Siglent SDL1000X/SDL1000X-E Electronic Load

[tv84]

In the EEPROM dump of a SDL1020X-E it's visible a siglent.jpg at offset 0x8084 (attached is the extracted raw file) besides the S/N, model, settings, etc... I guess people should be able to flash their logo with this command:

SYSTem:DEVICE:JPEGUG

Following the .JPG, right at offset 0xA6F3, there are these 7 fields:

05 00 00 00
F7 26 00 00
0A 00 00 00
53 44 4C 31 30 32 30 58 2D 45 - "SDL1020X-E"   ( Model )    SYSTem:DEVICE:PRODUCTUG

06 00 00 00
0D 27 00 00
07 00 00 00
53 69 67 6C 65 6E 74 - "Siglent"    ( Manufacturer)     SYSTem:DEVICE:MANUFTRUG

07 00 00 00
20 27 00 00
01 00 00 00
31 - "1"    ( Language )          SYSTem:DEVICE:LANGUAGEUG

08 00 00 00
2D 27 00 00
06 00 00 00
30 78 31 36 32 31 - "0x1621"    ( USB PID )      SYSTem:DEVICE:PIDUG

09 00 00 00
3F 27 00 00
06 00 00 00
30 78 46 34 45 43 - "0xF4EC"       ( USB VID = "Atten Electronics / Siglent Technologies" )     SYSTem:DEVICE:VIDUG

0A 00 00 00
51 27 00 00
01 00 00 00
32 - "2"     (Model ID)        SYSTem:DEVICE:PRODUCTUG:ID

15 00 00 00
5E 27 00 00
03 00 00 00
32 30 30 - "200"    ( Max. Power )         SYSTem:DEVICE:POWERUG

Legend:
Red - field # + field offset
Blue - contents size
Black - field contents

-----------------------------------------------------------

There are 6 different Model IDs:  (the green is the only one that i think is correct. Can anyone test others?)
1 - SDL1020X (4 digits)
2 - SDL1020X-E  (3 digits)
3 - (3 digits)
4 - SDL1030X (4 digits)
5 - SDL1030X-E  (3 digits)
6 - (3 digits)

[Alectricity]

Hi,

i would like to buy an electronic load and try to consider between the SDL-1021X-E and Rigol DL3121. it is important for me, that i can control the load (easily) via linux bash. currently i use lxi-tools (https://lxi-tools.github.io/). The Rigol states to support the LXI protocol. i assume, that the rigol can be controled via lxi-tools therefore.
does anyone here know, if the Siglent SDL1000X series can be controled by lxi-tools? i had a look into the programing manual (https://int.siglent.com/upload_file/user/SDL1000X/SDL1000X_Programming_Guide_V1.0.pdf) and see there SCPI commands. but i don't see any statement about lxi protocol support.

anyone know, if the SDL1021X-E can be used with lxi bash command?

thanks!
Alectricity

[tautech]

Hi,

i would like to buy an electronic load and try to consider between the SDL-1021X-E and Rigol DL3121. it is important for me, that i can control the load (easily) via linux bash. currently i use lxi-tools (https://lxi-tools.github.io/). The Rigol states to support the LXI protocol. i assume, that the rigol can be controled via lxi-tools therefore.
does anyone here know, if the Siglent SDL1000X series can be controled by lxi-tools? i had a look into the programing manual (https://int.siglent.com/upload_file/user/SDL1000X/SDL1000X_Programming_Guide_V1.0.pdf) and see there SCPI commands. but i don't see any statement about lxi protocol support.

anyone know, if the SDL1021X-E can be used with lxi bash command?

thanks!
Alectricity

VXI-11 only is the official word from Siglent.

[DeepLink]

SDL1020X-E bug

FW: 1.1.1.21R2
HW: 03-03-020

While logging to USB drive during battery discharge it crashes, display freezes and I have to recycle power
Battery discharge function is fine without logging

The USB drive is an 8GB FAT32, freshly formattet

Will try some other drives / other functions

Have other users seen this before?

BTW - the data logging is only one parameter, and the logging interval is locked to the display update rate in the trend display
Would have been nice to select a minimun of two parameters and been able to set the logging interval

/Hans

[Rydda]

@DeepLink
I have been logging battery discharge to a 16G Kingston DataTraveller 111 without problems.

SDL 1030XE
FW 1.1.1.21
HW 03-03-020 

[sjm]

Yeah... After some fooling around, I was able to write some nice software for the dc load with Rust:

https://github.com/sjm42/my-hacklab-rs/blob/master/src/bin/load_test.rs

I am hooking a 5W transformer to bridge rectifier + capacitor and then programmatically find which is the load current where output voltage is close enough to 80% of the initial voltage with 10mA load current, aka "almost idle".

The 5W figure was revealed by the test, and it makes sense when looking at the random xformer of roughly child fist size.

The algorithm goes like:

• measure initial voltage with 10mA load
• start with 10mA step
• increase current in steps until voltage drops below 80% of original, using 1.5x larger step each time so that it won't take forever.
• now, inside a loop: step load current towards threshold voltage and calculate number of steps:
• - if more than 1 step was needed to cross the voltage threshold, cut the step size in half.
• stop when the step is less than 10mA, it is close enough now.

Sample output:

https://p.siu.ro/t/urgi5o

When running the test I was observing voltage steps and ripples on a 'scope hooked up.
This was a fun experiment. You can get creative with stuff like this. 

-sjm

[zburmeister]

Hi All, first of sorry for the repeat post but I am new here and wasn't aware of this thread. Anyways...

I have been working with the Siglent SDL 1020X-E Electronic Load for a bit and so far so good.  However, I have not been able to get the analog monitor outputs for current and voltage to work on the rear of the unit.  I have connected both outputs to an oscilloscope yet there is no activity on the lines.  I have also looked extensively through the unit's settings but have not found a configuration to enable/disable the ports.  Perhaps it's a firmware issue, perhaps it's a bad device.  Does anyone have any advice? Thanks!

Here is a link to the user manual.  Items 9 and 10 on page 17 are the two ports I am referring to above. http://siglentna.com/wp-content/uploads/dlm_uploads/2019/04/SDL1000X_UserManual_UM0801X-E01A.pdf

[doppelgrau]

It's a bit hidden: Utility -> Config -> EXTC -> I_M_ON AND V_M_ON

[zburmeister]

Thanks! I'll give it a shot when I get to work on Monday.  I appreciate your help

[tautech]

It's a bit hidden: Utility -> Config -> EXTC -> I_M_ON AND V_M_ON

Manual:
https://int.siglent.com/upload_file/user/SDL1000X/SDL1000X_UserManual_UM0801X-E01A.pdf
Chapter 3 System Utility Function P69 > Set EXTC (External Interface) P81

[Omniata]

Just to add, did this on a SDL1020X-E today and it's worked, unlocked to SDL1030X 

[Bad_Driver]

Elektor Store offers the SDL 1020X-E with 15% off these days. (found on the german web page
while looking for the best offer)

I have the strong gut feeling that I need it …..

…and I bought it.

[RAWebb]

So I've finally given in and ordered an SDL1020X-E from Tequipment.net. Intrigued by the up-thread discussion on altering the functional load capacity and resolution by changing the onboard parameters.

Am I correct that the consensus is that the nameplated 200W and 300W instruments have the same hardware, so dissipating that extra power won't let the magic smoke out?

I did successfully apply the FLIR E4 hack some years back (Thanks, eevblog!) so willing to go where no manufacturer really wants me to go. Don't want to set the bench on fire, though ... would scare the Blessed Cat!

[Omniata]

I believe @Elasia confirmed on page 6/post 3: https://www.eevblog.com/forum/testgear/siglent-sdl1000xsdl1000x-e-electronic-load/125/

[oz2cpu]

THANKS again TV84 :-)
my latest baby is now also working

[trp806mo]

Here my two cents after buying last week a new SDL-1020X-E not hacked (last firmware 1.1.1.21R2 / HW 03-03-020) as even with the youtube’s reviews it’s very hard to have real figures on the behavior on this model.

