
Another good news, Mouser has informed me that my pending components are finally available.
Over the past few months, I've been jotting down ideas in a notebook for my work on the SDM3055 so as I don't forget anything.
The work on the SDM3055 has opened up another path for me, a medical one : EEG and ECG.

I have received questions about the SDM3045 and SDM3065 asking me if my firmware can work on these models. Theoretically yes for the SDM3045, but I can't test anything, I don't have this model. For the SDM3065 this is currently not possible, there are too many internal differences.
Obviously if I had permanent access to these 2 models I could adapt my firmware there, but each thing in its time.
First health then the SDM3055.


I'm doing my best, but you have to be patient.
Sincerely, I thank you for your support.

[tautech]

Alex, GWS.

Some reading for you here and in the thread and the one it links back to about SDM3045X:
https://www.eevblog.com/forum/testgear/sdm3045x-to-sdm3055x-e-improvement-thread/

[ZhuraYuk]

Wish this would be for SDM3065 as well. It is already fast enough to add one extra digit.

[christopher.robot]

Would love to know if we can support you in some way!

I certainly wouldn't mind donating just to support the work you're doing, regardless of the outcome.

Either way - wishing you a speedy recovery.