All You Need To Know About TL431 To Fix Stuff : Test in in and out of circuit, test feedback circuit

[@NorthElectronFix]

No words can express my happiness to learn directly from specialists like you Richard. Thanks once again. J

[@Dad-ij2qy]

This is one of the very best electronics videos I've seen on youtube because it shows how a zener diode, an LED, a resistor, and the TL431 3-pin precision voltage-reference I.C. work together to affect the circuit's output voltage at one specific, repeatable voltage. Kudos.

[@zyzaxxxch]

Very much appreciated. You are actually a very good lecturer and very knowledgeable. Fixed a bit of stuff but never knew about the TL431 and never thought to wonder what it was. The theory lessons reinforced by seeing a working example really helps. Thank you.

[@garypoplin4599]

30:13 - Thanks, Richard. It never occurred to me I could feed my power supply back into the output of a switch mode power supply to check the feedback circuitry! I didn’t catch that last time watching this. I learn something new from every one of your posts. I especially appreciate the _All You Need to Know_ series!

[@followthetrawler]

I've learned more about switch mode PSUs from you than I did in college. Your testing knowledge is invaluable for people bewildered by SMPS

[@adamborkowski3175]

I have just fixed power supply with 431 chip broken thanks to your video. Thanks a lot!

[@tonyedwards5782]

Thank you Richard your videos are excellent. Nothing like a hands on demonstration

[@romancharak3675]

FINALLY, I understand how a SMPS regulates the rated voltage. And, how to use a LM431 in circuits ! Thank you, Richard. That demonstration with the LEDs really clinched it for me. 🤩

[@NeilsNonsense]

What a great video. Not only all about the TL431 but also one of the best videos I've seen describing and demonstrating Zener diodes. Thank you & have a like :)

[@LearnElectronicsRepair]

Like Accepted 😁

[@TYGAMatt]

Excellent way to test the feedback circuit. I suspect that I'm not the only person here that's gonna steal that method. Thanks!

[@rongray8934]

Love how well you make these understandable to someone just getting started. I may have to watch it again, but what great teaching. Thank you

[@Gonzi-ze1sk]

You're really helping me learn a lot. I've been applying it to various SMPS and I've been able to detect failures in the feedback circuit.muchas gracias señor!

[@stever.9925]

Evidently, I've been under a rock; this device was introduced in 1977!. I've worked with zeners, LM78XX, and even LM723, but I have never heard of the TL431. I've used opamps to form window voltage detector circuits, but I was impressed to see in the data sheet that a window voltage detector circuit can also be accomplished with a couple of LM431's. Thank you for this fun introduction video.

[@justmc62]

Another great instructional video .. Thank you so much Richard for your clear explanation.. Really looking forward to your future OpAmp video 🙂

[@colinauta]

These are very clear, concise explanations and test methods. I started with electronics as a hobby in my early teenage years and studied it at university. Now many years later the veil of mystery surrounding the inner working of this area of a common SMPS has been lifted. I might even have a chance of repairing one of those in my collection of faulty units. Thanks a lot !

[@PhatTony-km3fl]

Thanks for another really useful informative post Richard

[@mr.gathuru828]

You are a good teacher, keep it up!

[@ethanclement9647]

Very nice nuts and bolts so to speak demonstration starting with basic zener diode functionality and moving up to the TL431.

[@tenlittleindians]

I'm one that always figured those were just another transistor and now I know better! Thanks.

[@alfredo8431]

This video is about as perfect as it can get

[@jayfowler4747]

Clever... i love the feedback method and will be using that next time.... thanks.....

[@d614gakadoug9]

Be warned! If you just configure the circuit with the resistors to set the behavior at DC it is very unlikely that it will work properly in a real circuit. You have to consider the entire circuit and its gain and phase characteristics or your power supply is likely to have dismal dynamic performance or worse, become a high-power oscillator. "Frequency compensation" is key.

[@t1d100]

Good information = thank you. I believe that you missed an important point, regarding the TL431's operation... You noted that its output might be somewhat effected by its tolerance rating and that is true. However, IIRC, Zeners are rather particular about current... They have a current draw range. And, their voltage output will vary based on where the current draw is, within their required current draw range. Hence, selecting their supporting resistor does require some attention. Too much current and they die. Too little current and they may not come on. With a precisely selected resistor, you will get their precise voltage output, within their tolerance range. I am working from (an old) memory, so I could be incorrect. Cheers!

