Products > Test Equipment

Siglent SPD3303X(-E) calibration FW1.01.01.02.07R2 - best practice

<< < (10/10)

Mortymore:
Worst case scenario: 32V * 3A = 96W, so an 100W resistor will do.

But, but for current calibration one will need to set for 10V and 3A, I believe, so we have, 30W and 3.33Ohm.
3 paralleled 10Ohm resistors will make a 3.33Ohm resistor, and wattage of each can be even lower.

In the safe side I'd use 50W resistors, and would not rely on their accuracy, because a good a multimeter is what's important.
If, for instance, one can't set exactly 2.500A, and 2.432A is what's being measured, then 2.432 is what should be set for calibration.

Hope I'm not disregarding something.

sy:

--- Quote from: Mortymore on June 20, 2024, 10:36:33 am ---Worst case scenario: 32V * 3A = 96W, so an 100W resistor will do.

But, but for current calibration one will need to set for 10V and 3A, I believe, so we have, 30W and 3.33Ohm.
3 paralleled 10Ohm resistors will make a 3.33Ohm resistor, and wattage of each can be even lower.

In the safe side I'd use 50W resistors, and would not rely on their accuracy, because a good a multimeter is what's important.
If, for instance, one can't set exactly 2.500A, and 2.432A is what's being measured, then 2.432 is what should be set for calibration.

Hope I'm not disregarding something.

--- End quote ---

This is my first time calibrating and I want to make sure I got this right before making a purchase so please correct me if I am wrong.

I'm going based of what you're saying, so if I had like 3 parallel resistors, the worst case would be:
P=32V*1A=32W (per resistor)

Then we add some tolerance to make sure the resistor is not operating at its maximum rating:
So choose something like 50W.

But then to get the resistance value, wouldn't I be instead doing:
R=32V/1A=32Ohms (per resistor)

This resistance would limit the current through each branch of the parallel network, while the wattage remains well above the worst case.

Also I'm not sure what types of resistors would be good. I've had a look at some wire wound ones but I've seen them go for like $20 a piece which is absolutely ridiculous!

And also, when I am testing this current, do I need to hook the load up and keep it running for an hour to stabilise, or is it leaving the power supply on for an hour before connecting the load?

TIA!

Edit:

Taking another look, I've found some of these $7 resistors. Would they be good enough to use for this prupose?
https://au.mouser.com/ProductDetail/YAGEO/AHA50AJB-30R?qs=sGAEpiMZZMtlubZbdhIBIP1s4nuCZf1cZfhdKdBp1aw%3D

Mortymore:
From OP...


--- Quote from: BreakingOhmsLaw on November 18, 2021, 12:21:29 am ---...

Current display:
...
Set PSU channel to 10V, 3A.
Connect to an electronic load and set it to 100mA constant current. Check actual current with multimeter and enter into the command below:
CALibration:CURRent ch1,1,<measured current value> or ch2 for CH2
Change the electronic load to 2.5A constant current. Again, actual current with multimeter and enter into the command below:
CALibration:CURRent ch1,2,<measured current value> or ch2 for CH2
...

--- End quote ---

From this description, for current calibration you will need to set the PSU for 10v, 3 A (current limit). And you will also need to draw 2.5A.
3 // 10 Ohm = 3.33 Ohm => 10V / 3.33 Ohm = 3A => P = V^2/R = 10^2/3.33 = 30W
2 // 8 Ohm = 4 Ohm => 10V / 4 Ohm = 2.5 A => P = 10^2/4 = 25W
And so on, for any given current you might need for calibration.

I built some dummy loads with power resistors and repurpose old CPU coolers and PSU cases. The one in the image attached has only 2x 8Ohm 100W resistors to load audio amplifiers for testing/repairing/adjusting. With a setup of a few power resistors, several resistance values can be achieved by different resistor associations in series or parallel, and that's handy for many purposes.

I didn't find trivial to calibrate the SPD3303, and only did it fully when I had a 5.5 digit multimeter and an SDL1020 available.
But with care, I think with power resistors as load, and given an hour to set everything, the calibration can be done.
I like the DL24 to test batteries and such... but I wouldn't trust it to calibrate this PSU

EDIT: Maybe this thread about dummy loads could be of interest:
https://www.eevblog.com/forum/beginners/dummy-load-for-bench-power-supply-resistors/

lugamodder:

--- Quote from: Wade2019 on May 29, 2024, 06:56:10 am ---I tested SCPI:

CAL:VOLT ch1,1,0.833     ERROR  The beep of the buzzer
CAL:VOLT ch1,1,0.83311     ERROR  The beep of the buzzer
CALibration:VOLTage ch1,1,0.8613   ERROR  The beep of the buzzer
CALibration:VOLTage ch1,1,0.861   ERROR  The beep of the buzzer

I am using the software NI MAX and my power supply is SPD-3303D    Seeking everyone's help

--- End quote ---

Have you managed to solve the calibration problem? I also have SPD3303D fw 1.01.01.06R2 , and it doesn't accept the "CALibration:VOLTage" commands."

SuperD:
I am right there with you. I calibrated once before and it was a bit of experimenting before I got it. I have v3 hardware and needed to use "SPD3303X_1.01.01.02.05" firmware and I was able to send the commands using "EasyPower_1.01.01.18".

I recently got a more accurate benchtop meter and went to calibrate mine again. But now I am stuck on the "*CALCLS #" cleared values permanently. I can send the power supply values in Putty or EasyPower, but the voltage on my SPD3303X will not adjust after the new values are set. If I go back to the "*CALCLS #" values, it's adjustable on the whole range, but the display numbers are way off.

Maybe a caution to users, I don't know if this would have prevented it. But if you don't have to reboot after *CALCLS, then you may not need to send *CALST after clearing either. *CALST is just to store permanently. But, if you don't have to reboot after the *CALCLS, and all you have to do is change the voltage to get the cleared values. *CALST may be an unnecessary risk if you get stuck not being able to write new values.

My theory here is, if you *CALCLS, then just change the voltage on the power supply, don't just *CALST and make the cleared values permanent. If you find you can't write in new numbers, just reboot the power supply and at worst you might be back to where you started on the factory calibration. Way better than the uncalibrated values. Worth a shot if you are just starting out.

As for me, what I think happened in my case is that I saved my cal values in an Excel sheet with my voltage table. When I copied the cell, I think I copied a crlf at the end. I think that may have messed up the calibration values stored in the PS somewhere. I didn't catch the cursor dropping 1 line until I tried pasting into Putty. By then I already had sent the command in EasyPower and NI MAX.

Has anyone dug any deeper into the config table? Thinking I need a backup of a good file. Or possibly to go in and edit the config file somehow, possibly removing the crlf. Jtag maybe? Reverting firmware to all versions I have, does not fix mine.

I am quite sure it's writing. I just can't adjust voltage after I set the measured voltages. When I do the power supply is stuck on the last saved voltage I set, even after a reboot. To get it back to "somewhat functional" (but very off calibration). I need to send the *CALCLS commands, and then *CALST. Now my 1 volt is really .84. Depressing.. But, my fault too.

My new numbers that I want to set, in case someone notices a mistake - Keysight 34465A
CAL:VOLT CH1,1,0.840
CAL:VOLT CH1,2,25.680
CAL:VOLT ch2,1,0.875
CAL:VOLT ch2,2,25.668

My numbers the first time - Fluke 87III
CAL:VOLT CH1,1,0.8401
CAL:VOLT CH1,2,25.67
CAL:VOLT CH2,1,0.855
CAL:VOLT CH2,2,25.75

Navigation

[0] Message Index

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod