Siglent SDL1000X/SDL1000X-E Electronic Load

[HamDancer]

sorry to but in - but have you checked if you have a MAC address clash? It is rare- but can happen - and more likely to happen if you have multiple devices from same company - usually of questionable origin..

Check the arp table on your computer as well after it stops - and check if something have changed in your computers arp table.

Interesting question, but I didn't have the other Siglent instruments turned on at the same time, so this seems unlikely. It also wouldn't explain why it works with DHCP and not with static IP, since the MAC address wouldn't change. The arp timeout on Windows is 15-45 seconds. I know the ethernet port shuts down completely because when I'm pinging it during shutdown it doesn't even respond to arp requests anymore and I start getting "destination unreachable" errors after the arp cache goes stale. But again everything works just fine when DHCP is on - I've been using it for hours today with no problems. Thanks for the suggestions!

[HamDancer]

OK - here's another bug I found in the SDL1030X-E. In List mode, if you select the "CR" function, the list is missing the units on the "Set" line and any number you enter in the list on that line will take effect and modify the waveform displayed, but the number itself won't actually show up in the list, which stays blank. See attached good (CV) and bad (CR) photos.

Maybe someone else can reproduce this one?

Rob

[Coliban]

Was this already mentioned?

Teledyne LeCroy T3EL150303P

https://de.rs-online.com/web/p/elektronische-lasten/2149913?cm_mmc=DE-PPC-DS3A-_-google-_-DSA_DE_DE_Test+and+Measurement_Index-_-Elektronische-Lasten+%7C+Products-_-DYNAMIC+SEARCH+ADS&matchtype=&aud-827186183686:dsa-1189795610816&gclid=EAIaIQobChMIzoK558en9gIVSOR3Ch2h3gJKEAAYAiAAEgL7kfD_BwE&gclsrc=aw.ds

[kaz911]

haha - slap the Teledyne Lecroy logo on - and double the price..

[kaz911]

EasySDL bug:

I just got my SDL1000x

Have anyone else noticed the time scale on the EasySDL 8.5 seems to not run in sync with elapsed time?

Picture : Screenshot SDL - I reset the graph just seconds before I started the battery drain test. Test timer is > 1h - graph is at < 50 min.

[JDubU]

Interesting question, but I didn't have the other Siglent instruments turned on at the same time, so this seems unlikely. It also wouldn't explain why it works with DHCP and not with static IP, since the MAC address wouldn't change. The arp timeout on Windows is 15-45 seconds. I know the ethernet port shuts down completely because when I'm pinging it during shutdown it doesn't even respond to arp requests anymore and I start getting "destination unreachable" errors after the arp cache goes stale. But again everything works just fine when DHCP is on - I've been using it for hours today with no problems. Thanks for the suggestions!

As a test, have you tried connecting your computer directly to the Siglent using static IP's on each?  You might need an ethernet crossover cable (or another network switch) in case the Siglent can't do cable auto-polarity.  This would eliminate your network switch, router and other network connected devices from the equation.

[HamDancer]

Interesting question, but I didn't have the other Siglent instruments turned on at the same time, so this seems unlikely. It also wouldn't explain why it works with DHCP and not with static IP, since the MAC address wouldn't change. The arp timeout on Windows is 15-45 seconds. I know the ethernet port shuts down completely because when I'm pinging it during shutdown it doesn't even respond to arp requests anymore and I start getting "destination unreachable" errors after the arp cache goes stale. But again everything works just fine when DHCP is on - I've been using it for hours today with no problems. Thanks for the suggestions!

As a test, have you tried connecting your computer directly to the Siglent using static IP's on each?  You might need an ethernet crossover cable (or another network switch) in case the Siglent can't do cable auto-polarity.  This would eliminate your network switch, router and other network connected devices from the equation.

No...there's a point beyond which I can't believe this is my problem. Not only does my computer and network work perfectly with four other Siglent instruments (and dozens of other devices), but if I touch the IP configuration on the SDL configuration page the ethernet wakes up, and this time doesn't go back to sleep again. That would make no sense if it were a problem with my switch or router. I just don't know why no one else is experiencing the same thing. It really smells like an uninitialized variable in the firmware to me, and I just happen to have a bad value stored there. Currently I have my router set to always assign the same IP address based on the MAC address (and it's the same address I was using as a static IP before), and things work perfectly. That's added evidence it's not my computer, cabling, or switch.

Thanks for the suggestions, though. At some point I'll try my SDL with a completely different environment - different cables, router, switch, computer - just to make sure it has to be the unit.

Rob

[HamDancer]

Here's another bug in the SDL1000. This one could be dangerous to your DUT.

• Go into List mode
• Program several steps with different currents/voltages/powers/resistances
• Change the number of steps to be one fewer, so you don't see the final step you entered
• Turn on the load and trigger the list to start executing
• As expected, the list will cycle from step 1->N and then repeat
• While executing step N, hit the trigger again to pause the List at the end of that step; the display will show that it has stopped on that step
• Hit the trigger to resume the List. Instead of resuming at Step 1, as expected, it actually executes Step N+1, a step that isn't displayed or active. Once Step N+1 is done, it cycles back to Step 1 and continues normally.

If Step N+1 were programmed to be something dangerous, like a high current, it could potentially damaged the DUT.

While I'm at it, a feature request: It would be very nice to have an SCPI command that told you which List step (and Program step) was currently being executed.

@tautech can you forward to Siglent?

Rob

[kerouanton]

Happy owner of a new SDL since yesterday, I found the only Siglent distributor in Switzerland was in my street, less than 100m away from home  .

I've read all this thread as usual with my other siglent devices (I've now got all their zoo puppies except the RF generator and VNA), thanks @tv84 for the useful firmware/scpi info 
I was quite lucky because the SDL height is exactly perfect to insert into my "mobile lab shack" as shown on the second picture.

And obviously I spent some time investingating the capabilities, it's quite nice and fits my current ( ) and future needs: Solar panels, mppt controllers, nimh, lifepo4, powerbanks, power supplies and more. A remarkable test was done with my Powerwerx SS-30DV, able to deliver 25A continuous and 30A surge at 14.1V : I'm quite happy with the SDL. I know there are some bugs documented above, but for now it's not a real issue for my needs.

The only problem I constantly have with Siglent, is their software.

I have the feeling either they outsourced the development to someone different for each device in an external company, or they hired a trainee  . I really would love Siglent to take their customer feedback in account for this, and that they start writing an unified software that could handle all their devices (let's say, like Visa-NI that has a single framework with optional plugins per product), or at least rewrite their applications to be more synthetic and similar in terms of UI.

I wonder if @tautech has already given a feedback to Siglent about this, but on my side I feel that they could miss some customer market share because of the low quality of their software. That's a pity because the hardware is well engineered.

Next time I'll visit my Swiss distributor neighboor, I'll have a talk too about this. (oh, btw it's funny to see that the other brand they sell is Agilent. Not exactly the same price/market (they have 220GHz-1THz Spectrum Analyzers for example. Don't ask for the price...), but that means they didn't reject Siglent as being "low-end chinese devices"

[HamDancer]

The only problem I constantly have with Siglent, is their software.

I have the feeling either they outsourced the development to someone different for each device in an external company, or they hired a trainee  . I really would love Siglent to take their customer feedback in account for this, and that they start writing an unified software that could handle all their devices (let's say, like Visa-NI that has a single framework with optional plugins per product), or at least rewrite their applications to be more synthetic and similar in terms of UI.

I agree with you completely...so much so that I took it upon myself to start writing such a software suite. It currently supports the SDL and SPD. It includes full instrument control (at least everything supported by SCPI), line plots, scatter plots, histograms, and, for the SDL, a detailed battery discharge report. For the SPD it adds presets, voltage and current limits, and easier ways of entering numbers.

I would love to have beta testers. If you (or anyone else) is interested, drop me a PM and we can discuss.

-HamDancer

[arcitech]

Hi gang, first post. (thanks for existing!)

TL;DR increasing the limit >300W seems to work be configurable and functional (but I make no claims about reliability/longevity etc.).

Having found myself testing LiPo packs at a continuous 10C discharge current, I found myself butting up against the 300W limit. For 6s packs (25.2V when topped off), I  was able to maintain this 10C test consistency for 6s packs <1200mAh, but anything larger would surpass the 300W ceiling, if only briefly at the start.

I'm happy to report that the

Code: [Select]

SYSTem:DEVICE:POWERUG setting happily accepts values above 300W. The unit will, like other changes mentioned here, require a restart before the UI allows inputs above the previous limit. I can also confirm that the unit was able to actually discharge at 420W for a few seconds -- despite a temp/rigged high static pressure discharge fan fitted to heat sink outlet to help the pusher at the other end, I'm keeping my magic smoke.

I'll likely dial my limit back down to ~330W since my present cap on 6s packs @ 10C is 1300mAh. The time above the 300W "standard" limit for my tests will be brief, since voltage drops several volts shortly after load, and I'm only twiddling this knob because I feel comfortable paying for my own mistakes.

Has anyone else already twiddled around on this front? All apologies if this is not news; I've read this thread a few times but haven't carefully studied it...

YMMV, and of course, proceed with caution if you want to try things out for yourself. (caution and... maybe a peltier retrofit)

[tv84]

It seems you're totally on unchartered territory. So, please post any new data and try not to burn the thing.

[arcitech]

My first couple 6s/1300mAh/10C tests w/ a peak of ~328W have gone just fine. It bears mentioning again that time above 300W is brief, as this is a CC load on a battery. (Don't mean to sound like a broken record, but don't want to cause anyone to cook their gear!)

Since the internal fan's ramp up is rather slow, I'll leave any more "drastic" tests for after I've affixed an external fan to the rear (and a thermocouple affixed to the rear of the heatsink). Previous temp measurements indicated an ~80°C ceiling during 300W CP tests. I just happen to have a 60mm Noctua fan showing up tomorrow which should be an appropriate match...

I'll also do my best to dig around and see whether there's any means of accessing a console on the unit (hopefully without cracking it open, as it's only a few days old). As a newcomer to test equipment, but a bit of an old hat in the linux space (w/ a dash of embedded/yocto), I'd like to wield some more established skills on my new collection of "crappy but hackable" Siglent & Rigol gear (e.g. collecting metrics & serving them through grafana). Perhaps a (truly) programmable battery test routine isn't out of reach...

[HKJ]

The only problem I constantly have with Siglent, is their software.

Have you tried TestController? It support a few hundred test instruments including many Siglent models and works on Windows, Linux and Mac.
On this forum: https://www.eevblog.com/forum/testgear/program-that-can-log-from-many-multimeters/
Download: https://lygte-info.dk/project/TestControllerIntro%20UK.html

[arcitech]

Still chugging along fine with the occasional trips north of the 300W line.

Added these doodads for a much more usable experience, if slightly awkward. Unsure where to obtain the flashy PCB banana option I think I've seen once. Wonder if there's room inside for some flush shrouded banana inputs to retrofit in place of the lugs... but doing it up R&S style might be a plan B. I like what defpom did but I'd want a spare pair to keep fresh before I drilled anything.

Haven't seen an M6 threaded nut to 4mm input out there but I've not scoured either...

[tautech]

Haven't seen an M6 threaded nut to 4mm input out there but I've not scoured either...

flash2b showed us some adapters here:
https://www.eevblog.com/forum/testgear/siglent-sdl1000xsdl1000x-e-electronic-load/msg2512482/#msg2512482

Then later upgraded to the 4mm banana plug nuts here:
https://www.eevblog.com/forum/testgear/siglent-sdl1000xsdl1000x-e-electronic-load/msg2539623/#msg2539623

You can buy these here:
https://nl.aliexpress.com/item/32778229464.html
They are a 6mm stud so the 4mm banana plug nuts wind straight onto the existing 6mm studs without any mods required. 

[arcitech]

Haven't seen an M6 threaded nut to 4mm input out there but I've not scoured either...

flash2b showed us some adapters here:
https://www.eevblog.com/forum/testgear/siglent-sdl1000xsdl1000x-e-electronic-load/msg2512482/#msg2512482

Then later upgraded to the 4mm banana plug nuts here:
https://www.eevblog.com/forum/testgear/siglent-sdl1000xsdl1000x-e-electronic-load/msg2539623/#msg2539623

You can buy these here:
https://nl.aliexpress.com/item/32778229464.html
They are a 6mm stud so the 4mm banana plug nuts wind straight onto the existing 6mm studs without any mods required. 

Thank you, but OTS electrical connectors on Ali Express without any amp rating spec doesn't, to me at least, instil much confidence when working with LiPos on the load.

I do have a short path from the load to a window (screen preemptively yanked!) in case a cell/pack thermal runaway gets underway: lipo in fire blanket out the friggin window onto asphalt. (a sand bucket is too heavy to huck out the window).

The load is also probably too heavy to eject swiftly. As comical as it is to think about, I don't want to have the neighbors see a shattered piece of electronics below. They might think I'm having marital problems. But I am pretty sure a pcb solution solves this that exists, and puts both inputs conveniently on the right edge to be sure inputs aren't blocked.

[Johnny B Good]

 I think you're overthinking the problem. If I was routinely testing LiPo packs, I'd simply keep my pair of 160mm long Knipex cutters handy to chop the wires (one after the other, naturally ) to effect a swift divorce from my precious SDL in order to throw it and the fire blanket into a metal waste bin which will allow me the extra 4 or 5 seconds to tip it safely out of our half landing window from which I can check for the presence of any potential accident victim(s) rather than blindly eject it from a more local but blind sided window.

 My hobby room windows' openings are the top panes making it awkward to safely eject such undesirable flaming objects from, whereas the half landing window opens at the bottom pane with a clear view of what's below making it a much safer option.

 I know there's a small but not insignificant risk of a battery pack fire with LiPos but just how many times a month do you expect to have to chuck a flaming battery pack out of the window?

[arcitech]

I know there's a small but not insignificant risk of a battery pack fire with LiPos but just how many times a month do you expect to have to chuck a flaming battery pack out of the window?

Zero times per month, zero times per life, but I tend to prepare for the unexpected since, way back in my younger years, some oily rags between trash bags, and a half day's time, left only two walls of my childhood home standing. I tend to prepare for the unlikely & unexpected as a result, thankfully w/o being a paranoiac about it (except when caricaturing myself online, of course).

[Johnny B Good]

I know there's a small but not insignificant risk of a battery pack fire with LiPos but just how many times a month do you expect to have to chuck a flaming battery pack out of the window?

Zero times per month, zero times per life, but I tend to prepare for the unexpected since, way back in my younger years, some oily rags between trash bags, and a half day's time, left only two walls of my childhood home standing. I tend to prepare for the unlikely & unexpected as a result, thankfully w/o being a paranoiac about it (except when caricaturing myself online, of course).

 I wasn't deriding you for looking for a "neat and effective swift disconnect" solution to ejecting a burning LiPo pack off the premises. Just suggesting a more pragmatic and equally effective solution between now and finally getting hold of the required parts to eliminate the need to always keep a pair of heavy duty cutters on hand in the event that such a flaming disaster were ever to strike in your lifetime.

 Either way, the key to preventing Sod's Law being invoked (as I'm sure you're already too well aware) is to always be prepared for such surprises (even if it's just keeping a humble pair of cutters on hand to prevent such dramas becoming a total disaster). 

[Bad_Driver]

„Hoping for the best but expecting the worst“ - Jack Reacher. 

[Dirk C.]

Moin,

is there anything else fixed or updated?


2022/9/27   1 .1.1 22    Compatible new EEPROM chip

Dirk

[BillyO]

So I did all the stuff in post #147.  It displays 4 decimal places but will not allow me to set current any finer than 1mA.  Did I miss something?

[TheDefpom]

So I did all the stuff in post #147.  It displays 4 decimal places but will not allow me to set current any finer than 1mA.  Did I miss something?

You cannot set less than 1mA on any of the units, what you get on the X models is the readback resolution to 0.1mA.

[BillyO]

Okay, thank Scott.