LM2937ET-3.3 voltage regulator with 5V output??

Moskog

I needed 3.3V for a Propeller project and found an LM2937ET-3.3 that was bought from Parallax some time ago. The regulator was fed from an LM7805. After soldering and testing the PCB the output on pin3 was still 4.99V, not 3.3 as I expected. OK, I thought the smoke has left the regulator, even though I never saw that happend. -And I expected the regulator to be very hard to kill. Then I removed it from the PCB and put 6-7 volts from the lab-supply into pin 1 and 2. Still 4.9 volts on output pin. Did the same to an another unused LM2937ET-3.3 and still the output was 4.9 volts. Without any load. Have I got some 5V regulators with wrong marking? Could that be possible?

Phil Pilgrim (PhiPi)

That regulator is very fussy about its external caps. The minimum output capacitance is 10uF for stability, and the ESR has to meet strict requirements. Did you read the datasheet and follow these requirements, or just tack a 100uF aluminum electrolytic on the output and hope for the best?

-Phil

Moskog

Phil Pilgrim (PhiPi) wrote: »

..Did you read the datasheet and follow these requirements, or just tack a 100uF aluminum electrolytic on the output and hope for the best?

-Phil

Hmm, I think you got me there, been too used to the LM78.. family. No time to read the datasheet or at least no time to read it carefully. Back to the work bench, thank's so much, Phil, I'll check it out.

Duane C. Johnson

Hi Moskog;

As others have said this regulator is a bit fussy.

Reading the spec for the LM2937 carefully you will find it has a minimum output current requirement of 5mA.
For the 3.3V version a small 630Ω load resistor should be added. (3.3V / 5mA = 660Ω)

Duane J

erco

Phil Pilgrim (PhiPi) wrote: »

Did you ... just tack a 100uF aluminum electrolytic on the output and hope for the best?

PhiPi very eloquently summed up the quiddity of my electronics knowledge with a single stroke of his pen.

Phil Pilgrim (PhiPi)

And erco, with even more eloquence, sent me scrambling to my dictionary with quiddity! I really hope that word shows up in a NYTimes crossword soon, so I can thank you for it, good sir!

-Phil

Moskog

Ok, replaced the capacitor with a tartar electrolytic and added a 560 ohm resistor and the voltage dropped to 3,3 V.

Duane C. Johnson

Hi Moskog;

Moskog wrote: »

Ok, replaced the capacitor with a tartar electrolytic and added a 560 ohm resistor and the voltage dropped to 3,3 V.

Just curious, which did the trick? Capacitor or load resistor?
Also, what is a tartar electrolytic? Did you mean tantalum capacitor?

Duane J

Moskog

Duane C. Johnson wrote: »

Hi Moskog;
Just curious, which did the trick? Capacitor or load resistor?
Also, what is a tartar electrolytic? Did you mean tantalum capacitor?

Duane J

Thanks for asking, Duane. First I replaced the capacitor but still almost 5V , then I removed the Propeller chip, because I was afraid I overpowered it. Whith the Prop removed the voltage drop to 3.3 of some unknown reason. (Back to 5V as soon as the Prop was reconnected) So I found a 560 ohm resistor, (closest I found to 660 ohm) and that did the trick. Now the output is 3.3 V, with or without the Propeller. The tantalium (tartar) is 100uF.

Phil Pilgrim (PhiPi)

Moskog,

It sounds like you have 5V connected to one of the Prop I/O pins that was overpowering the 3.3V bus through a protection diode. Removing the Prop removed that connection. Adding the resistor lowered the impedance that the apparently higher-impedance parasitic 5V had to supply, which swamped it, letting the regulator do its job.

I would definitely check your connections to make sure you're not feeding too much current into one or more Prop I/O pins.

-Phil

kwinn

I would suspect that propeller chip, or at least that pin with the 5V on it is toast.

Duane C. Johnson

Hi Phil;

Phil Pilgrim (PhiPi) wrote: »

Moskog,

It sounds like you have 5V connected to one of the Prop I/O pins that was overpowering the 3.3V bus through a protection diode. Removing the Prop removed that connection. Adding the resistor lowered the impedance that the apparently higher-impedance parasitic 5V had to supply, which swamped it, letting the regulator do its job.

I would definitely check your connections to make sure you're not feeding too much current into one or more Prop I/O pins.

-Phil

I agree, a pullup resistor to 5V would do this.
I would have thought the voltage would be more like 4.3V to 4.4V or so due to the .6V protection diode drop.
Maybe one of the protection diodes has is weakly shorted which would bring the voltage up to 5V.

The 5mA load, 660Ω, minimum is mainly for stability and better regulation.
I don't have the LM2937 but I have used other low drop out regulators and they at least worked down to no load.

Anyway, I think you got the minutia of the problem right.

Duane J

Phil Pilgrim (PhiPi)

The 0.6V drop only comes into play after a threshold current. At very low currents, the voltage drop can be insignificant, based on the diode's I/V curve.

-Phil