Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Challenge 3: Find the circuit flaw, or not ?

Status
Not open for further replies.

MrAl

Well-Known Member
Most Helpful Member
Hello again,

This might be the most difficult problem to answer, but when you dont have the answer yet i guess that's the way many problems can seem. I dont have the answer for this one either yet so lets see if anyone can come up with something interesting.

In the diagram the circuit is a very simple one LM317 circuit set up as a 1.25v regulator. The input voltage range is about 5v up to about 25v, so it's not a dead set value this time, but can be varied. That's to allow various input voltages to power the circuit, and since it only has to put out 1.25v (or close to that) it should be able to work down to 5v input, and i am pretty sure it does.

So the question is, can you find the circuit flaw in this circuit?

If i were to take a casual look at this circuit i would say, "There is no flaw really". But as luck would have it, that's not correct, and as of yet, i still dont have the answer.

If you would like to find the flaw yourself, then dont scroll down. If you want to see what the flaw is, then scroll down. I advise you to scroll down though because this one is entirely different than the other problems where looking at the schematic might reveal the answer. Looking at this circuit doesnt seem to help unless you've already encountered this problem in the past.




Scroll down...




The problem:

When the input is at 5v, the output is 1.25v (or close to that) and relatively stable. When the input voltage is raised up to around 7v, still stable and at 1.25v as expected. But once we get to around 10v or so (may be a little higher as per package) the output starts to drop down until it reaches maybe 1.22v.
Note that if we add a bottom resistor of 120 ohms the output goes to 2.500v (or close) and stays there pretty solid over 7 to 25v input.

This 1.25 to 1.22 volts decrease might not seem like much, but that's around 3 percent which is much more than the data sheet indicates, and i have not tested to see if it is stable with load.

So the real question is, does anyone have any information on why the voltage might drop when the regulator is used as a 1.25v reference?
 

Attachments

  • RegulatorProblems-03.gif
    RegulatorProblems-03.gif
    5.1 KB · Views: 174
Add a scope to the output while you are raising the input voltage.

I assume that, up to now, you have monitored the output voltage only with a DMM. If that is the case, the "voltage drop" seen in your DMM may in reality be oscillation.
You may need to add output capacitance for stability.
 
I attached a circuit diagram of an experiment card. At the diagram you should focus into the red rectangle. İf you wnat to keep stable the output of your circuit you can use D5 and C7 together or just C5 as schmitt trigger said. But, output stabilization is not one adventage of using D5 and C7 together, using of the D5 and C7 prevent your LM317 in an unexpected stuation such as short ciruit on load. That's why, I advice you to use D5 and C7.
 

Attachments

  • LM317.png
    LM317.png
    43.5 KB · Views: 158
Spice doesn't show any output voltage drop at all (even with various parasitic resistances in circuit; rather, it predicts a slight output rise as the input voltage rises. But that may just be an artifact of the sim model.
 
Do you have the required input and output stabilization caps? I have seen a similar problem in testing those devices because the caps weren't there and it was oscillating.
 
Hello guys,

I finally got to testing again this morning, and you all got it right...it is the lack of capacitance that causes oscillation and then the output looks like it drops using just a DC meter. I did not experiment with a lot of values just a 2uf cap on the output, and the oscillation went away completely. It could be just a 0.1uf cap would be enough, but at least we know for sure what caused this problem now.

So it seems very stable above maybe 2v even without caps, but without caps it starts to oscillator below that.

So this circuit flaw is fixed too now :)

And yes, the circuit simulation does not show this behavior. It stays constant right down to 1.25v even without a single cap.
 
Hello guys,


And yes, the circuit simulation does not show this behavior. It stays constant right down to 1.25v even without a single cap.

I had the good fortune of attending one of the seminars given by the late Bob Pease. He would always say that the best "simulator" would be a breadboard and scope.
As a fact, in his excellent book "Troubleshooting Analog Circuits", he included a pair of hilarious pictures of what he likes to do with Spice simulations.
 
.........................................
So it seems very stable above maybe 2v even without caps, but without caps it starts to oscillator below that.

So this circuit flaw is fixed too now :)
So the "flaw" was not connecting the decoupling caps as the data sheet states. ;)
 
Hi,

Schmitt trigger:
Yes, and that sounds interesting. Maybe you can relay the joke somehow.

Carl:
Yes, that was the only 'flaw'. So some circuits absolutely must have the caps. Battery chargers on the other hand dont seem to care as the voltage is always higher like 3v or more, and then the battery might act as a capacitor anyway. Never tried it lower than about 2.5v though with a Li-ion cell, no problems there.
 
I had the good fortune of attending one of the seminars given by the late Bob Pease. He would always say that the best "simulator" would be a breadboard and scope.
As a fact, in his excellent book "Troubleshooting Analog Circuits", he included a pair of hilarious pictures of what he likes to do with Spice simulations.

Hi schmitt trigger

Don't get me going about Mr Pease. A living legend. I would of loved to have met him. He knew Analogue inside out.

And he was humble too: https://en.wikipedia.org/wiki/Bob_Pease

No airs and graces. Died in his 1969 VW Beetle after attending his fellow expert's funeral. Namely Jim Williams....what a sad day that was for Analogue :arghh:

Regards,
tvtech
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top