Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

High precision, good accuracy Current Source Schematic

  1. Feb 22, 2012 #1
    I am looking for a schematic diagram of a compact, temperature compensated, stabilized, high precision (accuracy within 0.5% or less) current source.
    The target current should be easily set to 100/150 mA, no matter if discrete or integrated solutions must be used.
    This project will be related to supply current to the Hall Sensor.

    Thanks for any addressing.
  2. jcsd
  3. Feb 22, 2012 #2
    It should not be hard, just use a simple precision op-amp and a MOSFET.

    The op-amp is in non inverting configuration with

    1) +ve input as voltage control. Input voltage call Vref
    2)The output of the op-amp drive into the gate of the N-MOSFET.
    3) A resistor of say 10K connect from the source of the FET to the negative input of the op-amp.
    4) A 200pF cap from output of the op-amp back to -ve input for stability.
    5) A precision resistor of 0.1% or better from the source of the FET to ground. This is the resistor that set the constant current. The drain is the constant current source. The resistor called Rs.

    The current is equal to Vref/Rs. getting 0.5% is easy. You want better, get better Rs and better op-amp. You want the other polarity, use P-MOS. I am too lazy to draw, scan, upload the drawing. It should be easy enough following step by step. Any question, post back.
  4. Feb 22, 2012 #3
    Thank yungman!

    Now you've intrigued me and I really want to see the schematic from your own handwriting :smile:
    Seriously, can you advise me some schematic from the web or suggestions about components?

    And what about the temperature compensation? This is a quite critical aspect.
  5. Feb 22, 2012 #4
    Op-amp like OP-27 is not bad. But look at Digikey:


    Look at Analog Devices, Linear Technology.

    The thing you need to look for low offset voltage, that is part of your error. offset current and bias current is not important.

    If you really want to be picky, get 4 terminal resistor for Rs. 4 terminal resistor has two terminals( sense terminal) that hook up to the two end of the resistor body to eliminate the voltage drop on the lead that give you error.

    What you ask is not that critical, make sure you have at least 2 volts across the Rs, then the little offset error of the op-amp is not even important.

    Temperature compensation? get low drift component and you'll be fine. You can get some really precision resistors!!! Look into Vishay and particular Caddock resistors.
  6. Feb 22, 2012 #5

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    National's AN31 is a handy one to print and keep someplace in hardcopy. It's a collection of ad-hoc op amp circuits. It's so old the opamps they suggest are mostly outdated, but what a great "How To" resource.

    www.national.com/an/AN/AN-31.pdf [Broken]
    "The Handyman's Secret Weapon"

    current sources page 13
    adjustable capacitance multiplier page 24
    Last edited by a moderator: May 5, 2017
  7. Feb 23, 2012 #6
    Thanks guys,

    I see the light.

    From TI Op Amp Circuit Collection I've found this schematic:

    http://desmond.imageshack.us/Himg834/scaled.php?server=834&filename=precisioncurrentsource.png&res=medium [Broken]

    In needs a negative Vin, while there is only a +5 Vcc available.
    Also, I think the temp regulation is not so accurate here.
    Any more idea?
    Last edited by a moderator: May 5, 2017
  8. Feb 23, 2012 #7

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    looks from its datasheet like your sensor is ~2 ohms ?

    with only +5volts available there's not much headroom for opamps, but i've used LM324's there.

    see if this looks attractive to you

    bottom of page 19 current limiter
    of course you'd use a fixed resistor , maybe with a small trim..

    that 117 regulator should be okay with 5 volt supply, the regulator needs 3 volts across its I/O terminals, and 1.25 more for control terminal
    and your 2 ohm sensor will drop 0.3, leaving

    5 - 3 - 1.25 - 0.3 = 0.45 so you haven't a lot of room, but enough by the book.

    LM117 says temperature stability is 1% over whole range -55 to +150 (chart page 6) so it'll be important to select a stable current measuring resistor. Dont eat into your voltage margin with long connecting wires.

    Yungman's expertise is leagues above mine here.

    old jim
  9. Feb 23, 2012 #8
    Hi again Jim,

    Yes, the HGT-3030 from Lakeshore has 2 Ω as approx input resistance.

    I guess you're talking about this circuit (906316):

    This has been a mistake of mine. It looks like the output voltage is 3V3 rather than 5V, so maybe this makes things a bit more tricky. On the other hand, we can also provide a supplementary line of 5V, with very few additional components, I guess.

    Sounds good.

    Yungman is more than welcome!

    Just for speaking, what do you think about the following schematic from Maxim, which uses an op amp, a transistor, and a resistor ?


    Here there is an application note for DS4303/DS4305, but I am puzzled about the ADJ procedure as well as about how the BJT can operate within its linear range...

    Any further consideration is greatly appreciated.

  10. Feb 23, 2012 #9
    Your are too humble, I don't think so. But thanks

  11. Feb 23, 2012 #10
    I see you need current source from the top. This is the circuit I come up really quick. Double check whether I make any mistake:
    150118[/ATTACH]"] 34zcb2o.jpg

    The reason I kept pushing for MOSFET is because it does not draw any gate current and is a true current source with high output impedance. Using simulate current source implemented with a voltage source has a disadvantage of slow change with changing load impedance and is really not low impedance output.

    The idea of the drawing is a differential amplifier with reference to the top sense lead of Rs. The resistor Rf and R1 should be in single match package. So is R3 and R4. I think Vishey sell these precision resistor pair specially for this. I just put the ratio of 1:1, other ratio is possible. You have to work out the detail. One issue is the +5V supply which is a little low for this circuit. It is a little low for even the other ones here!!!!

    The adjustable battery is the current control where I=V/Rs. The op-amp voltage follower is used to buffer the sense lead of the Rs so it does not draw any current by the differential amplifier. It is important to choose an op-amp with low offset voltage drift as it contribute to temp. error coef. Rf, R1, R2 and R3 should be under 10K, the capacitor should be about 200pF for stability. The 1K resistor from the output of the op-amp to the gate is to isolate the output from the gate capacitance.

    Again, double check my work as I am having fun and it's your neck!!!:surprised:tongue2: :rofl:

    Attached Files:

  12. Feb 23, 2012 #11

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    I'm sorry maxim - it's page 19 of the document which is page 20 of the PDF file.

    How do you copy those drawings for posting?

    Anyhow look at bottom of page with number 19 at bottom which is page 20 of pdf document, for little circuit " precision current limiter"

    it's just the IC and one current measuring resistor....

    i will read your 4303 datasheet and see what it is.

    Basic approach is to set a reference , measure current with a precise resistor and wire an opamp to keep them equal.
    Only two difficult parts are a stable resistor to measure current, and a stable reference voltage for desired current.

    LM117 has a decent reference and opamp internal to it.

    Does this need to be computer controlled, adjustable by a knob, or set it once and leave it there?

    I've never worked with 3 volt opamps but they abound. One of them and a Mosfet should work, i think, if we give it a good reference like the MAX6520 and get a good current sampling resistor.


    If i can figure out how to draw something and post it, will do.

    Only trouble i see with your circuit from Maxim appnote is the measured current is referenced to V+ while the 4303's output voltage is referenced to circuit common - so if done exactly as shown current would fluctuate with supply variations.

    Would you and Yungman check on my thinking?

    having trouble getting this to post... hope it appears
  13. Feb 23, 2012 #12
    I like the MOSFET in the output stage, as well as two op-amp!
    No too complicated, despite the fact the V is only 3V3 or 5V.
    Iout should be 100 mA, thus using 3.3V as input voltage and Rs = 33 Ω low tolerance, I should be on the right side.
    What is not clear for me is the limit of V=5V as an issue you mentioned earlier.

    But ok! now is the time to find components.
  14. Feb 23, 2012 #13

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    Here's a current injector i built years ago.


    Yungman - Were R3 shunted with a 2 terminal reference and A1 changed to a 3 volt rail to rail amplifier, would we be there? Tweak R values of course.

    This worked great with LM324 but i had to give it higher supply, it needs 1.5 volts headroom to V+ on inputs , though they'll sense down to negative rail.
  15. Feb 23, 2012 #14
    Just a screenshot, save it as jpg/png image and upload to an external sharing server.

    So, this guy here?


    This stage is only for providing a stable current to the sensor Hall. The output from that sensor will be processed by a PC. No needs to be remotely controled, just set ot once.

    It looks ok for me.
  16. Feb 23, 2012 #15

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    oops i'll digest yours now

    be back tonite, need the daylight for car repairs...

    old jim
  17. Feb 23, 2012 #16

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    maxim - that's the one. Simple to the max, isn't it ??

    old jim
  18. Feb 23, 2012 #17
    Both this and my circuit use a MOSFET, I am worry about the gate voltage eat up a lot of the supply voltage. Some MOSFET need about 3V between gate and source to work, there goes 3 out of 5V!!!! I think both work if you can get it to work for 5V supply no matter what. I think the MOSFET can be replaced with a darlington hybrid with a p-JFET and a PNP. JFET is to really lower the gate current to get more accuracy and if the drain of the FET drive the PNP base, you don't loss any current at the process. Amps are easy to find now a days. I have been designing using 6V single supply op-amp that can work down to 1.8V single rail. A lot of them are rail to rail. It is the pass transistor that I am worry about.
  19. Feb 23, 2012 #18
    There are plenty of op-amp that work down to 1.8V single supply!!! That's not the problem. The uncertainty for me is the gate source turn on voltage is about 3V, that eat up your supply voltage. Look into it in more detail and see whether you can make it work for 5V. Also a hybrid darlington using a p-JFET and a PNP might help as the gate source voltage is in opposite direction and it might even help. You have to do the calculation, I am only giving a suggestion. Try look for low threshold P-MOSFET.

    Maybe it's not a problem. If you can find a P-MOS with operating gate source voltage of -3V, you still have 2V on Rs to play with. Say if the op-amp output is at 0V for max current, the source will be at +3V, so operating voltage for constant current output at the drain can be from 0 to about +3V. Maybe it's not a problem.
  20. Feb 23, 2012 #19

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    i think i used a n mosfet, could just swap opamp's +&- inputs for p installed other way..

    here's an interesting transistor..
    see curve Ic versus Vce, pages aren't numbered it's about middle of second page.

    will those 3 volt opamps drive as much as a milliamp?

    never mind, just looked at one.
    http://www.ti.com/lit/ds/symlink/lmv711-n.pdf [Broken]

    old jim
    Last edited by a moderator: May 5, 2017
  21. Feb 24, 2012 #20
    This forum is fantastic, and you're really lovable persons! :approve:

    So, being inspired by your good advices, I put together the latest project.

    It uses a low power Differential Amplifier (AD8276B) supplied by a Voltage Reference (ADR3425), a low bias current Op Amp (AD8603) for a feedback loop, then an external current source BJT (T1, to be defined).

    http://img861.imageshack.us/img861/8749/schematicdiagramsourcec.jpg [Broken]

    Please, have a quick look to that and let me know if makes sense.

    Later on I will write here my considerations and calculations for the component sizing, based on datasheet and app notes from Analog Devices.

    Of course, any idea, suggestions or criticisms are welcome!

    Last edited by a moderator: May 5, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook