Author Topic: DC Electronic Load  (Read 1010 times)

0 Members and 1 Guest are viewing this topic.

Offline KW

  • Contributor
  • Posts: 7
  • Country: my
DC Electronic Load
« on: November 06, 2018, 04:05:10 pm »
Hi...

I am trying to construct a DC electronic load. The circuit diagram is attached. The opamp in the circuit is configured as a voltage follower.

In this post, the resistor (HSA50R10J) is denoted as RM, whilst the voltage across the resistor is VM. Ideally, VM should track and be identical to Vset. Hence, the load current (ILoad) is given by Vset/RM.

When TL081 is used, the current increases linearly with Vset (see first graph in "EEV Blog 2.png"). However, when Vset is greater than 0.45V, the load current starts to saturate.

When the opamp in the circuit is replaced with TL071, the current also increases linearly with Vset (see 2nd graph in "EEV Blog 2.png"). In this case, the current starts to saturate at a higher Vset, i.e. 0.6V.

Ideally, the voltage at the non-inverting terminal of a negative feedback opamp circuit should be identical to that at the inverting terminal. However, as can be deduced from the experimental results ("EEV Blog 2.png"), the voltage at the inverting terminal can no longer follow up the increasing voltage at the non-inverting terminal after Vset is greater than 0.45V (in the case of TL081) or 0.6V (in the case of TL071). Consequently, the load current starts to saturate after Vset = 0.45V or Vset = 0.6V. Kindly advise why these anomalous behavior (e.g. voltage at opamp input terminals becomes different, current starts to saturate, etc.) happened.
« Last Edit: November 07, 2018, 10:38:38 am by KW »
 

Offline t1d

  • Frequent Contributor
  • **
  • Posts: 698
  • Country: us
Re: DC Electronic Load
« Reply #1 on: November 06, 2018, 04:15:33 pm »
Search this forum for "Jay-Diddy-B"'s electronic load threads. He does a good job of explaining e-load operations and he offers some great designs.

I have just finished building an adaptation of his design. I supplied the schematics and KiCad files.

There are other e-load designs, on this forum, if you want to search them.

https://www.eevblog.com/forum/projects/electronic-load-project-ltv171-irfp250-with-kicad-files/
 
The following users thanked this post: KW

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1015
  • Country: us
Re: DC Electronic Load
« Reply #2 on: November 06, 2018, 04:32:37 pm »
I found this chart on the spec sheet for your MOSFET, which looks a lot like your graph. Try putting another IRF620 in parallel and see if that fixes the problem.

 
The following users thanked this post: KW

Online Zero999

  • Super Contributor
  • ***
  • Posts: 13831
  • Country: gb
  • 0999
Re: DC Electronic Load
« Reply #3 on: November 06, 2018, 04:39:01 pm »
Hi, I am trying to design a DC electronic load for my project. The circuit diagram is at the attachment below. Since the voltage follower configuration was used, I found that the measuring results is not as theoretical results which the gain of voltage follower is as 1. By using TL081 operational amplifier, the load current become constant at 0.45A regardless increasing of Vset. When TL081 replace by TL071, the load current become constant at 0.60A regardless increasing of Vset.  Why the current will become constant regardless increasing of Vset and the gain of voltage follower is not as close as 1? Can anyone explain it? Thank in advance.
The figures you've given in your post are 1/10 those on the graphs.

If you're getting between 4A and 5A, then it's because MOSFET isn't turning on enough. Look at figures 1 & 2 on the data sheet. The op-amp won't be able to drive the gate with the full 15V supply and you shouldn't expect more than around 4A with 12V of gate drive and drain voltage of 12V, especially at higher temperatures.
http://www.vishay.com/docs/91027/sihf620.pdf

Use a MOSFET with a lower on resistance and higher current rating or several in parallel, as mentioned above.

Are graphs are the simulated values and the figures given in your post the measured values?

If so, then it's probably oscillating and your meter is not giving you the correct readings. An additional RC circuit will be required to stabilise the circuit.

See the thread linked below for an example.
https://www.eevblog.com/forum/projects/linear-led-driver-with-mosfet/msg1796558/#msg1796558
 
The following users thanked this post: KW

Offline HB9EVI

  • Frequent Contributor
  • **
  • Posts: 391
  • Country: ch
Re: DC Electronic Load
« Reply #4 on: November 06, 2018, 06:21:52 pm »
my guess'd be oscillations as well, if the opamp is not broken.

- I don't see any decoupling caps on both opamp rails
- I don't see a compensation network

so almost certainly oscillations
 
The following users thanked this post: KW

Offline MarkF

  • Super Contributor
  • ***
  • Posts: 1558
  • Country: us
Re: DC Electronic Load
« Reply #5 on: November 06, 2018, 07:51:42 pm »
Here is what I did:

   

   
« Last Edit: June 05, 2019, 09:37:28 pm by MarkF »
 
The following users thanked this post: KW

Offline KW

  • Contributor
  • Posts: 7
  • Country: my
Re: DC Electronic Load
« Reply #6 on: November 07, 2018, 10:41:23 am »
Are graphs are the simulated values and the figures given in your post the measured values?

Hi... Thanks for your reply. The graphs show the experimental results.

By the way, the original post has been rephrased to describe my problem in more detail.
« Last Edit: November 07, 2018, 10:44:45 am by KW »
 

Offline KW

  • Contributor
  • Posts: 7
  • Country: my
Re: DC Electronic Load
« Reply #7 on: November 07, 2018, 10:43:16 am »
my guess'd be oscillations as well, if the opamp is not broken.

- I don't see any decoupling caps on both opamp rails
- I don't see a compensation network

so almost certainly oscillations

Thanks for the suggestion. I will add decoupling caps (0.1uF) on both opamp rails. I will also learn how to add a compensation network to my circuit.
« Last Edit: November 07, 2018, 10:44:55 am by KW »
 

Online capt bullshot

  • Super Contributor
  • ***
  • Posts: 1773
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: DC Electronic Load
« Reply #8 on: November 07, 2018, 10:52:54 am »
Your circuit just works as it is expected to.
The MOSFET reaches its saturation current, this is why the current doesn't follow the setpoint anymore at some value, and the different values for TL071 and TL081 are the result of real life tolerances. One of the OpAmps is able to put out a bit more voltage at the upper end, so the gate voltage is a bit higher shifting the transistors saturation point a bit higher resulting in a somewhat higher saturated current.
Anyway, it's good practice to put a series (100R ballpark) resistor into the gate, and use some compensation network and decoupling. Did you check your circuit for oscillations with a scope?
Safety devices hinder evolution
 
The following users thanked this post: KW

Offline KW

  • Contributor
  • Posts: 7
  • Country: my
Re: DC Electronic Load
« Reply #9 on: November 07, 2018, 01:24:17 pm »
Your circuit just works as it is expected to.
The MOSFET reaches its saturation current, this is why the current doesn't follow the setpoint anymore at some value, and the different values for TL071 and TL081 are the result of real life tolerances. One of the OpAmps is able to put out a bit more voltage at the upper end, so the gate voltage is a bit higher shifting the transistors saturation point a bit higher resulting in a somewhat higher saturated current.
Anyway, it's good practice to put a series (100R ballpark) resistor into the gate, and use some compensation network and decoupling. Did you check your circuit for oscillations with a scope?

Thank for the suggestion. I will try to check the circuit for oscillations with a scope.
 

Offline brybot

  • Contributor
  • Posts: 20
  • Country: us
Re: DC Electronic Load
« Reply #10 on: November 07, 2018, 07:11:41 pm »
Expanding on 'capt bullshot' post:

I also believe you're operating in saturation.

The IRF620 is rated for 5.2A continuous, at 25C, or 50W maximum power dissipation. When you're hitting 5A at 12V, that's exceeding the maximum rating of the device, which also assumes a really good heatsink.

I'm guessing your FET is heating waaaaay up. Probably over 100C, at which point the continuous current rating goes way down. And peak wattage goes down at 0.4W/C for this chip.

Check out Fig. 4 of the datasheet. Even if you can turn your FET on fully and achieve the 0.8ohm Rdson, the heat is going to push it up and limit your maximum current.

Fig. 1 shows (at 25C) that you need relatively large Vgs to avoid saturation in your application. The IRF620 is designed for fast switching and low on resistance, which probably means it's not ideal as a load, unless you put quite a few of them in parallel with one another.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf