Help with finding an electronic variable resistor

[alcazar]

Hello everyone, i need to find a place where to buy (or where to buy the necessary parts and instructions) for an electronic varible resistor. In order to nake a series of tests in some batterys, i need to keep a constant current, and was looking forward to makeing a program which would change the resistance of such device (which would be connected in series with both the battery and load), but in order for that to work i need a variable resistor which could interpret the signals via an RS32 port or something alike. If ANYBODY knows anything, please let me know. Bye

PS: i need one that can go for several hundred ohms, the load is between 5kW and 20 kW, with a 48/24 source, so you guys can imagine the size of resistor i would actually need.

 

[DanDel]

Constant current tests for batteries are typically done with a load bank that switches in and out different small resistors as needed, keeping the current moderately constant. (see

http://www.alber.com/)

Building something more adjustable than that may not be practical/possible.

 

[Skogsgurra]

A little more information is needed to make an educated choice of technology.

1 How constant does your load need to be? Does it matter if current changes 1 %, 5 %, more, less?

2 Are transient deviations allowed? Say 10 % for 100 ms? Or is that too much? One percent for 1 ms? Please specify.

3 Do you absolutely need a resistor - or is regeneration into mains a possibility?

4 Is this a one off thing or do you need to produce many devices like this?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[swz202]

Try Analog Devices,

http://www.analog.com/

and search for "variable resistor"

 

[ScottyUK]

swz202,

I have reservations whether AD will have anything capable of dissipating 20kW, unless you know better?


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[alcazar]

Hello again, this is for a proyect i am trying to do myself, in order to cut down on costs for buying an electronic load (which i am positive will work for keeping a constant current from a battery). The main idea is to test the batterys conected to a UPS, testing both the batterys and UPS (to verify if it can give out the requested power with out failure). I work on a company that sells UPS devices in various power categorys, from 300 VA to 750 KVA. This is only to test the performance before they ship out to the customer (its more like a quality verification).
To SKOGSGURRA: the current can have a +- 5% difference, and about 10 % for 100 ms can be considered as a max.

 

[DanDel]

Don't try to re-invent the wheel. The Alber battery test systems are designed specifically for this.

 

[alcazar]

Jajajaja, that is very true, but i would really like to dig in a little more on this idea, if it isnt possible, i will just have to buy one that does this for me!

 

[Skogsgurra]

OK. +/- 5% and 10 % for 100 ms is an easy spec. You can do it with resistors and contactors - cheap an easy. Use a simple micro-PLC to control the contactors and measure current as voltage drop across the "foot resistor". Mini-PLCs are quite capable and Siemens LOGO!, which I have scrutinized can do the job.

But, there will be heat. And large resistors. Such resistors are available - google "starting resistor" or "power resistor". If you are happy, you will find them on eBay.


Regeneration is a better. You can do that using any old thyristor rectifier capable of the current. The voltage adaption may seem less-than-optimal - and it is. You could use a transformer between thyristors and mains for that. But I wouldn't say it is necessary.

Select a thyristor rectifier with digital control and RS232. You can then set wanted current via the serial interface. You can also (normally) read back actual current and voltage via the serial link.

This will usually give you something like 1 or 2 % accuracy and less than 15 ms step response.

And, do not forget to put a heavy diode in series with the rectifier. That is to protect battery from getting a gigantic overcharge if anything goes wrong.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[swz202]

oops...i totally missed the power requirement part.

I agree with Skogsgurra's resistors + contactors + PLC solution.