Voltlog #72 - DIY AC-DC Off-Line Converter With TNY276/TNY277 - part 1

[@Ivo--]

I appreciate when youtubers show their failures and not only their successes. My hat's off to you sir!

[@voltlog]

thank you for watching!

[@nor4277]

I like that it had a fuse and that it was spaced well, when component are to close it makes soldering much harder.

[@voltlog]

Thanks! An update video has been posted here kzbin.info/www/bejne/aqCnZGyXgqqkbsk

[@CRITRON]

The additional winding is used for the no load situation to reduce the power disipation of the swiching device. Also the parallel cap 560pF is acting more like a load than a transient improvement of the shotky... it can be removed. I would be fun conecting the RIGOL all 4 channels in various points of the circuit and then compare with the reference signals in the datasheet. The datasheet has a typical application of the IC, so compare and see what is missing to your design... good luck

[@station240]

Only thought I had, is to add a small capacitor between enable (EN) and source (S) to smooth out the feedback a little. Possible it's switching too much with no load. Usually these small PSUs have an led as a minimal load. I agree on trying the RC snubber in the datasheet, these are a lot more complicated than they first appear.

[@voltlog]

between having the coil start/end reversed and missing the rc snubber which one you think would cause the circuit to over-heat? I have followed recommended design from pi expert suite, which I think removes un-necessary filtering for low power designs.

[@josipzlk8019]

VoltLog current snubber is good enouth but it is nicer to use better snubber its just nice practice.There is nothing wrong with snubber.Also he was talking about cap that smooth power supply to the chip,ussualy you have one more winding to power the chip but this chip is diferent.I am not familiar with this chip and dont have time to research it right now.

[@poorchava]

I would check transformer winding polarity first. If that doesn't help I'd connect some 1R resistor in series with primary and try to observe voltage across it with scope in order to determine current. Perhaps the transformer primary has too low of an inductance and the core gets saturated. This could happen if for example the core was assembled poorly or the transformer you got is for 120VAC countries.

[@blackbrayn]

+1 - what he said!

[@voltlog]

An update video has been posted here kzbin.info/www/bejne/aqCnZGyXgqqkbsk

[@josipzlk8019]

coil polarity is shown with that dot on schematic and it does matter very much.

[@voltlog]

Do you think that if I had the coil polarity the other way around, the circuit would behave like it does?

[@josipzlk8019]

VoltLog my theory is that chip is overloaded from one transformer windings being conected backwards.It can output right voltage if conected backwards but almost no power so it strugles to keep voltage 5v. reversing windings shoud help if there is nothing else horibly wrong.

[@mrlazda]

You mess up polarity of coupled inductor (flyback transformer). You made one transistor forward converter, but without way to reset core, and core “walk−up” to saturation, which results in excessive current.

[@voltlog]

An update video has been posted here kzbin.info/www/bejne/aqCnZGyXgqqkbsk

[@uK8cvPAq]

With the coils phased the wrong way you could end up with a forward converter rather than a flyback converter.

[@voltlog]

An update video has been posted here kzbin.info/www/bejne/aqCnZGyXgqqkbsk

[@blackbrayn]

HI , is it posible that the windings on the primary side are too low in resistance overdriving the internal mosfet? And yes i remember reading something about the "polarity" too.

[@voltlog]

That's a logical question to ask, and it's the first thing that came to my mind too but that particular transformer is designed to be used with the TNY276 among others from that series. I didn't just pick it for no good reason. So I guess I'll try reversing the "polarity" and see how that goes..

[@BurmeseExplorer]

Hello Sir, I found my computer power supply got broken TNY276pn. I couldn’t find replacement IC in local market, is there any way to bypass this? Thanks.

[@TheDefpom]

To me it looks like the transformer is for EITHER the tny267 OR tny277, not a range between the two.

[@unnyash1433]

Can I use tny266 instead of tny264gn?plz help

[@berkcan2439]

Where can i find this kind of transformers ?

[@orides5976]

Great video! What optocoupler did you use?

[@voltlog]

I don't remember the part, it was either some common part or something recommended in the converter chip datasheet.

[@orides5976]

@@voltlog Thank you. I managed to read the part number in the video. It was a COSMO 1010 817. Pi expert software recommends a PS2501-1-K-A for this kind of PSU, but looks like the 817 opto worked fine for you.

[@voltlog]

@@orides5976 yeah, the 817 is a pretty generic optocoupler, which I probably had in my inventory.

[@battlecoder]

I'm definitely not an expert, but I see that the suggested schematic recommended 4.7M resistors for R1 and R2 and you used 2M Ohm. Could that be the cause?

[@voltlog]

thank you for the suggestion, that was not the issue. An update video has been posted here kzbin.info/www/bejne/aqCnZGyXgqqkbsk

[@alimeshkatifar6640]

it is very very good

[@trottolo90]

Hi...can i replace tny278 with tny276 ?

[@johnmij5828]

According to your pcb layout diode D4 is a 1N5819, this is a schottky diode and can't be used because the blocking voltage is way too low, you need to replace it with a high speed, high voltage diode, like the FR106.

[@voltlog]

+John Mij on the actual board I populated with FR106 ultra fast diode. Sorry for that..

[@sergokhuraza9503]

Очень хорошая дикция, я слегка знающий англиский всё понял

[@JoseCarlos-ep3yd]

Hello! How to test a tny277pn?

[@avejst]

An ESD mat please, I am afraid that you destroy your components by ESD, static discharge, so that you intend alter your circuits

[@voltlog]

what do you mean? I am using an ESD mat, and that black foam is ESD safe.

[@avejst]

Oh, sorry, It look like ordinary non ESD safe foram.

[@JyrkiKoivisto]

Maybe you should implement the rc-snubber circuit in it? R2 and C4 in figure 14 from the datasheet (www.power.com/sites/default/files/product-docs/tny274-280.pdf)

[@josipzlk8019]

Jyrki Koivisto or cap across zener too

[@voltlog]

It is my understanding that additional filtering is to be added either if you increase the output power or if you have tighter design constraints: for example lower output noise but in this case, I don't think the rc-snubber circuit could prevent this circuit from working, please correct me if I'm wrong.

[@JyrkiKoivisto]

Whenever a inductive load is being switched on and off it generates large spikes (which I'm sure you know). Your diode clamp might not be fast enough to clamp all the voltages. This is just my guess.

[@josipzlk8019]

Jyrki Koivisto well it must be decent fast diode and cap across zener helps with that too

[@voltlog]

The diode is FR106, decent 250ns recovery time. I will first try reversing the polarity on the transformer and see how that goes, if it doesn't work I'll start exploring the other suggestions. Thanks.

[@samwikiwiki8274]

Salute dich

[@FLAVIOBOIAGO]

tem dublado?