200 Watts Electronics load

[tohbas]

Hello

Is there somebody in here try to build Electronics load( resistor mode)?

i am looking for 200Watts electronics load.

Guide me please..

[akutisarizar]

hello Tohbas,

if you need electronics load in ac mode you can use electrical bulbs , but if you need in dc mode you had to buy wire wound resistor load as

http://www.made-in-china.com/china-products/productviewzeEJfahbItkM/Aluminum-Housed-Wire-Wound-Power-Resistor-5w-10w-20w-25w-50w-100w-200w-.html

http://www.alibaba.com/product-free/11845177/Aluminum_Housed_Wire_Wound_Resistor_5w.html

[Parmin]

Why wire wound resistor?

I have been using light bulbs for loads on both AC and DC.. they works great!
they are also very handy to discharge large capacitors.

[Walkura]

Are you sure Parmin ? 
For one time use why not, for regular use hooking up 20+ halogens is just a waste of time.
Right now i'm at work when i get home i'll post some schematics .

[Walkura]

I have a few links for you .
http://electronicdesign.com/Files/29/3239/Figure_01.gif
http://www-d0.fnal.gov/hardware/cal/lvps_info/curr_load/
And if you need really hardcore you can always go pwm & wirewound resistors .
With that setup i builded them in several KW's (if the pwm isnt a problem anyway)

Goodluck

[Parmin]

I admit I don't venture towards WATTS of power often. 
My usual job is in the nano watt range.

However, yeah, lightbulbs done their job admirably.
 As to hooking up 20 bulbs? 
well hehe no problem, there are plenty of bulbs holders available 

[Walkura]

You have a point Parmin
,although i do want to add a maybe important info .
For certain tests lightbulbs can be very usefull .
We used them on numerous accounts for testing under full load .
Although i have to add that lightbulbs have a very low *cold* resistance.
While heating up the resistance rises ,that currentpeak from cold to warm might destroy your test project .
In that aspect resistors are more reliable
Proffessionaly i design and build microprocessor controlled battery chargers .
The picture shows a test setup of a 133 Ampere charger (we still had to improvise)
Setting up aluminium profiles with halogens ,buckets of water with 300 Watt resistors & accu's ,ventilators to keep it sort of within reason.
Its not only far from practicle but its also hmmmmm dangerous .
Your current regulation is never fast enough to prevent a serious blast when you make a short circuit .
Powers like that can go very bad ,very fast (i burned my fair share of cable ,housings etc. by now)
The preliminary solution i designed worked very good and is regulatable up to a Kilowatt continues input power .
At the moment i'm working on a bigger version with some extra's like,
Controlable (and measuring) by plc or computer ,analog optical isolation between powerstage ,logic & computer/plc -
300 Ampere - 45 Volt - capable to withstand spikes up to 150 or 200 Volt .

[Ikaros]

With electronic load I suppose that tohbas meant a device with a potensiometer or other control to adjust the power to be consumed by a resistor of some kind (or transistor).

I want to design such a device long time but had not the time. The idea is to use a power transistor array with the emitter driving a small (0,1-1 ohm) resistor. The voltage on the resistor would be fed back to an op-amp in order to control the current to be flown through the transistors.

A big heatsink is necessary for this.

Also the op-amp and rest electronics will be powered by the circuit under test.

This can be use in AC also with the addition of a bridge.

[Biggles]

I want to design such a device long time but had not the time. The idea is to use a power transistor array with the emitter driving a small (0,1-1 ohm) resistor. The voltage on the resistor would be fed back to an op-amp in order to control the current to be flown through the transistors.

A big heatsink is necessary for this.

Yes.. an emitter follower:

http://en.wikipedia.org/wiki/Common_collector

Big heatsink and high rated components depending on your current and voltages.

[Ikaros]

Not quite. The load is connected at the collector and the emitter with a small resistor (in value not in powe!) to ground. From the emitter the feedback adjusts the current flow with the help of a circuit (around an op-amp maybe and a voltage reference).

[Biggles]

Yes, sorry, my mistake.


You mean like this:

http://en.wikipedia.org/wiki/Current_source

[Ikaros]

ifyou mean the idea of this: http://upload.wikimedia.org/wikipedia/en/9/97/Op-amp_current_source_with_pass_transistor.png

yes. It needs som modifications of course. I just don't have the time now.

[Walkura]

A few posts back i putted some links that bring you to a ready schema .
http://electronicdesign.com/Files/29/3239/Figure_01.gif
http://www-d0.fnal.gov/hardware/cal/lvps_info/curr_load/
http://www.edn.com/article/CA629310.html

Like i stated before ,if you need real big power consider pwm regulation ,wirewound resistors + mosfets .
Powers of several Kilowatts is no problem with a setup like that .

[Ikaros]

That is the basic idea according to these schematics. I just wanted to add some fancy features like digital displays and adjustments to make it more impressive and easy to use.

[Walkura]

In itself its not that hard to add a current sensor and a lm35 .
The measuring Volts Amperes & temperature is easy enough to program in a MCU and put it on lcd.
I been working on something like that but then the PWM variation .
This due to the amount of Amperes (that design was for 300 Ampere upto 45 Volt)
For the time being its on hold cause i lack time to finisdh it .

[Biggles]

I think it could it be made as simply as a microcontroller and a high current transistor.

The mcu can sample the volt drop across the load ( or current sense resistor ), and turn the transistor on and off as required. Some mcu have comparator built in to achieve similar result.  The purpose here is simplicity and small parts count with much features ( in software )

[Ikaros]

Yes it is not difficult to add MCU control and indications, I just don't have the time to make a good looking (and working instrument).

The amprere is not an issue I suppose, as it is possible to just put in parallel and without any resistors more FET's. Just be sure to heat sink them.

Thinking green, if a peltier is attached on the heatsink, will it convert heat to electric power?

[Walkura]

Transistors is ofcourse an option but you are very fast in trouble with proper cooling .
I'll attach the simulation i made for that setup to this post .
When you run the sim and calculate the heatsink required it will be a vulnerable system .
Not to mention those darlington's come with a price of almost 10$ each .
I builded a few prototypes ,SG3525 based and MCU based both were very rugged .
(those were part of a discharger-charger for nickel battery,s 50 Volt - 20 Ampere)
You aint going to heatsink a few to3 transistors with the intention to burn away 1 KW .
Parallel mosfets also create a vulnerability ,if  you blow 1 you'll blow all .
I chosen to split the resistors and switch 5 x  4 or 5 Ampere any cheap mosfet can do that and a to220 radiator is enough .
When you intend to use them in lineair mode you get again the problem of keeping them cool .
(not to mention no dificult control creating an analog voltage for the gates)
Right now i'm sitting at the wrong computer and i don't have multisim here .
Later i will go through my files and can send you the rest of the schema's and pcb,s .

[Ikaros]

With PWM isn't there any problems to the device under test? I mean if the power transistor of the load is switching the maximum load even for a small fraction of the time, the DUT will supply its max current and not the rated. I hope I expressed my query correctly.