E36313A Triple Output DC P.S. - Review

Table of contents

RoadTest: E36313A Triple Output DC P.S.

Author: geralds

Creation date:

Evaluation Type: Power Supplies

Did you receive all parts the manufacturer stated would be included in the package?: True

What other parts do you consider comparable to this product?: Momentary I've no equal or similar power supply which with i can compare with the E36313A.

What were the biggest problems encountered?: All was easy to use just some (important ? ) menu items are very in background than some other menu items. E.g. in the System - User Settings: the power supply functionality as E3631xA ( all channels are separated) instead as E3631A (CH2 and Ch3 as dual voltage +-) This was set as E3631A (may be during checking the Benchview I switched to this functionality without seeing this). Then I have had a bit a problem to find back to the default settings. I found it then in few minutes.

Detailed Review:

Dear Keysight, Dear Element14 Team,


Thank you for selecting me to participate in this E36313A power supply.

The E36313A is an excellent laboratory device that will do an excellent job in my electronics lab.
I am very happy about it because it fits 100% for all my development work.


My first impression is described in the link above, so I can now dive into the impressions of the functions.

First, following this message from Hendrik Lipka in this link, I immediately tested this error description:
Output (Out)1 - connecting to main power makes a voltage impulse about 10V to 16V as often as you connecting, its varying the waveforms.
I also tried this with the Out 2 and 3. Out 2 and 3 does not make a voltage impulse.

I also tried the power button and the channel buttons.

The spikes are coming only on the ch 1, if you plugs in the power cord.

But in details, the Out 2 and 3 also are bringing out a very small waving signal.

My interpretation with the LEDs:

The channel LED flashing is just the setup power-on routine - the on-board computer is driving up.

During this time the channels are not switching on to the outputs.


My power supply is pretty quiet, has no hum, and the fan is very quiet.
If the loads are rising up about to 70% to the possible full loads the fan motor will turn on.
Also here the fan motor is blowing very silent.
Update, Jan 8, 2018:
Luis Ortiz noticed that the outputs 2 and 3 also produce a spike to the outputs.
Software fault?! - the software freezes and the E36313A stops working.
This is dangerous because the outputs stay in their last state.
When the outputs are switched on, the outputs remain switched on.
Firmware: 1.0.4-1.0.0-1.04

The error comes when you press the "back" button twice.
The error does not occur every time, the software works normally for many hours.
Normal: After pressing the "Back" button to get to the main menu, "lock / unlock" should be displayed in the main menu.
It seems to be a bug in the keyboard routine, namely that the keys pressed event for the up (back) loop is not properly programmed.
Then I have to disconnect and plug in the power cable.
As I write this test and test the functions within the buttons, I think that software is working a bit slow.
I do not want to say that it would be a mistake, it's just a feeling as a technician.
For example: if I click through the keys, then a "beep" always sounds. That's a good thing, and for me it's a bit of getting to know the software.
This works very clean, except for the now found faultless, only slightly slower in feeling than I know in the feeling with embedded systems.
Ok, everyone programs differently. And this software has a lot of work to do.
I say this with the thought, because if I press the "Back" button more slowly, with a greater time interval, this error does not come, at least this has not come.
That's just my personal feeling, because among other things I've programmed keyboard software.

This is followed by the second error, as you see in the video above. The software immediately moves back up to the home position to the main menu.
In this case, then, correctly, not the output is turned on.
But that should be prevented if you plug in the cable again.
The E36313A should only go back to standby and with the power button you have to turn on the E36313A.
Or, ok, this can be a feature - fast restart the power supply.
Two devices under this one hood.
This is an absolutely fine feature for the engineer's workbench.

The E36313A is a direct upgrade to the E3631A.

The E36300 series offers an E3631A mode, which is a compatibility mode that allows the E36312A and E36313A to emulate the E3631A in programming.

This means that the code written for the E3631A can be used directly on the E36312A and E36313A without changing a single line.

The E36313A has three independent outputs which provide positive output voltages in normal operation.
The E3631A has two independent outputs, with output 1 providing 6V / 5A (positive) and outputs 2 and 3 combined as dual outputs with voltages + - 25V / 1A.
Programming Ranges:
The following table shows the maximum voltage, current, and power that can be programmed for each model.
UnitParameter+6V Output+25V Output-25V Output (1)Note
Output IdentifierP6VP25V


Output Number12


1: Applicable for E36311A or when SCPI ID is set to E3631A (E36312A and E36313A).

In normal operation, the E36312A and E36313A will output as +25V.

VoltageMAXimum6.18 V25.75 V0 V2: Applicable for E36313A
MINimum0 V0 V-25.75 V
DEFault (*RST)0 V0 V0 V

5.15 A

10.3 A (2)

1.03 A

2.06 A (2)

1.03 A

2.06 A (2)

MINimum0.001 A0.001 A0.001 A
DEFault (*RST)

5 A

10 A (2)

1 A

2 A (2)

1 A

2 A (2)


How to change the operation modes on the E36313A:
How to change to Tracking mode:
Normal operation mode:

For the test of the E36313A I thought to test the power supply not in the "clinically" pure laboratory, but real approximate under workshop conditions.

So I put the cable lengths longer, several meters, as in the laboratory of a few cm.

For the tests, I used power resistors with fixed values tuned to reach full power limits.



For the switchable load tests I used the 2nd channel as an example for PSU power, similar for the other channels as well.

The 3rd channel was used to switch the loads in channel 2.


The first measurements:

Power consumption - in switched off condition



Power consumption - in switched on condition - main menu - outputs off



The Load - fixed resistors _ 2x 12R _ 1x 0,68R _ 3x 100W; first built



Power consumption 272W- all outputs on - full load




Load regulation, Transient Response as an example with the channel 2 from the power supply.

I thought about doing this test with an active load, as I suggested in my application.
However, I decided to do this test with relays and resistors so that the contact bounces from power on and off will be visible, when the power supply regulates load.
Rather rarely an active load is used in the real circuit, but rather complex circuits with relays, e.g. rather capacitive or inductive character, or real as an ohmic load as well.


The circuit:





Probe with RC-Filter_ Transient Response; Ripple Voltage





Load regulation response_ through the shunt resistor from 100mA to 2 A and back

Settings of the oscilloscope: time 200µs/div; volt: 10mV/div


The shunt resistor has the value 0.5Ohm. Oscilloscope settings: time 200µs/div, voltage_DC 20mV/div 1:10 probe

Load Regulation Response _ switching on and off the current of channel 2 - rising and falling edge1



Load Regulation Response - rising and falling edge 3



Transient response _ contact bouncing - fast regulating during contact switching while the load changes

Settings of the oscilloscope: time 200µs/div; voltage_AC 10mV/div modified probe 1:1 with RC filter


We can clearly see the fast regulation, the overshot and undershot are eliminated very fast;  also the power supply does not switch off the output; oscilloscope settings: time 200µs/div voltage_AC 10mV/div modified probe 1:1 with RC filter.

Transient response _ contact bouncing_ from 100mA to 264mA and back to 100mA



Transient response _ contact bouncing_from 264mA to 2000mA_2 and back to 264mA




Ripple voltage in CV mode and CC mode

Settings of the oscilloscope oscilloscope: time 1ms/div; volt 5mV/div




BenchVue - Remote controlled E36313A via software connected with USB or LAN or GPIB port

One of the finest feature is to control the power supply via software in an automated system.


Here there some impressions:

The BenchVue





BenchVue_Datalogger_ Export to Excel



Remote Control via mobile smart phone

The link for the BenchVue Mobile App:



The installing is very easy, just follow the advice.

Because  some secure items i cannot offer the settings complete, sorry.


In this video it will showed how easily can be controlled the E36313A via mobile smart phone


Some screen shots and images of the BenchVue mobile app:





Power Supply Settings and measurement - Mobile app



Power supply Datalogger - Mobile app






The E36313A, workbench power supply, is a very fine instrument, ideal for the laboratory in development and in the workshop for mass production under somewhat harsh conditions.

In the performance tests, I have chosen such values of the resistors, so that with all channels at the same time the limits can be reached, or something can be exceeded.
Therefore, I can say that I would give full marks for this attitude.

What I'm going to do, are the tests with the sense lines, I have to modify the leads.
I will add this as an attachment in this road test.

-> The firmware is so far well consistent and intuitive easy to use, with a software error has appeared here.

This software error has paralyzed the whole power supply, so I had to turn it off and on.
But during this action, all outputs have been active in their state all the time.
So, if the channels were on then they were on during this error and could not be turned off for safety.
This error must be eliminated very quickly. This can cause a lot of damage to the application.
For this reason, the power supply does not get the full score.
The other reason is that when plugging in the power cord, the outputs bring a voltage pulse to the outputs.
Again, the application may get damaged.

That was my impression of this always absolutely recommendable power supply E36313A.
That's how we know Keysight!
Top quality down to the last detail, yet remaining on the ground of the facts, without quirks and unnecessary gimmicks.

Thank you for allowing me to participate in this great RoadTest!

Best regards