[@RexxSchneider]

The concerns about overcurrent causing overheating and failure on a zener diode are real, and obviously of greater concern, the higher the zener voltage is. However, all zeners have a dynamic slope resistance, that is the change in voltage as the current changes. In addition, they have a "knee", the region where the zener is just starting to conduct which makes the voltage vary a lot as the current begins to increase. Neither of those are desirable in a voltage reference, and the TL431 improves dramatically both in its slope resistance (typically 0.15Ω) and in the abrupt change from non-conduction to conduction as the reference voltage is reached. For comparison, a 3.3V zener may have a slope resistance of as much as 100Ω,

[@apk55]

@@RexxSchneider Low voltage zener diodes also have a large temperature coefficient. As one who has designed equipment subject to wide operational temperature range in the aerospace industry (-55 to 125 deg C) this is important. Unless the zener diode has a voltage range of 6 to 7V - specially selected and run at a precise current a TL431 is far better.

[@saidbaci4344]

Very educational, well done Richard, thank you.

[@NorthElectronFix]

Thanks Richard.

[@darkdepth1991]

I never learned this bit at school, thanks so much Richard !

[@alunroberts1439]

Thanks for all this I did think I knew a lot from fixing CRT TVs amp and radios but then as a long time student in live only if you are clever you know there is so much more to learn then you find there is even more to learn.

[@stevedebeukelaer1424]

Nice video great explaing how it works and very interesting how you can test ii in circuit.

[@CliveTrezona]

Thanks Richard. I've been waiting for you to do a video on the 431.

[@Lyndalewinder]

Great video Richard - experience shows through!

[@ToltecMerc]

Lots of good practical information. I enjoyed and subscribed! Thanks!

[@ThriftyToolShed]

Excellent video. I enjoyed it very much and learned a lot more about the 431 for sure. Thanks for sharing!

[@knightwar3]

You've demonstrated 2 good ways for testing the component i like the testing device you just made I'll make one for sure

[@gadgetmind]

I have a few of these in my parts box but have never found a dead one. One day ... maybe. BTW I tend to put test clips on the LED side of the opto isolator fairly often to 1) see if the supply is running if I can't easily get to a DC output, 2) see what the output voltage should be if the board isn't marked and the board is dead. When supplies are running well (or my injected DC is as rated output voltage) you typically see 1V on the cold/LED side of the 817 isolator. I'm going to have to do more poking around on the cold side as you're doing here.

[@tonyturujlija]

You are really good man....thank you for your videos. Thanks from Sweden.

[@darrenmurphy6251]

when i was a kid a --few decades ago-- i wanted to build a dual 0-30v 3a power supply with selectable series tracking for split rails , i looked at how commercial units did it and could not get my head around the complex reference shifting and regulating the negative rail instead of the positive ect , for simplicity of built and fault finding i wanted both chanels to be symetrical, and dual gang pots dont track well and dont come in 10 turn , so i wanted voltage feedback through an led based opto isolator to slave one channel but couldnt get any linearity out of it i didnt know what the TL431 did, just knew that it was as common in the 90,s as a TL494 in a smpsu !

[@d614gakadoug9]

There are some optocouplers on the market that are specifically designed for linear applications. They have one emitter and two quite well matched "receivers." You use one receiver for the output and the other one used in a closed-loop driver circuit for the infrared emitter. The closed loop method with the good matching nulls the non-linearities. I've used them in a few circuits and they work quite well if you do the design properly. They are a lot more expensive than general purpose optocouplers. To get good performance you also need to use good op amps.

[@garypoplin4599]

12:15 - Wow! With only two discrete components and a variable power source, you’ve just created a zener identifier. Neat.

[@kiore22]

Very helpful, thank you

[@kabandajamir9844]

So nice the world's best teacher thanks

[@Dutch_off_grid_homesteading]

Heya, the tl431 is totaly new for me and in so many packaging's they come I like this tl431.very nice video richard just great

[@keithking1985]

The best TL431 video iv seen. You have Covered zener diodes plus the 431.. not to mention that little test circuit with the header pins(great idea). P. S. Iv also seen the TL431 symbol on schematics look like an SCR diode with the reference pin come out at an angle from the cathode just like the control pin on an SCR... Great video👍 best explanation 👌👏

[@LearnElectronicsRepair]

Thank you 🙂

[@d614gakadoug9]

Just recently I saw, in a video, a TL431 drawn like an SCR. Sigh! When a TL431 is being used in the feedback path of a power supply, I think it is best to draw it as two separate sub-circuits - the 2.5 volt reference and the amplifier with the output transistor. For people well-versed in this stuff it doesn't make much difference but for people who are not I think it makes the operation clearer.

[@JaenEngineering]

​@@d614gakadoug9I agree that using the wrong symbol in the schematic is annoying as hell. Drawing out the equivalent circuit though is only useful as a teaching aid or for circuit analysis in the design stage. For production and service schematics the proper only the proper symbols should be used.

[@RetroClaire]

Great explanation 🙂

[@kc1fx]

Thank you for this video. I have a 12v 4A SMPS that I use for cameras and the LED on the output would just flash in abt a 1 sec interval. I pulled the TL431 from out of the circuit and in using the A to GND and KtoR with a 1K resistor to 5v, my reading is .77 on the K pin. I should be seeing 2.5v based on your information. I'll be replacing the TL431.

[@Bibekanandaroy-tn5jm]

Many many thanks sir ❤

[@warwickbunn1250]

So nice to see the pen and paper drawings back, we've missed you 🥰 Another great video. Thanks Richard. I have a makita battery charger im trying to understand and fix. Having checked the basics, capacitors, resistors etc I'm now into areas i don't understand yet. But this video has given me fresh impetus to check the TL431 etc. so we may have success yet 😊

[@JohnCena-ez7lr]

Thank you! 🙌

[@johnlenerd]

The TL431 can oscilate during start and/or if there is a low Va/Ia going through it. Also if the capacitance of the capacitor across it is below Y or higher than X according to Texas Instruments. Higher than 10uF, I like to put at least 22uF across. A video showing this oscilation and when it happening would be very Nice. Love the videoes. Keep them comming.

[@thehobe150]

Clever repair technique!

[@jozsiolah1435]

I found a variometer in some phone and camera power supplies, that can alter the voltage of the output for the user’s needs. Probably all power supplies have the option, if not by variometers, but with a proper resistor, or socketed resistor can do the same job. Not SMD, usually easy to install. This modification is usually useless, because everything is based on the 5V standard.

[@woodrowbrimm2805]

Very informative.

[@paparoysworkshop]

At 27:00 I think the variance in your readings are caused by your fingers touching the traces of your little tester, therefore altering the readings. Coating the traces with some fingernail polish would help to insulate them.

[@RixtronixLAB]

Nice info, thanks for sharing with us, well done :)

[@weerobot]

That was Excellent...

[@andrewdewar8159]

Awesome!

[@PeterMilanovski]

Was waiting to see you place a potentiometer in the circuit to get adjustable voltage control... I need a 4.2V and a 3.6V supply, a cheap with plenty of current SMPS usually have the TL431 with it's associated resistors, I know that a potentiometer can be used to drop from 5v to 3.6v but I haven't seen it done and thus haven't gotten around to trying to figure it out myself....

[@sarathai2876]

Great video🎉

[@reg2590]

Richard, I have a simple 6 led PIR night light that I have replaced the Alkaline AA batteries with Ni- Cads. I find that one battery always discharges more than the rest . This of course will damage that battery over time. I believe a TL431 circuit will provide protection but - Do I apply that circuit to each of the 4 series connected batteries or can I monitor all 4 and shut off when a low voltage (1v) per cell is reached ? Secondly, would this make a good follow on video. Great channel - keep up the good work...

[@LearnElectronicsRepair]

Thanks for the suggestion I'll take a look at it

[@adangizomaffed4056]

thank you

[@leepovey9084]

Thanks

[@JTKhopefuls]

How much current is flowing through the LED,Zener ,resistor circuit ?

[@dreamfly555]

Why @16:01 the LED is lit at 1.955V? Also similar things @18:17. If it behaves like a Zener diode, it shouldn't conduct until 2.5V. But obviously it did conduct at less than the 2.5 reference voltage. Is the meter on the left not displaying the correct voltage?

[@_droid]

You know you can disable the sleep on that meter simply by holding the select button when you turn it on.

[@Lightrunner.]

Hi Richard 👍🖖 Nice Video and explaination . 🍹🍻

[@andymouse]

Interesting and useful but at 17:30 we are demonstrating that we can get 5 volts at the 431, because that's what you used the calculator for but that didn't happen as your circuit wasn't the same as the Texas one in that R1 should connect to the cathode, as it was you had a potential divider between VCC and ground and the REF in the middle so the ref will change with VCC and so does the output which aint the point. Other than that a lot of good takeaways !...cheers.

[@Sydney268]

Yeah I noticed that too, nice video all the same 👍

[@kriswillems5661]

I had this comment too. But read the replies to my remark. In a switching power supply this component is not used as a zener at all, but as a comparator that switches on or of the led based on a threshold voltage. It does not act as zener - it acts like a switch driven by a threshold voltage. Richard did not mention this.

[@andymouse]

yes, basically a op amp or comparator @@kriswillems5661

[@d614gakadoug9]

@@kriswillems5661 No, it is not used as a comparator and it does not switch in normal operation when used in the feedback path of a switchmode supply. It is used as a voltage reference and an error amplifier. The amplifier and the optocoupler normally operate in the "linear" region, with the amplifier increasing or decreasing the current through the IRED of the optocoupler and hence the current through the transistor of the optocoupler. It would only "switch" (swing to one or the other of the limits) under "large signal" conditions.

[@kriswillems5661]

Yes, I was not sure about this. This makes sense. I would like to see scope image of the voltage on the cathode. How such a loop works (switching or amplifying) very much depends on the time constants of the individual parts and the behavior of the feedback input of switching IC. Looking the schematic I do indeed see an Rr and Cr in the feedback, which does make it amplifier and not a comparator. Also in the datasheet of the switching IC the control input is called "Error amplifier and feedback current input pin for duty cycle control", which confirms what you say. That said, just calling it a zener in this configuration is a bit confusing. It's an error amplifier with voltage reference. Thank you. @@d614gakadoug9

[@sraros]

For those that make a lot of repairs on PSUs: How common is it for TL431 to fail?

[@afunnymonkey]

I am curious if anyone knows why he stopped using the KM601 in his videos.

[@LearnElectronicsRepair]

Don't know really, I still have it. I think it is just that I have an emotional connection to my Fluke, and nothing beats it for continuity/diode test mode. Apart from Mr Bllep maybe which has moe functionality and is even faster 😉 And the KM601 tend to chew through batteries a bit. But intertestingly enough Kaiweets just sent me a KM601S and a KM602 to show to you guys. These are rechargable via USB (I think) so let's check them out

[@Dutch-linux]

timestamp 12:30 so for that LED the voltage does not really matter it is the current? in other words you can put a LED on 20 volts as long the current wont go over the current the LED can handle ?? just asking for the ones here that may not know this .

[@jutouberundso]

Yes. LEDs definitely have a maximum voltage and it is not much higher than the voltage at which it starts conducting. For example a LED could start producing light at 3V and reach its maximum current at 3,5V. But if you restrict the current, it will not reach that voltage and you can use higher voltages to drive the LED. Most of the time a resistor in series is used, to limit the current. If the value of the resistor is high enough, the voltage drop at the diode will not be higher than the maximum voltage. If a LED with max. 3.5V and a resistor (that is not too small) in series are connected to 20V, the voltage drop across the resistor will be 16.5V or a little bit higher and the LED is safe.

[@d614gakadoug9]

LEDs are very much like any other PN junction diode, except for the light emitting part. Below a certain applied voltage, which depends on the specific design and materials used to make the diode, there is very, very little current conduction. At some applied voltage current flow begins to rise. If you continue to increase the voltage applied across the diode the current will begin to rise sharply. The curve is more or less logarithmic (the change in current through the diode is proportional to the log of the change in the voltage across it), but there are additional effects that come into play. So no, you can't apply 20 volts directly across an LED because the current would rise to something vastly too high for it to handle and it would be destroyed practically instantly. You can use a 20 volt power source but you MUST use a resistor or other means to limit the current. If you just used a resistor, most of the voltage would be dropped across the resistor - perhaps 17 volts across the resistor and 3 V across the LED for a white LED. You use that 17 volts to calculate the resistor you would require to set the LED current, simply applying Ohm's law.

[@absurdengineering]

A TL431 is not a Zener or anything like that. It is a reference voltage source and a low-voltage low-gain op-amp cleverly packaged into just three pins. You can use it as a “Zener” just as you could so use a low-voltage op-amp with a 1.5V battery connected to the non-inverting input. In a switched mode supply, TL431 is THE voltage error amplifier and has a key impact on the performance of the voltage regulator aspect of the supply. It finds use in switch mode supplies because it’s cheap and fairly well understood among the power supply designers. It does the job well enough. When thinking of TL431, think like this: op-amp with (+) input tied to a 2.5V reference, (-) input brought out to the R(eference) pin, output driving a current source connected between positive and negative supply rail, and those rails then brought out as A(node) and C(athode) pins. It’s a compromise, but a reasonable compromise. And it works like any other low bandwidth op-amp would. You can use it to amplify speech audio for example (it’s too slow for HiFi lol).

[@kriswillems5661]

Exactly. But Richard made a mistake. R1 from that feedback loop should be tried to the output of the opamp (thus the cathode) and not Vcc as Richard did. R1 and R2 make a voltage divider without load on it, because the reference input has a very high input inpedance, because it is an opamp input.

[@RexxSchneider]

But note that the TL431 behaves exactly like a zener diode when the anode is taken positive relative to its cathode, i.e. it conducts at about 0.6V, allowing it to be a direct replacement for a zener in any application. An opamp is rather unlikely to behave as cleanly when its supply pins are swapped.

[@RexxSchneider]

@@kriswillems5661 It depends on the application. For a shunt regulator/fixed voltage reference, R1 and R2 should indeed make a voltage divider between anode and cathode. However, as part of the feedback loop of a regulated power supply, you do need the top end of R1 to be connected to the PSU output, so that the TL431 runs open-loop and switches on or off like a comparator, as the output voltage varies about its set value (making the voltage at the Ref pin vary close to 2.5V).

[@kriswillems5661]

​@@RexxSchneiderThat makes sense.

[@LearnElectronicsRepair]

I accept that TL431 is not a Zener diode (internally) it is more like an op-amp with a voltage reference and this is clearly visible on the Texas Instruments datasheet. But I believe the point is that it *behaves* like a zener diode in real life applications and that, as the title suggests, is really 'All You Need To Know To Fix Stuff'

[@williamjohnson6592]

Why would you use white for positive lead. Is this confusing?

[@LearnElectronicsRepair]

Those were just some old wires I salvaged out of a bit of old scrap CAT5 LAN cable so the colours are what they are. But if you are fixing stuff you should get used to this sort of thing. For example on motherboards, sometimes the silk screen shows a solid white area for the negative ends of electrolytic capacitors and other times (different manufacturer maybe) it shows a solid white are for the positive end of the capacitor! The moral of this is *never trust the colour of the wire to indicate the polarity!!!*

[@kriswillems5661]

You use the component a bit in a weird way. Normally R1 is connected to the cathode, not to VCC. If you connect R1 to the power supply (VCC) the reference voltage increases when the voltage of your power supply increases, and thus the zener voltage will vary with the voltage of the power supply.....which is not what you want if you want it to act as a voltage reference.

[@kriswillems5661]

Normally there's a third resistor R3, that is essential.....R3 is in the supply line. And R1 is connected to the cathode, not Vcc.

[@LearnElectronicsRepair]

I connect it exactly as shown on the Texas Instruments datasheet www.ti.com/lit/ds/symlink/tl431.pdf though I do agree when used with an optoisolator there is usually another resistor that connects to the cathode of TL431 to ensure it has Vcc before the LED turns on @kriswillems5661

[@kriswillems5661]

The application schematic is on page 25 of the datasheet. Do you see Rsup and do you that R1 is connected to the cathode? In your schematic Rsup is the LED and its series resistor.

[@kriswillems5661]

Would you mind telling me which page of the data sheet you used? I honestly don't understand the way you use it.

[@kriswillems5661]

Maybe I can prove it to you this way : in your schematic increase the supply voltage and measure the zener voltage when you do so. The zener voltage will go up if the supply voltage goes up. Now connect the side of R1 that is connected to Vcc to the cathode of the zener and LED instead (both cathodes are tied together). Now, when you increase the supply voltage the zener voltage will not go up. You can do the measurement. (Your led and series resistor take over the function of Rsup)

[@ivolol]

They DO NOT work the same either way around. Their sharp zener cutoff only occurs with positive voltage connected to the cathode. This is why you are getting a random-ish value one way around (2.6V) and an exact 2.49-2.5V the other. They're made of a bunch of BJT transistors and (zener) diodes in the package, and you can bet those don't work 'either way around' either.

[@LearnElectronicsRepair]

*Sorry but you are wrong* When I was testing the TL431 I have the Ref and the Cathode connected directly to each other to give a 2.5V reference. That is the two outside pins. The Anode is the center pin, yeah? So whichever way round I put the TL431on the tester, the 1K resistor from my bench PSU is always connecting to the C and Ref pins and the ground is always connecting to the Anode 😉 @ivolol

[@ABehrooz]

​@@LearnElectronicsRepair I've never worked with the TL431 but the internal schematics for LM236 and LM336 shows a lot of transistors inside. I guess the Cunningham's Law is in effect.

[@raytech6049]

Precision variable voltage zener.😂😂

[@user-rs8zg8ey2b]

Zenera are semi horible on the stability of its output voltage over load, just look at the curves, the 431 is much better.

[@samueldaniel7]

Thanks