Round A: I monitoring
Green : Current probe E3N (BW 100KHz)  -- connexion between the DC load and the power supply
Blue : Imon of DC load
Red : Voltage diff probe DP1007 on my power supply (Am TTI PL155)

Configuration 1 : DYN 100ms 4.9A/50 ms 0.1A// rise :0.010 // Fall 0.020  // Average 2^6

The I monitoring is not the direct waveform of the output current but a digitalization depending of the averaging (menu config)

Configuration 2 : DYN 100ms 4.9A/50 ms 0.1A// rise :0.010 // Fall 0.020  // Average 2^8

The averaging changes a lot the shape of the waveform so that it's oly a trend ("as specified in the user manual")

[trp806mo]

Round B : V monitoring

Blue : V monitor
Red : V on the power supply with voltage probe

Configuration : DYN 100ms 4.9A/50 ms 0.1A// rise :0.010 // Fall 0.020  // Average 2^8
I configure my current limitation at the limits in order to generate some low falling/rising timing on the output voltage on the power supply


The same result as for I monitor : We don’t have the direct analog shape but a poor digitalization of the output depending of the configuration of the averaging and with a lot of delay  same as for the current useless or only in DC but in this case, you can take a DMM …).

[trp806mo]

Round 3 : I Rmt (which handles directly the output ) :
Red : real voltage on power supply
Green : real current with probe
Blue : voltage for managing the Current trough Vrmt

Configuration 1 : 110Hz // 5A // 36V  // Average 2^8 // square input

The Vrmt is digitalized and is not a direct analog voltage reference for the regulation

Configuration 2 : 110Hz // 5A // 36V  // Average 2^8 // sinus input

I “estimate” the BW at around 10 Hz. The averaging configuration doesn’t change the BW
[ Specified attachment is not available ]

[trp806mo]

Round 4 : Rising/ Falling management

Red : real voltage on power supply in AC
Green : real current with probe

Configuration 1 : DYN  // 5A range// 4.9A during 4 ms at 0.5A/µS   //0.1A during 4 ms at 0.5 A/µS  // // Average 2^6

Rising rate: 0.35 A/µs Falling rate 0.22 A/µS, lower than the setpoints and it’s not coming from the power supply

Configuration 2 : DYN  // 30A range// 4.5A during 4 ms at 2.5A/µS   //0.1A during 4 ms at 2.5 A/µS  // // Average 2^6

Rising rate: 0.61 A/µs Falling rate 0.34 A/µS , still lower than the set points but higher than the 5A range configuration
In this configuration the right management of the rate begins around 0.5A/µs for the rising rate (0.360 for 0.4 setpoint) and for the falling rate there is a lot of difference between the setpoint and the real falling rate but you can match the desired value if it’s below of 0.34A/µs in my case by tweaking the setpoint and checking the reality with a scope.

Configuration 3 : CC / set point 5 A / 2 different rate

By pushing the ON/OFF button in  the CC mode, the SDL manages the rate (no bug)

Remarks :
There are two ranges (0.001-0.1) and (0.1-2.5) and you can’t mix them. The range 0.001-0.1 is much more precise: for 0.099A/µs for both rate I have 0.109 and 0.118 which is correct.




For a very low rate we can notice the threshold of the beginning of the regulation (around 0.6A)

[trp806mo]

Round 5 : Overshoot in CC

Red  battery voltage in AC
Green : Current with probe

Configuration 1 : DYN  // 30A range// 12A during 4 ms at 2.5A/µS   //0A during 4 ms at 2.5 A/µS  // Average 2^6 on a LiIon battery

No overshoot. I can’t go higher than 12A (for 16V) as my SDL is not hacked

Configuration 2 : DYN  // 30A range// 1A during 4 ms at 2.5A/µS   //0A during 4 ms at 2.5 A/µS  // Average 2^6 on a LiIon battery

Overshoot of 270 mA for a set point of 1 A.

Configuration 3 : DYN  // 30A range// 0.3A during 4 ms at 2.5A/µS   //0A during 4 ms at 2.5 A/µS  // Average 2^6 on a LiIon battery

In fact, this overshoot is always present but can be hidden in the ripple noise of the regulation. For 0.3A of set point, its value is around 140 mA

Remark :
The overshoot disappears if the lowest setpoint is not 0 but higher than 10 mA for 0.3A or higher than 20mA for 1A for example:


Or (nor) if we modify the rise/falling rate : hereunder 0.01A/µs for 0.3A :

[trp806mo]

Round 6 : Noise/ripple

Green : current probe in AC
Red : voltage probe on the DC load not the battery

DMM is K2000 for only 4KHz of BW while we have 22Khz of noise ripple

Configuration 1 : CC  30A range  /12.5A set point / RIS/FAL : 2.5A/µs / averaging 2^8

Ipp : 390mA  Irms =125 mA

Configuration 2 : CC  30A range /12.5A  set point / RIS/FAL : 0.01A/µs

Now we have 110 Ipp for 22.6 mA rms still for 12.5A. We always notice 22.6Khz of noise ripple frequency

Configuration 3 : CC  30A range /9.9A set point / RIS/FAL : 0.004A/µs

We are in the noise of the probe. With a DMM (Fluke 45 calibrated) we have around 4mA rms of noise ripple

Configuration 4 : CC  30A range / 1A set point  / RIS/FAL : 2.5A/µs
Real : 0.999 A for  16 mArms  of ripple noise on Keithley 2000

Configuration 5 : CC  30A range /1A  set point / RIS/FAL : 0.004 A/µs
Real 0.999 for 2.3 mArms  of ripple noise on Keithley 2000

Configuration 6 : CC  5A range /1A  set point / RIS/FAL : 0.004 A/µs
Real 0.999A Curiously, the ripple noise is higher on the 5A range and the value is around 4.7 mA rms

Configuration 7 : CC  30A range /0.4A  set point / RIS/FAL : 0.004 A/µs
Real : 0.399A for 2.1mA rms of noise ripple On the 5A range the noise is slightly higher at 2.5mA

Configuration 8 : CC  5A range /0.05A  set point / RIS/FAL : 0.004 A/µs
Real : 49.X mA (X not stable ) for 1.9mA of noise

Configuration 9 : CC  5A range / 3A  set point / RIS/FAL : 2.5 A/µs
Real: 2.998 A for 3.001A on SDL with 18mA of noise ripple (limitation du to DMM) hereunder real waveforms on a resistor of around 1 Ohm in yellow and my curent probe in green  :





The noise depends on the range (5/30) , the rise rate, the fall rate and the output current. It can rise to 4/5% of the setpoint  (a Keithley DC load doesn't exceed 5 mArms on the full range)

[trp806mo]

Round 7 : Accuracy of setpoint in CC mode
configuration 30A range / 150V range/ rising/falling at 0.004A/µs / sense not activated

With the above ripple noise, I have no interest to measure the precision, accuracy, linearity and co, so I’m using a DMM (BW of 4KHz in AC [K2000] for 22Khz of noise-ripple) and give you some figures :
For 12A -> read 11.989A on SDL and 11.975 on DMM
For 5A > read 4.995 on SDL and 4.995 on DMM
For 0.2A -> read 0.197 on SDL and 0.198 on DMM

[trp806mo]

Round 8 : Accuracy of set point of V in CV mode

Configuration : /30A range / 36V range/2^14 / sense activated:
No ripple noise , the noise on the graph hereunder is coming from my Picoscope + DP1007
 


For a set point of 30V : I read 30.008 on SDL and 30.006 on DMM ripple 6.8 mVrms (I is 2.5A for information)
For a set point of 25v I read 25.007 on SDL and 25.006 on DMM  ripple : 6.8 mVrms (I = 5A as I put my power supply in current limitation)
For a set point of 5v : I read 5.003 on SDL and 5.003 on DMM ripple 2.3 mVrms (I = 5A as I put my power supply in current limitation)
For a set point of 1.2v : I read 1.202 on SDL and 1.103 on DMM ripple 1.5 mVrms (I = 5A as I put my power supply in current limitation)

remark :
The resolution of 1mA or 1mV for the setpoint is accurate

   

[Bad_Driver]

Here my two cents after buying last week a new SDL-1020X-E not hacked (last firmware 1.1.1.21R2 / HW 03-03-020) as even with the youtube’s reviews it’s very hard to have real figures on the behavior on this model.

Thanks for this work. You put a lot of effort in it. Since I‘m awaiting my SDL for tomorrow and having no experience with other loads - what‘s your conclusion? Is the SDL enough bang for the bucks? Or should I have had spend my money for another one? Please share your perspective with us!

[RAWebb]

Yes, I saw that earlier however it only mentions that the main boards are the same. I'm curious whether anyone with access to both models could say whether, for example, the FETs are the same temperature rating on each or if the fan on the 300W units has a higher rated volumetric flow rate.