Author Topic: [Closed] Flyback transformer testing and replacement (Read 4304 times)

gkmaia · « **on:** January 05, 2019, 12:33:55 am »

Can anyone help me diagnose this transformer or suggest a replacement?

Primary coil takes 5v and the feedback coil takes another 5v. I cannot get any output voltage so I assume it is faulty.

But it outputs to a 3v power regulator and a 1.5v power regulator so I suspect at the same 5v range.

As you can see by the photos. I get no output no matter what.

It has printed on it 3305 n181 and part number is S1673305 and unfortunately I cannot source a datasheet or the part.

Ian.M · « **Reply #1 on:** January 05, 2019, 01:41:56 am »

Its a transformer - it doesn't work on continuous DC, and if you put too much DC current through it you'll burnout a winding.

You need to know what sort of driver circuit its intended for then pulse it at an appropriate frequency to check it. However its almost never the transformer that's bad in low power low voltage (in and out) SMPSU circuits. It will almost invariably be bad electrolytic capacitors, a blown semiconductor or occasionally an open-circuit resistor. If the transformer is bad, it will usually show visible signs of distress, e.g. discoloured insulation from overheating or cracked core.

gkmaia · « **Reply #2 on:** January 05, 2019, 01:49:06 am »

Transformer is T1 here on the photo. DMMV takes 5v.

DMMP on high 5v disables the circuit DMMP on low enables Q9 and takes 5v DC to the transformer.

That should output something in similar range. But it does not.

How you suggest I test this transformer. In the circuit it does not flow current.

Ian.M · « **Reply #3 on:** January 05, 2019, 06:13:05 am »

Ass you have it our of circuit, start by checking the DC resistance of all its windings to confirm there are no open circuits. Looking at the photos, from the dimensions and wire gauge one would expect resistances of tens or hundreds of ohms. Then check the transistors and the three diodes, and check R41-R43. Check the ESR and capacitance of electrolytics C24, C26, C27 (or replace if you don't have an ESR meter and cap meter). If no faults are found, put it back and measure the voltages in the part of the circuit to the right of T1 on the schematic, at each component pin and look for any discrepancies that may indicate a cracked track or dry joint. If you've got a scope, check the signal at the base of Q9 when DMMP transitions from high to low to see if its attempting t start oscillating.

gkmaia · « **Reply #4 on:** January 05, 2019, 08:58:16 pm »

I have removed Q10 and tested as a diode with my DMM. These are the results:
CB = 0.6v
EB = 0v

Ian.M · « **Reply #5 on:** January 05, 2019, 10:17:21 pm »

That means its failed B-E shorted. That is unusual for a switching transistor with large base resistors. Carry on looking as the Q9/T1 circuit should work with Q10 failed that way or removed.

Are you *SURE* you didn't mix up the two transistors?

gkmaia · « **Reply #6 on:** January 05, 2019, 11:11:11 pm »

I am sure. I removed Q9 yesterday and tested. It passed.
Removed Q10 today and tested it. It failed.
Wrapped both in tape with identification.

Ian.M · « **Reply #7 on:** January 05, 2019, 11:48:38 pm »

OK. If its also leaky C-B it may be pulling Q10 base down preventing oscillation. Try it with Q10 removed or with a good replacement fitted (type is non-critical, any silicon NPN with similar Hfe and Ic_max will do if you match the pinout).

james_s · « **Reply #8 on:** January 06, 2019, 12:38:28 am »

I'm not sure what you'd expect by feeding DC voltage into a transformer and measuring DC voltage on the secondary, of course you'll get 0V on the output. That's a high frequency transformer, you need to drive it at several kHz and the raw output will be AC.

gkmaia · « **Reply #9 on:** January 06, 2019, 01:44:51 am »

Now I am struggling to replace Q10 as it is obsolete. Originally a 2SC2812L7-TB. I have attached the closest datasheet i've found.

Thought about replacing by those. Would you be able to let me know if I choose the right replacement bellow? If not, why they are not the best ones.

I thought about these as it has closest hfe and frequency
https://www.digikey.co.nz/product-detail/en/toshiba-semiconductor-and-storage/TBC847BLM/TBC847BLMCT-ND/6198789
or
https://www.digikey.co.nz/product-detail/en/toshiba-semiconductor-and-storage/TMBT3904LM/TMBT3904LMCT-ND/5403487

Ian.M · « **Reply #10 on:** January 06, 2019, 02:22:24 am »

For Q10, frequency does't really matter. Any small signal NPN with min. h_FE>200 will do. To get you moving forward today, you can probably pull one off a random scrap board and all you need to check is that the pinout matches, its Iceo leakage is too small to read on an ordinary transistor tester, and its h_FE is high enough.

gkmaia · « **Reply #11 on:** January 06, 2019, 05:14:56 am »

Thanks Ian. I just got one more question before buying the parts. I want to get some spare Q9s.

For Q9 service manual refers as "PNP Chip 2sd1623T-TD"
The attached Datasheet the 2SD1623 is a NPN, while the PNP is the 2SB1123.
If you look at the schematic it is treated and seems to behave as an NPN.
On the tests I've made current from B+ to C- and E-, which is NPN.

So I think I the service manual may be wrong...

Ian.M · « **Reply #12 on:** January 06, 2019, 05:43:33 am »

Quote from: gkmaia on January 06, 2019, 05:14:56 am

So I think I the service manual may be wrong...

Trace enough of the actual circuit round it to confirm the schematic is correct then you can safely assume the error is in the BOM.

TheNewLab · « **Reply #13 on:** January 06, 2019, 06:01:06 am »

Quote from: Ian.M on January 05, 2019, 06:13:05 am

Ass you have it our of circuit,......
Check the ESR and capacitance of electrolytics C24, C26, C27 (or replace if you don't have an ESR meter and cap meter). If no faults are found, put it back and measure the voltages in the part of the circuit to the right of T1 on the schematic, at each component pin and look for any discrepancies that may indicate a cracked track or dry joint. If you've got a scope, check the signal at the base of Q9 when DMMP transitions from high to low to see if its attempting t start oscillating.

OOPS! typo in the first sentence. LOL

I tend to be writing post challenged..

Now I am really password retarded....my friends fall over laughing when I have to enter a password to access some site..

Too often, the site blocks me out after making so many tries

( i know, off topic)

Ian.M · « **Reply #14 on:** January 06, 2019, 06:13:47 am »

TheNewLab · « **Reply #15 on:** January 06, 2019, 06:17:14 am »

OK, This isa my domb questions day...

I thought a flyback transformer is that thing you find in older TVs that are ussually round like an automobile coil..(TV is a little smaller. That has 4,5, 0r 6 usually, round outlets coming out of the tops.

IF that is not a flyback transformer, What on earth is it???

floobydust · « **Reply #16 on:** January 06, 2019, 06:27:10 am »

That is a colour TV flyback transformer integrated with built-in rectifier diodes, and focus anode and screen voltage dividers (two potentiometers). It's all in one.

gkmaia · « **Reply #17 on:** January 06, 2019, 07:02:37 am »

Thanks Ian! If I were to change the title of my post to reflect what that transformer really is. So the title is more accurate for everyone how would you call what I call "flyback transformer"?

james_s · « **Reply #18 on:** January 06, 2019, 09:06:44 am »

"Flyback" is the topology of the power supply, a flyback transformer is one that is designed to be driven in flyback mode. While a television HV transformer is often referred to as a flyback transformer, it is not the only type.

gkmaia · « **Reply #19 on:** January 07, 2019, 10:51:15 pm »

Thanks for all the answers! I think I understand it better now.

I am now trying to understand how the current flows on the circuit and what would be the expected voltages on A and B considering I have a couple of regulators that would take up to 15v. I imagine my B should be something around 10V ac right and A around +5v DC?

floobydust · « **Reply #20 on:** January 07, 2019, 11:30:53 pm »

It would be easier to take measurements using VSS as multimeter common. Otherwise it is confusing because GND is actually going to the output of IC5.
At "A" you need enough DC voltage for the regulators - at least 2V above their output rating.
SC17710YDA (3V)
SC17710YHA (1.5V)

So at minimum 5VDC needed at "A" so "B" will go up to 6Vp (due to D6 voltage drop). I would guess 7-9VDC is used.
If you start a repair thread with the model number, more people can help.

gkmaia · « **Reply #21 on:** January 08, 2019, 12:01:28 am »

When I was drawing the current flows that looked very confusing. Why having a voltage regulator that sends 1.5v back to T1?

I will do that, just a bit afraid of people thinking that is just myself asking the same question twice. If you think that is cool i will do it. I am new here so don;t want to get people upset.

Ian.M · « **Reply #22 on:** January 08, 2019, 04:19:57 am »

Uh Oh! .
I did say at the beginning "If the transformer is bad, it will usually show visible signs of distress, e.g. discoloured insulation from overheating". In your new thread ( https://www.eevblog.com/forum/repair/finding-a-replacement-for-an-undocumented-transformer/ ) you posted this photo:

which clearly shows at least one badly burnt winding on the transformer. Chris Leyson thinks its dead, and I agree.

Options at this point, include obtaining the OEM part, maybe used off a donor board from the same device if the manufacturer cant supply spares, rewinding the burnt windings in situ, hand winding a new transformer on a suitable core (but that requires a *LOT* of very careful deconstruction and testing to establish the core characteristics and gap, and a fair degree of craftsmanship), or patching in an off-the-shelf isolated DC-DC converter.

I strongly suggest that you only maintain *ONE* forum thread for this repair going forward. Its up to you which one you wish to continue, but please post a notice and link to the other thread in the thread you want to close.

gkmaia · « **Reply #23 on:** January 08, 2019, 04:57:23 am »

Ok! how does the notice works? Is there a section in the forum where I can post it?

bsfeechannel · « **Reply #24 on:** January 08, 2019, 05:04:47 am »

Quote from: TheNewLab on January 06, 2019, 06:17:14 am

OK, This isa my domb questions day...

I thought a flyback transformer is that thing you find in older TVs that are ussually round like an automobile coil..(TV is a little smaller. That has 4,5, 0r 6 usually, round outlets coming out of the tops.

IF that is not a flyback transformer, What on earth is it???

The origin of the flyback transformer is the TVs.

It is said that they got that name because, in TVs, they are part of a circuit responsible for the horizontal deflection of the electron beam in CRTs. The inductance of their primary together with the capacitance of the so called retrace capacitor (a.k.a. safety cap) determines the fly-back pulse duration, which controls the rapid return of the blanked beam back to the left of the screen.

It is also used to generate extra high voltage and other voltages for the CRT.

When the first modern SMPS were invented, one of the topologies used by the designers was exactly the already familiar flyback circuit. The components were available from the shelf.

But I suspect that the other topologies, buck, boost, etc. were also inspired by the flyback circuit somehow.

gkmaia · « **Reply #25 on:** January 08, 2019, 06:49:39 am »

That is a NOTICE to inform this thread is CLOSED

https://www.eevblog.com/forum/repair/finding-a-replacement-for-an-undocumented-transformer/

bsfeechannel · « **Reply #26 on:** January 08, 2019, 11:38:54 pm »

I can still post to your topic. Although it is a bit too late for this thread, next time, you can use the MOVE TOPIC button at the bottom left of your page. It apparently moves the entire thread with all posts, images, quotes, etc. intact to another board and leaves the old thread locked with just a link to the new location.

It only works if it's you that created the topic (i.e. you are the OP).

See details here:

https://www.eevblog.com/forum/blog/what_s-this-move-topic-icon-all-about/

gkmaia · « **Reply #27 on:** January 09, 2019, 12:12:35 am »

That is ok. I think this thread was great to pin point the issue with Q10. And the new one has got much more focus on the Transformer side of things, which is what I wanted for it initially.

I got to tanks Ian for helping me find the Q10 issue.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: [Closed] Flyback transformer testing and replacement (Read 4304 times)

Share me