Attempting repair of TV tuner - Emitor Megalook

[Bicurico]

Hi,

I have two Emitor Megalook field meter. One has a broken TV tuner and thus cannot render analog TV channels. Everything else (spectrum, analogue SAT, DVB-S, DVB-C and DVB-T) works fine, so this is not the end of the world. Still, I would be happier if I could repair this fault.

I am pretty sure it is the TV tuner board, used for showing analog CATV channels that has an issue. This is the "TV LOOK" board. The reason I don't just swap this board with the one from the working Megalook is that there are a few cables soldered, so I cannot extract the "TV LOOK" board without dessoldering them. Naturally I don't want to risk ending up with two broken units, so the good one is a "don't touch" device, regarding this repair.

I checked the board and neither cables going to or from the board are defect nor is there any visible damage.

When activating this analog TV function, the screen just goes white with some flicker, but no sound or picture/noise.

I photographed the board as good as possible and searched datasheets for all major IC's. The complete archive can be downloaded here: http://www.mediafire.com/file/3d9n27r7pppalgs/Emitor_Megalook_-_TV_Look_Repair.zip

I would like to have some opinion on what would be the best way to start finding the culprid.

Attached is a picture with some notes and the (?) indicate the IC's I would like to test - but I don't know how.

I don't have any spare parts, so I would need to order whatever IC I believe is broken.

Any help is appreciated.

Thanks,
Vitor

[BigBoss]

I think the pins which you have marked as "Connector from CPU Board with Settings" are just the pins of a SAW filter.Because after MCL633 Transformer, the signal is redirected to TDA9880T ( IF Demodulator +Sound IC ).
ACD0900 is a TV Tuner IC that is controlled by a PLL IC ( I cannot see the part number) that is marked by [?]
There is also a 4MHz Crystal for TDA9880T ( left hand side )
I think there is an IF amplifier besides to Xtal ( SOIC-8 case) to amplify the IF signal to drive the Demodulator IC.

If you have a good oscilloscope with a high impedance probe, you can track the signal through the signal path.

[Bicurico]

I have a Rigol DS1054Z (hacked) with the shipped probes, but have not used it much.

I am not experienced on how to use it for fault finding...

This is what I am trying to learn here, so bear with me.

Thanks,
Vitor

[BigBoss]

I have a Rigol DS1054Z (hacked) with the shipped probes, but have not used it much.
I am not experienced on how to use it for fault finding...
This is what I am trying to learn here, so bear with me.
Thanks,
Vitor

First thing first, you should download the datasheet of the components in order to get the connections and technical specifications.Then draw a block diagram by investigating the PCB and obtain roughly main signal paths and supply connections.After that check first the supply voltages and if it's possible the currents by referred to datasheets.If no abnormalities are observed on DC bias, you can track the signal path mentioned in the datasheet.

[Bicurico]

That is what I kind of have been doing...

I made a block diagram of the Megalook, showing how each PCB is connected and determining its function.

This is how I got to be pretty sure that it is the "TV Look" PCB, that has a problem.

Also, I downloaded all relevant datasheets.

So now I would do a diagram showing how each IC is interconnected and then check the supply voltages.

I was asking to understand how experienced people would do it. I see now, how time consuming it can be!

It will take me a long while, as I am very busy now. But I am sure that the repair will be some cheap components at the end...

Main thing is to not break anything in the process.

Regards,
Vitor

[Bicurico]

Hi,

Today I had the time and nerve to take a further look into this device.

At the end, I got it to work properly , however I the fix is not permanent and I don't understand the fix at all.

I have a second unit and I connected a composite video cable to the EXTERNAL VIDEO INPUT of the second device. The purpose was to touch different points on device 1 (broken) to see if there was a video signal present.

By accident I touched the VIDEO OUTPUT of the TDA6880T and suddenly I got video. I touched the AUDIO OUTPUT of the TDA6880T and after a few times touching, I got sound as well.

The fix did not last very long - after running the device for 10 minutes it reverted back to the old state.

By touching the two pins with the video cable, I can get it back to work.

Also, if I touch the XTAL pin with my multimeter probe, I will get VIDEO as long as I keep touching it.

All of this does not work always, one has to keep trying.

The fix will survive a power off, removal of battery and power supply, but after like 30 minutes it is gone and the old state is back.

This proves that the TVLOOK board is not totally damaged and that all other boards are OK.

What I don't understand is what is wrong?

Can this be a strange damage of the TDA6880T?

Broken XTAL?

Any thoughts?

Attached is a picture of the relevant section of the board, the two black arrows show where I had to touch. The video cable had like +0.2mV on the tip - a different potenial as it comes from the other device's EXTERNAL VIDEO INPUT connector.

Thanks,
Vitor

[Rasz]

magic finger repair suggests:
-intermittent connection
-loss of capacitance

your fat fingering fix survives power off so capacitance is off the table, means most likely you have cracked joint somewhere

[Bicurico]

Yep - bad solder point!



The device is to be considered repaired!

I am not sure what pin had a pad solder point, as we are talking SMD here.

So in the end I took a soldering iron with small tip and resoldered most of the TDA9880T pins and a few other suspect looking ones. To my surprise, the device started working correctly!

Regards,
Vitor

[Bicurico]

arrrgggghhhhh!!!! 

After 1h without power, it is not working again.

WTF!!!

What on earth is wrong with this.

Intermittent problems are the worst.

Is there any recommended proceedure?

Regards,
Vitor

[Rasz]

another shotgun debugging method is a hair drier on running system, see if that fixes things. This could mean hairline crack somewhere (maybe even one of the components). also try pushing components with a plastic stick instead of finger = removes  capacitance from the equation

or download TDA9880T datasheet and do it hard way, checking if 4MHz oscillator is running (scope HighZ x10), verify all the small caps are ok etc

[Bicurico]

I did all you suggested before.

Now the TV mode works if I select D/K audio instead of B/G - of course I don't get any sound.

This is still better than before, but it leaves me to the conculsion that the resoldering of pins is the way to go. I may have missed one last faulty pin.

Right now I am leaving the device turned on for a while, to see if heat makes any difference.

Anyway, this is a great lesson in cracks and bad solder points.

Regards,
Vitor

[Bicurico]

The saga continues...

I resoldered some further pins and went through the TDA9880T ones.

It is working OK again. Let's see how long it lasts this time - my hopes are high, as it seems now that the issue since the last resoldering was probably one joint that was leftover and did the sound selection.

Actually I am quite surprised about this: on one hand fortunately there was no broken component, on the other hand having such kind of intermittent issue is a real PITA.

You think it is OK, then it fails, then you get it back working, it fails again,...

And nothing makes really sense.

Thanks for all input and hopefully this was the last time I had to open the device (lots and lots of screws).

Cheers,
Vitor

[Bicurico]

This has become an obsession... 

I resoldered every pin and the result was a temporary fix.

Back and forth, alternatively the video, the audio and the videotext worked - in any combination.

Now the video works, but with some interference. As a result the videotext does not render properly. Also, no audio.

I started to understand some of the circuit and measured it. It seems OK. Also, I don't think there is any bad solder point left.

What I do suspect right now:

1) The TDA9880T might be damaged (DOA) and has an intermittent problem inside. The problem is that it costs me around 25 Euro to order a new one and I don't want to risk this one, without being 100%. Unsoldering and soldering a replacement is a risk, too. I do have a hot air soldering station, but I prefer to avoid using it, as damage can be easily made in a fraction of time.

2) Some issue with the AGC? There are a few other components in viccinity, which could be bad. No idea how to measure these faults.

3) The chrystal. Should be 4.000MHz, but when I tried to measure it with my DS1054Z I only got around 2.8MHz. This can of course be due to me measuring it wrongly. Also, I understand that you cannot measure a chrystal while it is in operation, as the probe will actually influence the frequency. Again, getting this chrystal out and replacing it, will no be easy, as it is soldered from underneath. Possible, but not easy.

4) Something else I have not thought of, yet.

Any ideas?

Regards,
Vitor

[Rasz]

hair drier, cold air

[Bicurico]

I could use my temperature hot air soldering station.

The question is: at what temperature should I blow the air and should I heat up selected components? Which ones?

Can SMD capacitors and resistors suffer from a heat-related defect?

Also, I forgot to mention: when the device was fully working AFTER resoldering the pins on the TDA9880T, I left it turned on for like 30-60 minutes. After this period, the image started to get brighter and brighter and suddenly the defect was back (no image).

Turned it off for a while, still no go. Then toggled all modes back and forth and suddenly it was working again.

So again, it is not clear to me, if it is temperature related.

Then it failed again: no image. But I found out by accident, that changing sound from B/G to D/K would magically produce image. That definitely is not temperature related. Also, what has the sound mode to do with the video?

Then the sound came back.

Then it was gone again.

Last night I resoldered ALL pins and I am sure they are OK. Still no sound plus slight interference. Can it be due to remaining flux? How to clean off all excess flux?

Is it possible that the TDA9880T only works when HOT or COLD? Can it be broken internally to such extent, that it sometimes works OK, sometimes almost OK and sometimes not at all?

I really hate these intermittent issues...

Regards,
Vitor

[Emo]

Hi Vitor,

It sounds like your finger does deliver some charge on a pin suggesting a resistor normally setting a voltage on a pin has cracked or lost resistance. after some time the charge vanishes and your problem is back
Edit: you could try probing some pins with a high value resistor approx 1M connected to ground or some other voltage. you might be able to locate the area.

[Rasz]

Last night I resoldered ALL pins and I am sure they are OK. Still no sound plus slight interference. Can it be due to remaining flux? How to clean off all excess flux?

depends on the flux, 99% ipa

just blast hot air on it and see what happens instead of wondering
at this point sounds like PLL has trouble locking/is wondering around

[shakalnokturn]

The solders look dull on those colored components (SMD inductors?).
Check with a magnifying glass all solders, or just reflow...
Check/reflow shielding solders.
Check supply voltages right to the IC pins.
Try working the other way round, once it is working and open, try to trigger the fault by prodding/twisting the PCB.
Once you're dead sure it isn't a bad contact, (or at least not one you have access to) you could consider a bad component.
Some tracks run out under the shielding, check that there is no way it could short the track there, I'd feel safer with a piece of kapton tape there.

[EPTech]

Hi there,

I would try replacing the varicap diode (above the ACD0900) first and then the crystal.

Happy hunting.

[Bicurico]

Hi,
Today I tried my luck again.
I managed to measure the crystal and it has exactly 4.00MHz according to the Rigol.
After lots of messing around, at some point, pressing the TDA9880T down with some force resulted in getting a picture.
I checked and it does not seem to be a fault in the PCB. I rather think this IC is broken internally, having some kind of bad contact inside.
I checked everything else and even the voltages on the TDA9880T pins are correct.
It is a strange fault, nut I think I will take the risk and order a replacement.

Regards,
Vitor

[Bicurico]

*** UPDATE ***

I bought a TDA9880T IC on eBay - took almost 2 weeks to arrive from Germany 

Anyway, just replaced the existing one with the new one and the device now works 100%. Picture quality is great, including colour on external TFT monitor.

My conclusion: the TDA9880T was broken, it had some intermittent defect, probably some internal fracture on a conductor.

I am happy now!

Regards,
Vitor