Inexpensive current source?

Discussion in 'General Electronics Chat' started by bdknight, Apr 4, 2008.

  1. bdknight

    Thread Starter New Member

    Apr 4, 2008
    I am a newbie hobbyist in electronics, working through an undergrad electronics text, building each circuit as I come to it to build some practical intuitions as I slog through the math.

    Many of the circuits have current sources, but I am having a lot of trouble finding inexpensive current sources. Those that generate a large range of currents (between 0 and 1 A) are often over a thousand dollars! I could potentially afford up to 500 dollars. Ideally I'd like it to have DC and AC capabilities, but if that will price it out of my range then I'd accept a DC current source.

    Anyone have any suggestions, or know of any products that fit my requirements?

    As an aside, is it possible to simply create a current source by using the Thevenin-Norton equivalences using a standard DC battery, and calculate the voltage I need from my batteries to plug in to get the appropriate Norton current? Please forgive my ignorance, I am really just starting out with all this stuff and I really want to be able to plug in the components I'm reading about to build everything I see, tweak things, play around, really build my intuitions via exploration.
  2. studiot

    AAC Fanatic!

    Nov 9, 2007
    You sure have expensive tastes bro!

    Can you not build your own?
    Linear scale ohmeters use(d) constant current generators in this range, some had pulsed circuits to cut down on average power.
    Most modern bench power supplies have a constant current mode in this range as well, although you would need to provide accurate metering and perhaps replace the adjustment pot with a multiturn type for fine control.

    Constant curent devices are also used as constant current loads for testing purposes, are you perhaps referring to these beasts, which are a tad more expensive?

    Tell us more of your requirements, are you looking for a bench instrument?
  3. bdknight

    Thread Starter New Member

    Apr 4, 2008
    Thanks for replying. Yes, something for a home hobby electronics bench.

    I am a neurophysiologists and this is stuff I should know, so I am self-teaching practical electronics. I have set up a little electronics hobby bench at home (so far, an oscilloscope, various voltage sources, and all the basic equipment), and am simply working through basic electronics books, building the circuits, many of which have current sources. I just found this book which I am going to make my new guide, as it seems more practical than the electrical engineer oriented basic electronics books I've been using (I've been using this book so far, which is probably not the right one to learn practical things).

    I have gotten as far as node/mesh analysis, Thevenin-Norton, that kind of thing, but so many of these example circuits have current sources, and so far I've just skipped them, but I want to tweak around with those too!

    You asked:
    "Can you not build your own?"

    I really don't know much yet, so perhaps not? I did find this online:

    You also said:
    "Most modern bench power supplies have a constant current mode in this range as well, although you would need to provide accurate metering and perhaps replace the adjustment pot with a multiturn type for fine control."

    So you mean something like this? I bought something like that and it acts very strange when I try to use it to "clamp" the current in a circuit at a certain value--it seems to click on and off like I am draining too much power and it was breaking a circuit internally.

    I am at work now and will go home tonight and try to provide more specifics on that (perhaps I already have what I need), get out my multimeter and see what current it is putting out when I hook it up. I did this about two years ago and can't remember all the details. Wifey is away for the weekend so I'll have time to to fiddle with it.

    Again, thanks, and I am embarassed by how little I know. I find it weird that the standard things like powered breadboards and the like all have multiple votage sources but no current sources, but all the intro textbooks have tons of circuits with current sources. Is it that current sources, in practice, are not really used very often? Should I not even worry about it?
  4. Audioguru

    New Member

    Dec 20, 2007
    A current source or constant current circuit works by adjusting the voltage until the current remains the same as the setting.

    If the load's current is less than the setting for the current control then the current circuit does nothing. It will click on and off if the voltage setting is not high enough for the load to maintain the current setting.
  5. SgtWookie


    Jul 17, 2007
    Since you already have a few voltage sources, it's really pretty easy and inexpensive to make a constant current source using an LM317 adjustable voltage regulator and a variable resistor (potentiometer) of suitable wattage.

    National's datasheets for the LM317 and variants can be downloaded from this page:

    I've attached an excerpt showing National's schematic of their handy constant current circuit.

    Note that the minimum resistance is 0.8 Ohms.
    Vref is normally about 1.2V.
    The minimum output current for the LM117/LM317 is 10mA. The LM317L's minimum is 5mA.
    You'll need to use a heat sink for the LM117/LM317.
  6. studiot

    AAC Fanatic!

    Nov 9, 2007
    Curiously, although there are more current drive devices in existance, electronics tends to be about voltage drive.

    When we want voltage drive we often convert the current to a voltage by passing it through a resistor and using the voltage developed across the resistor. this is done in transistor collector resistors, analogue meter range multipliers and many other places.

    So perhaps you would be wise to delay your purchase until you know exactly what you need.

    Current drive is used in old fashioned TV tubes deflection coils, old fashioned meter circuits, old fashioned teletype current loops and old fashioned telephone circuits, old fashioned doorbells and generally in power electrical engineering for heating, lighting and motors.

    Notice the preponderance of the 'old fashioned' in the electronic side.

    I can't imagine what a neurophysioligist would do with 1 amp of curent; 50 milliamps is considered the threshold of fatality.

    Many circuits in elctronics are supplied from a constant current, to provide stable reference voltages, high impedance loads or linear charging for capacitors, but these sub circuits are usually incorporated into the main circuit, preferably close to the thing they are supplying.

    I am signing off for the night now, but no doubt others will have comments.
  7. Wendy


    Mar 24, 2008
    Transistors are current sources too. There is something like this...


    If Vcc were 12V, R1 = 1 ohm, R2 = 1 Kohm, then the current would be approx. 0 - 1.0 Amps. The transistor would need heatsinked, and R1 would need to be at least 15 Watts.

    I think what everyone here is saying is they're easy, you just need to define what you need in the form of specs, both in what the maximum voltage will be if there is no load, and what maximum and minimum currents are needed.

    I could build one for around 5 dollars, $20 if I wanted to jazz it up in a professional box and all.

    If you wanted to go all out and design a power supply with current limiting, then a variable regulator would go after the constant current source.
  8. bdknight

    Thread Starter New Member

    Apr 4, 2008
    Thanks for all the responses. I will have to take the time later to respond more properly, but a couple of people have asked what I want more specifically, so let me try to clarify a bit.

    At lab, I have a voltage source with a knob that I dial and it changes the voltage across the two leads.

    I want that for current, with a range between 0 and 1 amps.

    If that is clear, then stop reading. For those who want more, please keep reading.

    I want this to explore basic circuits (e.g., an RL circuit with a current source) so I can build intuitions about basic circuits, and then go on from there to more complicated stuff like OP-Amps. I don't care if this is old-fashioned. I want to learn the basics and play around with V and I sources.

    Since it is for hobby learning (not to apply to brains of my rats), I think I'll be working in resistance ranges between 1 ohm and 1 Mohm across the circuits. Open an introductory circuit book for engineers, look at any random circuit in the first hundred pages (node/mesh analysis, RC circuits, anything), and about half of them will have current sources. I want to reproduce such circuits and verify the behavior, switch things around and just build intuitions.

    That said, I will take a look at the above recommendation on what to build, and think more. This post was just to clarify what I'm doing.

    Incidentally, the thing I stupidly bought a couple of years ago is a Tenma DC-regulated power supply, as seen here. I obviously don't know how it works.
  9. Wendy


    Mar 24, 2008
    Nice piece of equipment, something most of us would drool over. It has a constant current mode built in. You short the leads, and set the dial for what you're wanting, then set the maximum voltage.

    Good equipment isn't generally a waste of money, there is an old expression, you get what you pay for. That chunck of hardware will do a lot during your expermentation.
  10. bdknight

    Thread Starter New Member

    Apr 4, 2008
    That's very good to know, not least of which because it set me back a lot of money! It is advertised as exactly what I'm talking about (it literally has a current knob), but for some reason it doesn't behave as I expected. I am going to play with it now, maybe take some pics to post here. The documentation that came with it didn't give any more information than the web page I linked above.
  11. studiot

    AAC Fanatic!

    Nov 9, 2007
    To help you get a feel for the subject consider the following, using nothing more than Ohm’s law and expressions for power in a circuit.

    All generators, voltage or current, have an internal resistance in series with their output.

    A perfect current generator has infinite internal resistance.

    A perfect voltage generator has zero internal resistance.

    Practical generators fall short of this ideal.

    However we may approximate a current generator simply by placing a large resistor in series with our load. Large in this context is greater than 10 times the load.
    This technique is used a great deal in circuit design.

    Now for some specifics.

    Power = I^2R so for a 1 amp current
    At 1 ohm power = 1 watt
    At 10 ohms P=10 watt
    At 100 ohms P=100 W (getting large)
    At 1000 ohms P=1KW

    Resistors in semiconductor circuits are normally well above 1Kohm so you can see that pushing 1 amp through would result in enormous expenditure of power.
    Current generators in semiconductor circuits are usually in the 100 microamp to 100 milliamp range, (or 100 times less still within individual integrated circuits). Work out the power levels for yourself and compare.

    Magnetic windings in motors, deflection coils, loudspeakers, solenoids and transformers tend to have resistances below 100 ohms so lend themselves to current drive without such huge power requirements.

    Talking of transformers brings me to your other request – Constant Alternating Current.
    The moment we start to talk about AC other factors come into play.
    Resistance is replaced by impedance
    The AC signal has a frequency to be taken into account. What frequency would your proposed generator run at?
    The signal also has a waveform. Many current drive waveforms are ramp waveforms.

    Keep talking we will get there.
  12. bdknight

    Thread Starter New Member

    Apr 4, 2008
    Ahhh, OK. So being more realistic I should focus on something that pushes no more than 100 mA through, and I'll use small-ish resistors (I'll reconfigure the circuits in the "text" to conform to specs). Note it isn't that important that it literally be a dial, but even a little push-button dealie that I could select from a prearranged set of currents would be fine. I guess I assumed getting a current source analagous to my little voltage function generator that goes from 0 to 30 V would be a trivial thing. I hadn't thought it through in terms of power.

    As for AC, I hate to say it, but it would be cool to have a range of frequencies (sine wave would be sweet, but any periodic signal would be better than nothing if it could be kept cheaper). Ideally, up to 250 Hz or so, but that doesn't really matter. Again, anything to help me fiddle around with little toy circuits I build.

    I am sure even with the AC I am probably being naive as I am thinking of something just like my function generator at lab but that spits out current rather than voltage waveforms.

    Incidentally, that thing I bought, the Tenma 72-6628 power supply, advertised as a working in "constant voltage or current" operation, confused me. I thought that meant I could set a constant current and it would output that current (within the voltage specs of the thing). It doesn't do that. It seems to simply maintain the voltage I set it to, but the voltage regulator is current limited. In other words, it's just a freaking (current limited) voltage generator. Perhaps a really good one, but I just don't even need it. I think I just blew out my multimeter fiddling around with it though. :) I hope I can recoup my losses and sell it...If anyone has experience with this, and I am wrong about it, i.e., it can actually do what I'm talking about here, please let me know.

    I appreciate all this help.
  13. studiot

    AAC Fanatic!

    Nov 9, 2007
    Some more fag packet calculations. Did you do the ones I suggested last time?
    I am encouraging this as this sort of thing is the nitty gritty of getting a feel for a subject.

    Ohms law tells us that for 1 amp flowing the voltage across

    0.1 ohm is 100 millivolts
    1 ohm is 1 volt
    10 ohms is 10 volts
    100 ohms is 100 volts
    1k ohm is 1000volts.

    See how quickly they stack up?

    Your Tenma psu puts out 30 volts. This means the max resistance it can supply 1 amp to is

    1 amp 30 ohms
    0.1 amp 300 ohms
    0.01 amp is 3k

    Try it by placing a few resistors across the supply and measure currents and voltages.

    Now perhaps you can see why such devices are in limited supply and have limited use?
    An individual cc generator has a limited range of application, you nearly have to design one specially for each and every use!

    Years ago I built a simple variable generator for testing the breakdown voltage of transistors and diodes. It puts out 1.5 milliamps max at several hundred volts.

    I also built a linear scale ohmeter with switched ranges. It could supply up to 1 amp on the lower millohms ranges, rather less as the ohms increased.
  14. bdknight

    Thread Starter New Member

    Apr 4, 2008
    OK, its starting to sink in. This is very helpful. I asked at another forum and they got all weird saying there is no reason I would ever want a current source like I'm describing (like the IT guy that said there is no reason I'd want to download and save all my emails onto my hard drive from MS Exchange). Thank you for your patience in explaining the reasons that, in practice, what I'm asking for is something that would require a ridiculous power source, so perhaps I should set my sights a little lower.

    We set up a car battery at lab to pass 16 amps through a solenoid. It's pretty bad @ss. But as you said, we had to build it specifically for the application.

    I've decided that I will re-start all this by working through a more practical electronics book rather than the more mathematical circuit analysis book.
  15. studiot

    AAC Fanatic!

    Nov 9, 2007
    I said we'd get there.
  16. studiot

    AAC Fanatic!

    Nov 9, 2007