Where can I  get an ideal current source device ?

Hi,

The ideal current source, is something that doesn't exist in real life as a standalone component (and in real life it will

only appear to be an 'ideal' current source over a defined range).

For example in real life, a current source circuit may behave like an ideal current source only for a certain current value

(e.g. preset to a value between 0 to 1mA only), and only for some defined load impedance (e.g. from 0ohm to 10kohm).

Furthermore, a real life current source relies on a voltage source (e.g. 12v supply) to function, so the fact that the rest of the

circuit is powered by a 1.2V supply doesn't mean that you won't need another supply too (the current source requires

power to function).

If you wish to step-up a lower voltage to a higher voltage, your best bet is to google for a topology of a 'boost converter', which

is a DC-DC converter which relies on a switching device (MOSFET) and an inductor.

I understand what you are saying shabaz, but I am ready to build this circuit, and cannot without a constant current device that operates with opposite polarities. Yes, ideal constant current sources are ideal and don't exist in reality, but there must be a device manufactured that can operate with AC at its terminals, even if within tolerance.

By 'ideal' I didn't mean precision (i.e. I didn't mean to a perfect accuracy or small accuracy tolerance or 0.1%, 5% accuracy, etc). I understood you

don't mind if it was accurate to 1% or 5% or 20% or 50%.

I mean to an arbitrary current range under any arbitrary load condition. All real-life current sources will only function as long as the load is within a defined range, e.g.

between 0 ohm to 10kohm and as long as the current value you require is set to a value within a certain range (e.g. 0mA to1mA, regardless of tolerance).

As soon as the load goes out of range, the current will drop, and hence it is no longer behaving as a current source.

Also, as mentioned, the current source requires a power source to function (it can't create energy from nowhere).

The current source in SPICE is theoretical only. As an another example, you could set the battery voltage in SPICE to be 1MV, but that doesn't mean a

1MV battery exists. Same with a current source, except that a current source can't be created through just chemistry; it needs power but also some

active component(s) (unless you consider a resistor as a crude current source - which is only acceptable within certain conditions, i.e. where the load

resistance does not change much). This is basically ohm's law.

Also, just because the SPICE symbol allows you to conveniently drop in the ideal current source anywhere in your circuit, doesn't mean you can do the

same with the real life current source. The real life current source will expect the 'top' of the symbol to be connected to

a power supply that can supply the demanded current, and usually that is from a DC supply (and usually it will need to be of a higher voltage for practical current sources).

Real life current sources can work from AC but then they are three-terminal devices using a rectifier bridge and a capacitor, to convert to DC. This still

doesn't get you to what you are asking for. Basically SPICE is allowing you to do stuff you can't always implement in real life. It is just a simulation (and sometimes

a very incorrect simulation, since it relies on simple models of the physical world).

OK, bear with me shabaz. Precisely, I need a constant current source that will maintain a current in One direction while its terminals change polarity. I do understand the theory behind constant current sources, as I have done too many years at university. You are right, a real constant current source will not handle two million volts or 100,000 amps, and it has to be operated within tollerance to function as specified. Is there one that is manufactured using MOSFET or BJT technology, whos parameters allow for a constant direction, constant magnitude current whilst its terminals change polarity. This is my question. Its voltage breakover or current limit are not of concern. I will build circuitry around this. Think about the question and then give an answer. Its a hard one, and I am rarely stumped, but I am at the moment. If I cant get an adequate answer, I may have to research it myselfe. 

@peter,

There is no such device - ie there are no simple component 2 terminal current sources - why not - because it would be a source of energy when the current is in one direction and  a sink for energy in the other. A battery is an approximation of  a two terminal voltage source, when the current is in one direction energy comes out of it , if you force current in the other direction then energy goes into it - if it's  a re-chargeable battery it charges up.

It is possible to make (complex) circuits that work rather like a 4 port current source (two terminals for the power in and two for the current source bit).

There are several components that operate as two terminal current sinks (ie they only work in one direction, the one where you put power from outside) but this won't help you.

MK

Thankyou Michael. Yes, a rechargeable battery maintanes a constant voltage regardless of current direction, and can be considered a constant voltage source. Is there a constant current equivelent (There must be) ?

Well no ......... and yes:

A perfect 1V voltage source in series with 1 ohm behaves EXACTLY like a perfect 1A current source in parallel with 1 ohm. http://en.wikipedia.org/wiki/Th%C3%A9venin%27s_theorem

I don't think this will help you much becaue I think you want a very high parallel resistance, which would be equivalent to a large voltage in series with a large resistance.

I'm afraid that in real life there are no physical equivalents to the current sources in spice.

Just in case there is another way to solve your problem - what are you trying to achieve (eg high efficiency, low cost etc etc).

MK

Neither. Just a constant current source that will maintane a constant current in direction and amplitude, regardless of terminal voltage (plus or minus). If you look closely at the LCR circuit, you will see that I can develope 6,000,000 volts and 200,000 amps from a 1.2 V DC voltage source and a 1 A switched current source. If I used a 10 mA switched current source, I could develope 60,000 volts and 2,000 amps. If I used a 100 uA switched current source, I could develope 600 V and  20 amps, enough to drive a 1500 W, 36V DC motor comfortably. I need a constant current source.

Neither. Just a constant current source that will maintane a constant current in direction and amplitude, regardless of terminal voltage (plus or minus). If you look closely at the LCR circuit, you will see that I can develope 6,000,000 volts and 200,000 amps from a 1.2 V DC voltage source and a 1 A switched current source. If I used a 10 mA switched current source, I could develope 60,000 volts and 2,000 amps. If I used a 100 uA switched current source, I could develope 600 V and  20 amps, enough to drive a 1500 W, 36V DC motor comfortably. I need a constant current source.

I think you are being misled by Spice here, the current source in Spice can provide infinite power, there are no real physical parts like this.

We would all like perpetual motion parts but we aren't going to get them

MK

OK. I bet that someone, somewhere in the world, at some point in history, has built a flyback circuit that has reached a million volts, or more, allbut more complex than this. The point here is that in principle it works, and if I could get a device that supplied one amp at 600V AC, using low loss components, I could get huge currents and voltages. Even with the circuit as is. The idea is that, running DC motors from series connected batteries poses problems, for instance, a 36 V motor requires three 12V lead acid batteries, and one is constrained to this. There is no choice.  Also, if a single cell fails, the circuit fails. And we all know about the memory effect using Ni-Cad's. The inductance in the circuit would be a winding of a brushless motor. By using the inherent inductance, and carfull timing and phase adjustments using feedback and inteligent control under changing loads, a 36V motor can be made to run on 1.2V. The advantages are that all the cells are in parallel, and I believe this offers advantages in reliability, efficiency and complexity. So the discussion is not about whether Spice is 'real', it is about whether there is a constant current source device that supplies a current that apposes its terminal voltage, whether it operates at 1 mA or 10 amps or 1 volt or a million volts. If there is not one out there somewhere, I will eat my shorts.