Author Topic: E3634A 4 wire connected to electronic load displays incorrect monitor current  (Read 682 times)

0 Members and 1 Guest are viewing this topic.

Offline Helix70Topic starter

  • Supporter
  • ****
  • Posts: 310
  • Country: au
  • VK4JNA
Hi all,

I have an Agilent E3634A that I have repaired, but there is one fault that eludes me. Everything works correctly with static loads, and also when the sense terminals are linked at the terminals.

When I run separate sense leads and power leads to an electronic load (Tenma/Korad) set to constant current 4A, and set the E3634A power supply to 2V at 5A limit (CV), the monitor current reads 1.0-2.0A despite the load drawing 4A as expected. Changing the current limit on the 3634A down to 3.9A forces into CC mode and then the monitor reads accurately as expected.

When I probe the sense terminals when I see the fault, there is a 60 kHz ripple of about 0.5V visible, which disappears when CC is forced or the voltage is higher in CV mode.

Anyone else able to replicate? Is it the PS, the load, the idiot operator? Any other ideas?

Regards,

Josh
 

Offline Ugur

  • Regular Contributor
  • *
  • Posts: 116
  • Country: tr
Have you tested it by connecting a resistor to the output of the power supply?
 

Offline Helix70Topic starter

  • Supporter
  • ****
  • Posts: 310
  • Country: au
  • VK4JNA
Yes, as I said in the initial post, works correctly with static loads.
 

Offline Ugur

  • Regular Contributor
  • *
  • Posts: 116
  • Country: tr
Sorry I missed it. In this case I think there is a problem with the electronic load. Have you tried connecting the electronic load to another power supply to be sure?
« Last Edit: November 14, 2024, 08:45:02 am by Ugur »
 

Offline Helix70Topic starter

  • Supporter
  • ****
  • Posts: 310
  • Country: au
  • VK4JNA
I am probably hoping for someone who can test it on their power supply. The electronic load works fine, and it is a very specific problem, but thanks for your advice.
 

Offline Kean

  • Supporter
  • ****
  • Posts: 2468
  • Country: au
  • Embedded systems & IT consultant
    • Kean Electronics
This oscillation sounds like it could be due to inductance in your wiring.

Can you try making the load carrying leads between the PSU and E-Load as short as possible.  Also twist the sense wires together.
 
The following users thanked this post: Helix70

Offline gburdzin

  • Contributor
  • Posts: 25
  • Country: ca
For your reference...

I tried a similar test with my E3633A power supply set to output 2V with a current limit of 6A, and a Maynuo M9712 electronic load set to CC mode of 5A.  In both cases, with the sense terminals connected at the power supply terminals (via the shorting bars), or with the sense terminals connected via wires to the terminals of the electronic load (shorting bars removed), I am seeing the correct current reading of 5A on the display of the power supply.

 
The following users thanked this post: Helix70

Offline Helix70Topic starter

  • Supporter
  • ****
  • Posts: 310
  • Country: au
  • VK4JNA
This oscillation sounds like it could be due to inductance in your wiring.

Can you try making the load carrying leads between the PSU and E-Load as short as possible.  Also twist the sense wires together.

I can certainly influence the readings by moving the leafs. I tried twisting them together for each load and sense pair, didn't really help. There leads are 1.5m, I will try shorter when I can. I can work around it, just trying to determine if there is another fault lurking, I have replaced 2 precision resistors so far, and wanted to know if this is normal or strange. Thanks
 

Offline Helix70Topic starter

  • Supporter
  • ****
  • Posts: 310
  • Country: au
  • VK4JNA
For your reference...

I tried a similar test with my E3633A power supply set to output 2V with a current limit of 6A, and a Maynuo M9712 electronic load set to CC mode of 5A.  In both cases, with the sense terminals connected at the power supply terminals (via the shorting bars), or with the sense terminals connected via wires to the terminals of the electronic load (shorting bars removed), I am seeing the correct current reading of 5A on the display of the power supply.

Any ripple on the sense terminals with a four wire setup?
 

Offline gburdzin

  • Contributor
  • Posts: 25
  • Country: ca
There is no noticeable ripple to me.
 
The following users thanked this post: Helix70

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
Hi all,

I have an Agilent E3634A that I have repaired, but there is one fault that eludes me. Everything works correctly with static loads, and also when the sense terminals are linked at the terminals.

When I run separate sense leads and power leads to an electronic load (Tenma/Korad) set to constant current 4A, and set the E3634A power supply to 2V at 5A limit (CV), the monitor current reads 1.0-2.0A despite the load drawing 4A as expected. Changing the current limit on the 3634A down to 3.9A forces into CC mode and then the monitor reads accurately as expected.

When I probe the sense terminals when I see the fault, there is a 60 kHz ripple of about 0.5V visible, which disappears when CC is forced or the voltage is higher in CV mode.

Anyone else able to replicate? Is it the PS, the load, the idiot operator? Any other ideas?
This can happen when the control loops of the PSU and DC load start to interact. Connecting a big capacitor (> 1000uf) across the terminals can help to stabilise the control loop. You have to keep in mind that a DC load in constant current mode has a high impedance (lookup the definition of a current source) for which the PSU isn't designed.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 
The following users thanked this post: Helix70

Offline Kean

  • Supporter
  • ****
  • Posts: 2468
  • Country: au
  • Embedded systems & IT consultant
    • Kean Electronics
And a similar issue was just reported by another user with different equipment.

https://www.eevblog.com/forum/testgear/siglent-sdl1000xsdl1000x-e-electronic-load/msg5718259/#msg5718259

It is a reasonably common occurrence, and I believe often goes unnoticed.
 
The following users thanked this post: Helix70


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf