Re: LM323-K adventure and lesson


"martin" <martin.reynolds@xxxxxxxxx> wrote in message news:1173223390.079402.260420@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
On Mar 6, 3:16 pm, A...@xxxxxxxxxxx wrote:
On Mar 6, 2:57 pm, "RonKZ650" <RonKZ...@xxxxxxx> wrote:





> I don't know why one LM323K brand would be significantly more
> tolerance than another, as this is why there are LM323AK available
> with 3% tolerance so 4.85-5.15v. As to the comment of adding a diode
> between ground and the reference pin as suggested, it does in fact do
> exactly what was described. In my test a few minutes ago, the
> regulated output rose by .67v by adding a 1N4007 diode.
> On Mar 6, 3:25?pm, c...@xxxxxxxxxxx wrote:

> > On Mar 6, 3:56 pm, A...@xxxxxxxxxxx wrote:

> > > On Mar 6, 11:31 am, "hagis_u" <ss1941...@xxxxxxxxx> wrote:

> > > > While playing some TZ on the weekend the pin went completly dead. > > > > ?I
> > > > removed the backglass and noticed the CPU board LED's were all > > > > off. ?I
> > > > metered the 5 Vdc test point on the power driver board...0 Vdc. > > > > ?I
> > > > then decided to check the output on positive terminal BR2...6
> > > > Vdc...time to replace it. ?So I thought might as well replace BR1 > > > > and
> > > > theLM323-K, 5 volt DC regulator and update these 14 year old > > > > parts.
> > > > After replacing BR1, BR2, and theLM323-K, my TZ reset when I > > > > pressed
> > > > the flippers (I replaced all the 15000 uF caps last year). ?My > > > > first
> > > > thought was "What the hell is going on now". ?I put my meter on > > > > the 5
> > > > Vdc test point and got 4.73 Vdc, and I noticed the CPU board > > > > would
> > > > reset whenever the voltage dropped below 4.72 Vdc. ?I proceeded > > > > to
> > > > check all the parts I had replaced, the connections, reseated all > > > > the
> > > > connectors and no luck the same result. ?A friend of mine who > > > > works in
> > > > the electronic components industry asked me what parts I > > > > used...he
> > > > said all looks good...who is the manufacturer of yourLM323-K? ?I
> > > > looked at the name on it "ST micro". ?He looked up the data sheet > > > > and
> > > > the spec indicated an output of 5 Vdc plus or minus 0.3 Vdc.
> > > > Depending on ?your luck the ST part will produce an output > > > > between 4.7
> > > > and 5.3 Vdc, at 4.73 Vdc I was within their spec. ?He gave me a > > > > Linear
> > > > Technologies (LT)LM323-K, whose spec is 5 Vdc plus or minus 0.1
> > > > Vdc. ?Put the new piece in and the output is at 4.91 Vdc, all is
> > > > well.

> > > > Lesson learned allLM323-K'sare not created equal...a frustrating
> > > > lesson. ? It was all worth it in the end, the TZ is working > > > > properly.

> > > Thanks for posting this.

> > > For your consideration, (not convientient in the TO-3 case styles, > > > but
> > > works well in the TO-220) you can add a 1N4004 Diode in between the
> > > reference pin and ground. (anode to reference pin, cathode to > > > ground)
> > > This will raise the output voltage .6 to .7 Volts which will > > > prevent
> > > them from continuously resetting.

> > > AL

> > Um, eh, well...
> > actually Al i do believe you are incorrect there.

> > Listen to TOPcast #7 and the suggested
> > way is described. ?The LM323 is a *variable*
> > regulator, but you don't use a diode to
> > adjust the output voltage. ?It is done using
> > resistance.

> > Again, TOPcast #7 explains > > it.http://marvin3m.com/topcast/past.php-Hidequoted text -

> > - Show quoted text -- Hide quoted text -

> - Show quoted text -

Ron,

As you're probably aware, the forward bias voltage drop of a diode is
fairly constant, regardless of current fluctuations.

I originally used this when designing bench linear power supplies.
Placing a diode in series with the output would protect it from any
overvoltage connections, but would drop the output about .7 Volts.
Placing an additional one on the reference pin would put it right back
up. This will also work with zener diodes if you need to "stretch"
the output of a fixed regulator in a pinch. (i.e. +12volts out of a
7805). You can also knock down output voltages by stringing diodes
in series.

Regards,

AL- Hide quoted text -

- Show quoted text -

The approach does work, but then the board is guaranteed to be running
over the maximum spec.


Putting diodes in series with the output will always result in hot diodes.
If the regulator is running under voltage -- then you should adjust the output voltage by installing the diode forward biased on the regulator's ground pin. There is a very small quiescent current going thru the regulator's ground pin - and there is little fluctuation with this current regardless of output load.
You can adjust the output voltage a number of ways. Easiest is to use a diode. This raises the output voltage by an equivalent of the diode's forward voltage drop. This can be a pretty high increase - as much as 1 volt depending on 'cheapness' of the diode (0.7V is typical). This diode is lightly loaded - hard to predict what the actual forward voltage drop will be... and it will usually be considerably less. Can also do this with a resistor for a more predictable result. Voltage increase (typical) = Resistor x 0.012.or Voltage increase (max) = Resistor x 0.020. A 33 ohm resistor would give you a typical 0.396 volt increase and a maximum 0.660volt increase. Aways go by worst case... Quite common to do this - even the app notes for LM323's do this. And - look at a typical older Bally board that used LM323K's -- they had voltage threshold adjustment resistors.
Too bad Williams didn't provide a place to install a voltage adjust resistor.

-- Ed

.

Relevant Pages

  • Re: Voltage Regulators
    ... And the divider voltage must be the desired ... output voltage minus the nominal difference voltage between ... through the reference pin does not distort the divider ... resistor of the reference divider, so it must be no more ...
    (sci.electronics.basics)
  • Re: Need help building a split voltage supply
    ... the symbol for a diode is a filled in black arrow ... If you put them in series, then the voltage across the resistor shown ... One other thing, a Zener diode is a special diode, which is used ...
    (sci.electronics.basics)
  • Re: Op-amp for power supply
    ... First, that diode will have a huge diffusion capacitance when conducting, it ... same is true for the transistor. ... So for the voltage control output this is bad. ... If I replace the diode with a resistor, ...
    (sci.electronics.design)
  • Re: C-multiplier again
    ... then reduce the series resistance R3 to the electrolytics. ... It turns out that low noise op amps set an absolute maximum voltage ... John Larkin raised the issue of diode leakage and added capacitance ... Changing the value of the series input resistor, R3, ...
    (sci.electronics.design)
  • Re: Need help building a split voltage supply
    ... > line on the actual diode. ... > you put voltage across it. ... > If you put them in series, then the voltage across the resistor shown ... > across it than the voltage rating of the zener, ...
    (sci.electronics.basics)