Connect with us

Placement of Ammeter in Active LM338 Regulator

Discussion in 'Power Electronics' started by Electromotive, May 20, 2020.

Scroll to continue with content
  1. Electromotive


    Feb 14, 2018
    Just having built two 0-13V regulated power supplies and having excellent results with the LM338 regulator, I need to upgrade two older units built with the LM317 in order to get more than 1 amp of output.

    I have a question on the proper placement of the Ammeter. I understand that the ammeter could be on the input or output of a component if the component was passive. However, I need to know if the two circuits shown, A and B, are the same or different.
    Thank you...
  2. Nanren888


    Nov 8, 2015
    (1) Putting the ammeter after the voltage regulation point introduces a current-dependent voltage drop. If you want to monitor the output voltage, then you probably want the voltmeter on the output, that is, after the ammeter. This may apply to the voltage sense of the regulator too so it sense the output voltage, not before the ammeter.
    (2) Is the meter marked "A" a voltmeter? Are you wanting to use a voltmeter as an ammeter by measuring the voltage across a shunt?
    If so it will see Ohms law is the way to calculate the voltage or shunt value as you have shown.
    Any linear regulator will require some headroom, that is voltage drop across it. The input will need to be greater than the output voltage.
    The LM338 data shows a "input-to-output voltage differential" with a minimum of 3 Volts. This is the quoted minimum for operation.
    That is, if you supply 12 volts, the output will only go up to something like 12-3 = 9 volts.
    davenn likes this.
  3. duke37


    Jan 9, 2011
    The two circuits are slightly different.
    A measures the current taken by the load and by the LM338
    B measures the load current only. The volmeter could be connected to the output for more accuracy.

    The sense resistors will probably only be passing 5mA or so, they are not specified.
  4. davenn

    davenn Moderator

    Sep 5, 2009
    ^^ that is the really important bit ^^
  5. Bluejets


    Oct 5, 2014
    Ammeter shunt is rather large and affects the circuit as well.
    Would be normal to have a rather low value shunt (0.1R) and metering to match.
  6. duke37


    Jan 9, 2011
    It would be nice to know what the current will be and what accuracy you want.
    A better circuit would be the B circuit with the voltage sense connected to the output. You should look up the requirements for stabilising capacitors.
  7. Electromotive


    Feb 14, 2018
    Gentlemen. Thank you for responding...

    - I have 4 home built regulated power supplies. Two were recently completed with the LM338 fed by a 4 amp 12vac transformers.. Running engines on an 0-scale model railroad, even multiple headed, the maximun current read 1.5 amps with the Ammeter at the Vout line nearly last, just before my reversing switch. The ammeter is as purchased and the voltmeter is connected just upstream of the Ammeter. The maximum voltage at full throttle is read as 13 volts by the voltmeter. Applying pressure to a running 12v motor prior to stalling gave me 4 amps max. I am extremely pleased with these circuits.

    AS SHOWN BY "A" - The other two home built older regulated power supplies using the LM317 are fed by 12vac 2 amp transformers. The ammeter was a voltmeter converted via ohm's law to read current. These power supplies run the engines a little anemic and only draw 1 amp. The voltage max. is higher than the 13v by maybe a few additional volts. I am trying to improve the outputs of these two units to closer to the outputs of the "B" units recently completed.\

    Would amp output and running improve by simply moving a real ammeter to the downstream of my "A" circuits?

    I am trying to duplicate with the "A" LM317 units the better performance of the "B" LM338 units by rebuilding, using LM338 circuitry.

    According to these excellent comments, I may have 3 issues....

    1 - Replace the Ohms law Ammeter with a Factory ammeter...

    2 - Relocate the ammeter to downstream of the LM338, as in "B"...

    3 - For both power supplies, place the voltmeter downstream and after the ammeter...

    Does it look like I have properly analyzed my issues based on your good comments?

    ps - The picture is the DC output of one of the LM338 regulated power supplies. IMG_20200505_150640288.jpg
  8. duke37


    Jan 9, 2011
    For this application you do not need any accuracy of better than 5%. I do not think that moving the circuit parts around will affect tne accuracy greatly.
    The LM317 has a limit of 1A but can be increased with a pnp boost transistor. Alternatively change to a 3A LM338.
  9. bertus

    bertus Moderator

    Nov 8, 2019

    Here is such PNP boost circuit that @duke37 mentioned:


    In the drawings the LM78XX is given, but it will work just fine for the LM317.

    When you want a single regulator with more current capability, have a liik at the LT1083, LT1084 or LT 1085:



    Attached Files:

    narkeleptk likes this.
  10. Hunter64


    Nov 20, 2018
    With the PNP-boost solution your current protection is gone.

    With a 4A transformer a LM350 is a better solution. Shutdown at 3A.
  11. bertus

    bertus Moderator

    Nov 8, 2019

    @Hunter64 , the second schematic I posted shows how to implement current limiting.

Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day