TL431 + LM317 + PNP = linear question

The 317 current source loads the 431, however like indicated 100nF bypass can cause instability in the 431, much better to go with 10-22uF and put a diode over adjust-output.

Attached the stability chart for the TL431
 

Attachments

  • TL431 problems.PNG
    TL431 problems.PNG
    26.5 KB · Views: 80

TL431 data sheet, product information and support | TI.com

https://www.ti.com › product › TL431





TI's TL431 is a Adjustable precision shunt regulator. Find parameters, ordering and quality information.

You seem to have it backwards, the 431 controls the 317.
Look at the data sheets, then decide.

317 is a variable voltage regulator. It needs to have an external input / reference to control its output.
I know the datasheets well.. Plus the application sheet from TI covering the instability region. Also I have looked thru the current On semi sheets and even older ones from places like Fairchild..

Which, I think most anyone using a TL431 knows well.. This linear ONLY displays the instability with no load on it. From 75mA up its perfectly stable but has the 3mv pk-pk noise that apruplty starts the moment the sawtooth instability waveform goes away. See vid i posted.

So. The instability seems to be quenched with any normal load for this supply.

I would not say it was unstable. It is. It is a bit noisy, but not horrible so. The noise is also wideband rather then harmonicly nasty. Regulation is reasonable for its current range.

This circuit does work..

The question is why use these parts in this way. There are seemingly better circuits using fewer parts and lower cost.

So why this circuit ?

Can anyone tell me why they used such a high resistance on the divider for the TL431 adj ? Thats higher then any datasheet or any application I can find. It gives the TL431 gain. WHy did they do this ?

Why no Neg feedback ? Would have been easy to do to gain better reg. No extra cost really. It seems it was a mindful choice. Can feedback degrade some performance parameter ?

Why use a LM317 + NJW0302 ? They could have used say a power FET or darlington with the TL431 as shown a lot all over the place. WHy use a LM317 + NJW0302

This linear supply SEEMS to have picked all these things for a reason. I don't know what that is and that is why I am posting.

I am now posting on some engineering forums hoping to find someone who can explain some of the above questions. A good start would be why push the TL431 towards gain. Why nor use feedback. Why use a LM317.
The 317 current source loads the 431
A resistor could do this right ? Why use the LM317 ?
 
TL431 is a 3 Rupee part here, 317 is 10 Rupees, in no name quality.
That is dirt cheap, particularly in large quantities, as these are retail prices.
That is a reason for the use of those parts, maybe a Zener design could have been used.

Approximately 75 Rupees buys you 1 US Dollar at today's prices.