Constant current programmable load, concepts and ideas...

Go To Last Post
12 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I will need soon to design a programmable load to test the 2.5v 60A power supply I have built for my little electrolysis project, and I was wondering if you freaks had any opinions as to what is considered best practice in this case. I do not need something very complicated, merely something to simulate a resistive load with a draw I can control from an AVR linearly from mA to 60A peak.

Is it easier/cheaper to buy an expensive instrument if I will use it very seldomly? Can I make it simple enough that it will not cost me more than about 100$ to build, counting board in for about 10$?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

My latest project was a dummyload :)
It can do 10A upto 20V and linearily derating to 5A @ 40V. The current can be controlled by manual pot or one of the 2 AVR PWM-channels. It incorporates an analog mux for rapid switching, is RS232 controllable (isolated), auto-fan, load and drive indicator, and outputs .csv-file at selectable intervals.

The current-spec can be upgraded in a (relative) simple way.

Contact me via PM if you're interested. Is it a project for personal use, or commercial ?

Nard

Attachment(s): 

A GIF is worth a thousend words   They are called Rosa, Sylvia, Tessa and Tina, You can find them https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There is also the option to rent equipment. But be careful. The rates of renting companies can be so high that in the end you payed more in rent than a new one would have cost you.

Stealing Proteus doesn't make you an engineer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Plons it is entirely for personal use, although I am not excluding the possibility of later reusing it for something else... :p

I don't want anything near as sophisticated as yours for the moment, I am mostly curious about how the load is typically handled? Is it dissipated through resistors or is there a better way to do this? How about using a bunch of high power BJT's?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Elaborating on the idea above, I just realized I am just thinking about an inverted switching current mode regulator, am I not? PWM a bank of MOSFETs to sink current to ground on ON cycle, and not on OFF cycle, keeping frequency high enough that load can be kept near linear to the source... Interesting..

Am I right in assuming I could achieve great sinking efficiency using a multiphase system much like the power supply itself? See attached shot...

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

What part in your schematic will do the dissipation ? And when the fets go from on to off, what will the drainvoltage do ?

In my dummyload, the fets dissipate the power. It uses 4 BUZ11's, and for yours I would suggest 8 fets of a more umpf-type. I made PCB-designs for it: one for the drivers and one for the controller. Tomorrow I will put some stuff online.

Nard

A GIF is worth a thousend words   They are called Rosa, Sylvia, Tessa and Tina, You can find them https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Obviously the example above was oversimplified... ;)

This seems to me like a good candidate for the FETs:

http://search.digikey.com/script...

Has low gate charge, low drain source resistance, low capacitance, and I think with their rating I could do with 2 per phase. Maybe use an external precision resistor in series with the drain as thermal relief, and as current sense for the different phases?

Do I HAVE to dissipate the energy as heat? Could I not sink it to earth via a thick conductor to the house's main GND, using the FETs only to regulate the amount of current being dropped? I understand that the drain voltage would fluctuate with the switching, but will this really be noticeable by the source with the inductor in series, and running at a frequency of around 1MHz?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The basic approach for a constant current load is something like the attached picture. Just vary the input voltage Vin and the current in the FET is controlled by the opamp to Vin/R1.

Any FET can be used, but for 60A, you have to take in consideration the size of R1 and the on resistance (Rdson) of the FET.

Rdson + R1 <= 2.5/60 = 0.04167 ohms

Remember that the Rdson increases with junction temperature (fig. 9 of the data sheet of the FET you showed - IPD060N03L).

Attachment(s): 

Felipe Maimon

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Just an idea (because we had "Earth hour"):

You may feed the energy back to your supply
by using a switched mode supply. Clearly
its a crazy idea and makes no sense at your
power level. But it really would decrease
the heat generated by the system.

Normally the heat is dissipated in transistors, or
in resistors.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

UNiXWHoRe wrote:

Quote:
Do I HAVE to dissipate the energy as heat?
Yes. Unless you do Ossi's trick. But the laws of physics ALWAYS apply. Your PSU delivers 2.6V, you want to load it with 60A --> 150W
My dummyload uses the schematic as posted by fmaimon. But then for 4 fets. The fet you selected does not come in TO220. And 43W per fet, to be dissipated on PCB-copper ? I guess not.

Nard

A GIF is worth a thousend words   They are called Rosa, Sylvia, Tessa and Tina, You can find them https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ok I think I got it now... However I think it will be much much simpler to use my actual intended load as a dummyload... I can adjust the current simply by playing with the electrode plates spacing (much like a pot), and it has about 2 liters of water to dissipate heat...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If that gives you sufficient control-possibilities .... that IS the best solution !

Nard

A GIF is worth a thousend words   They are called Rosa, Sylvia, Tessa and Tina, You can find them https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips