How to calibrate Siglent SPD1000X / SPD3303X / SPD3303X-E series power supplies?

[BreakingOhmsLaw]

So, a a quick update on this:
My SPD3303 suffered from the fault that it starts resetting after a while. This is due to the overheating bridge rectifier that Dave found in his review.
I've bought mine from siglent.eu webshop in the Netherlands.
Sent it for repair, got it back around three weeks later from Siglent Germany, repaired. Fault report stated "replaced faulty bringe rectifier".

Can't complain about their service.

[ono]

I was calibrating my "upgraded" SPD3303X-E today according https://www.eevblog.com/forum/testgear/how-to-calibrate-siglent-spd1000x-spd3303x-spd3303x-e-series-power-supplies/msg2131678/#msg2131678 on macOS using PyVISA and I got stuck on CALibration:VOLTage SCPI command producing beep despite I was sending perfectly fine commands. So essentially I experienced same thing as described by OP @MartyMacGyver.

Since I wiped voltage and stored wiped values with *CALST prior putting new calibration values, I thought I just screwed my power supply badly and it will show garbage voltage readings from now on. 

But then I started reading around the web if there is a some easier way to access SCPI command line, and yes you can connect to port 5025 (see https://www.siglenteu.com/operating-tip/instrument-socket-and-telnet-port-information/) to have telnet like console. (On Linux/macOS you can use nc 192.168.xxx.yyy 5025). Using this connection method all CALibration:VOLTage got accepted (no beep) and I was able to perform all calibration steps (V reading, I reading, CC) I have now great calibrated device that is at most 1-2mA off from my HP 34401A and Stamos S-LS-60 (aka Korad KEL102) with current readings and CC regulation. Previously CC was off by 8mA with small currents.

Also power cycling the unit is absolutely not necessary. New calibration values start to work immediately upon changing the voltage or current setting. *CALST is only needed to make them persist after power off.

Since others may trip on this as well I decided to post my solution here, despite this thread is pretty old. @BreakingOhmsLaw you could consider updating your instructions recommending direct TCP port 5025 telnet/nc access as the most reliable way that does not require any software.

[BreakingOhmsLaw]

Since others may trip on this as well I decided to post my solution here, despite this thread is pretty old. @BreakingOhmsLaw you could consider updating your instructions recommending direct TCP port 5025 telnet/nc access as the most reliable way that does not require any software.

Thanks! Updated the procedure.
https://www.eevblog.com/forum/testgear/siglent-sds3303x(-e)-calibration-fw1-01-01-02-07r2-best-practice/msg3819530/#msg3819530

[MathWizard]

I haven't had any problem with my hacked 3303x, besides one time I damaged the select-able 5V rail. But I've hardly stressed the PSU, most stuff I've powered was pretty low power, and for fairly short duration tests/etc.


But my cables and connectors aren't the best, and can sometimes drop a few mV or more, I know that's usually fine. But has anyone looked at the circuits in detail, and thought about adding voltage sensing out at the DUT ?

 I'm not sure what would be involved with that, maybe too much stuff would have to be changed.

I only have one, brandname bench PSU (I made one too) so I'm in no rush to open it back up, but I did want to mod the 3rd rail a bit, I think by adding some warning LED's or some protection that was not added, or something, I can't remember, I mapped some of the circuit when I fixed it ages ago.

[mawyatt]

The 3303X is quite a nice supply for the bench, we have three to support our work. We haven't experienced any of the use issues that would make us think about using these for development work with expensive sensitive ICs. The voltage and current settings are accurate and they don't have any nasty overshoots, however having remote sense capability would be nice.

Think the new Siglent supplies tautech mentioned awhile back may have this feature.

For very high current needs we have the Korad KWR102 and 103, which feature remote sensing. We created "special" cables for these which have sense leads built in, this works quite well and gives us an accurate voltage at the cable ends, even with higher currents (30 amps!).

Best,

[tautech]

Think the new Siglent supplies tautech mentioned awhile back may have this feature.

They will.

[MathWizard]

Yeah I guess getting or making some better cables would help. I need some better banana plugs.

But this reminds me, all my main gear, which is Siglent stuff, needs to be opened up and cleaned with the vacuum and toothbrush, or compressor. My 1204X-E, AWG and DMM have never been opened or cleaned in 3-4 years, and I did some house renovations, so I'm sure they have dust in there. I guess all their warranties are up.