17IPS20 No Led Voltage Solved With Video :)

[budm]

Re: 17IPS20 No Led Voltage Solved With Video

BTW, this the info on the VCC pin 6V LDO output.

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

Pin 16 with power off to Tv shows 1.89 on the 20k scale to earth.

[budm]

Re: 17IPS20 No Led Voltage Solved With Video

Pin 16 with power off to Tv shows 1.89 on the 20k scale to earth.

OK, we do not have shorted out MLCC at the VCC pin, at this point I would replace the IC since the 6V LDO is not working.

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

Ok -will order one and try that and report back. Thanks again for the ongoing help - its much appreciated.

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

This TV will be the death of me I think!!
The ordered MP3394S turned up early so fitted it today although the traces on this Vestel board are really flimsy. Two pads lost so had to substitute with fine wire to the new IC pins. Fired up board and almost nothing in the way of voltage - no 156v and the IC pins showed little or no voltage. Rechecked soldering in of IC and it seemed ok. Checked connecting cables between PSU and main board and as I did so the backlights all fired up and stayed on! Poor cable connection I suspected but at least proved that the culprit had been the MP3394S.
With the LEDs running checked voltage at beginning and end of array which was very low on both ( I assume this is because the 156v is being fully used by LEDs?).
Thought I had better check each IC pin for voltage so I could report back to you and just as I began there was a quiet electrical crack from board followed by another!! Quickly switched off and found that there had been a loose trace where the wire had been soldered from Pin 14 (GATE) and this had dislodged and shorted to the trace from Pin 13 (ISENSE) !!
The cracking noise was 5 resistors going up in smoke:
R509 [10R] Marked 1R0 = 1 Ohm ( I think ??)
R518 [10R] Marked 1R0 = 1 Ohm
R557 [10R] Marked 1R0 = 1 Ohm
R514 [10R] Marked 1R0 = 1 Ohm
R552 [330k] Marking ?
Cant see any other physical damage but until I replace these 5 resistors it will be difficult to know if the board will fire up again or even if the mosfet or IC will be damaged.
The circuit diagram is very misleading in general as it shows some of the resistors on the board as 330k but then if you look at them on the board in that position they are marked 154 which is 150k ohms so without being able to see the marking on the burned R522 I am struggling.
Any comments on the above disaster and what other damage may have occurred?
Also can you provide any greater clarity on the actual spec for the resistors shown as blown?
So close yet so far away - but at least it ran and every LED was lit!
Thanks again.
John

[budm]

Re: 17IPS20 No Led Voltage Solved With Video

The schematic is for making the BLANK PCB, the value of the component will not always match the schematic because the blank PCB will be stuffed in may ways to make many version of the PCB assembly, VESTEL never release the BOM of the PCB assembly so you can easily know what parts are being used on the particular PCB assembly.

According to the schematic, that resistor R552 is just a part of bleeder resistors network to discharge the caps.
Those four 1 Ohm resistors are connected in parallel to form 0.25 Ohm current sensing resistor.
The MOSFET connected to those resistor is still good? I would replace the IC again since we do not know what damage is done to the IC.

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

Thanks for the points and comments.
Fortunately I did order two MP 3394s ICs but am reluctant to just change it given the very poor pads and traces it connects to without trying the set first with the resistors replaced. If I do replace it I will probably snip off each pin at the IC then remove IC an dthen unsolder each pin from pad.
I have attached the best of the photos I earlier sent you of the area concerned showing the resistor markings as they were.
I think it is clear the parallel mounted resistors are marked 1R0 so 1ohm.
R522 and an adjacent resistor (R502) look to be identically marked as 100 on the photo (highly magnified view of the board shows the burnt R522 to have only three digits and the first one is clearly a 1 and the other two digits as probably the remains of the top and bottom of two zeros).
According to my reading smd resistors marked 100 are 10R or 10 ohms not the 330k as shown on the schematic for R522. That seems a hell of a difference between what I suspect and the generic schematic marking!

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

Take 2.
Have ordered the smd resisitors but in interim have soldered on equivalent through hole resistors to see if the board would boot up. No joy! No 156v on anything including the mosfet. Vin measured 11.8v on IC and backlight pin at 3.2v but the rest of pins showed almost no volts.
Despite earlier post I pulled the MP3394 and re-installed the original one which at least we knew allowed the 156volts but now it functioned as the removed new one ( ie 11.8v at Vin + backlight 3.2v otherwise nothing). So possibly another component is also blown but I cant find anything and all fuses are fine. Re-tested D 503 which was fine. The 12v, 3.2v and 24v on PSU and Main board are there ok. I think I will hold off fitting the remaining new MP3394 until I hear your comments on these two posts as I am obviously overlooking something I suspect.

[budm]

Re: 17IPS20 No Led Voltage Solved With Video

You have Vin so you should have the 6V on VCC as explained earlier, that should be present as long as Vin is present.
OK, I mess up my post #66, I was referring to R552, not R522 that burnt out as per your report.
R522 is connected between the Source pin of boost converter and pin 13 I_SENSE pin of U500, so for R522 to burn up, too much current has to flow through R552 and pin 13 of the IC to ground so it my mean that pin 13 of U500 is damaged, or C519 break down and shorted out, please check the resistance between pin 13 and ground to see what you get.

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

The IC in situ when the short happened is now removed so all I could do was test it between its Pin 13 and its ground pin and that showed open circuit. Testing the one remaining new IC similarly it gave infinity.
I then tested the re-installed old IC and it showed open circuit too on Pin 13. Unsoldering its Pin 13 from circuit and then testing it to earth still shows OC. Now as it wasn't oc when in situ before either it has gone oc since re-installation or there is another factor I am not aware of? Not sure of the best way to test C519 in situ. Putting the MDM on it at 200ohms shows 0.11 with no rise in that reading whereas testing a cap along board similarly but at a higher range the reading rises as you hold the probes in place as I would expect

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

Further update.
The cap along the board I measured was C 514 which according to the schematic is the same capacitance as C519 so does not perform the same.
Have also now full removed the original MP3394 and re-measured it out of circuit it now shows infinity on 200k setting and not OC.
Out of circuit:
So Pin 13 to Pin 4 (GRND) on original faulty IC on 200k = infinity. On 2M = 0.680
So Pin 13 to Pin 4 (GRND) on ?blown/ faulty IC on 200k =open circuit. On 2M = OC
So Pin 13 to Pin 4 (GRND) on new unused IC on 200k = infinity. On 2M = infinity.
I am sure you can explain that better than I can!!

[budm]

Re: 17IPS20 No Led Voltage Solved With Video

C519 is shorted since it shows 0.11 Ohm.
Pin 13 could have blown open circuit after over current flown occur.
How about Vcc pin after installing the new IC? You should have 6V on that pin.

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

Have now reinstalled the last new MP3394 (very difficult as pads and traces now very poor!).
When I test Pin 13 to the board earth I get Infinity but if I put the black probe on Pin 4 (grnd) I get OC !! Why is that?
If I fire up the board with the shorted cap still in place to test Vcc am I liable to damage the IC ?

[budm]

Re: 17IPS20 No Led Voltage Solved With Video

Have now reinstalled the last new MP3394 (very difficult as pads and traces now very poor!).
When I test Pin 13 to the board earth I get Infinity but if I put the black probe on Pin 4 (grnd) I get OC !! Why is that?
If I fire up the board with the shorted cap still in place to test Vcc am I liable to damage the IC ?

You are not making any sense right now.
Per your post: "Not sure of the best way to test C519 in situ. Putting the MDM on it at 200ohms shows 0.11"
C519 shows 0.11 Ohms on 200 Ohms scale, that cap is connected to pin 13 of the IC U500 and ground, so the resistance between pin 13 and Gnd should also show 0.11 Ohms.
I have never seen meter display 'OC', most will show 'OL' for over range for the meter to display.
You cannot try to run the board if you know you have shorted parts in place.
Please look at the schematic to see what are are connected to that IC U500 I_Sense pin 13 of the IC, also pay attention as to what the value of R522 is installed on your board.

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

Hi
I think some of our messages or thought processes are overlapping hence the confusion. Also I am using some abbreviations of my own making I think.
I always understood (probably wrongly) that smd caps could only be tested properly out of circuit hence my comment of not sure of best way to test C519 in situ. All I could do was to put the black lead to one end of cap and the red to the other which showed 0.11 ohms with no increase of reading if probes left on.
The rest of that sentence I was wrongly describing what I got (Tired I think!).
My meter shows 1 for infinity, open circuit or over range. The meter shows 00.0 for continuity or shorted items. So testing the new IC in place today Pin 13 to an earth point on the board showed 1 ie no continuity but if I place the black lead directly to Pin 4 (GRND) on the IC the meter reads 00.0 ie continuity or short. That is what I don't understand - why two very opposite findings from Pin 13 to earth/grnd? I would think the 0.11 resistance would not show as I was probably in the 200k setting.
Hope I have that right this time.
When I asked you about testing Vcc with the shorted part in I was responding to your question earlier today of " How about Vcc pin after installing the new IC? You should have 6V on that pin" but at that point there was no IC in the circuit as I had said so thought you wanted me to report the Vcc when new IC put in.

[budm]

Re: 17IPS20 No Led Voltage Solved With Video

Hi
I think some of our messages or thought processes are overlapping hence the confusion. Also I am using some abbreviations of my own making I think.
I always understood (probably wrongly) that smd caps could only be tested properly out of circuit hence my comment of not sure of best way to test C519 in situ. All I could do was to put the black lead to one end of cap and the red to the other which showed 0.11 ohms with no increase of reading if probes left on.
The rest of that sentence I was wrongly describing what I got (Tired I think!).
My meter shows 1 for infinity, open circuit or over range. The meter shows 00.0 for continuity or shorted items. So testing the new IC in place today Pin 13 to an earth point on the board showed 1 ie no continuity but if I place the black lead directly to Pin 4 (GRND) on the IC the meter reads 00.0 ie continuity or short. That is what I don't understand - why two very opposite findings from Pin 13 to earth/grnd? I would think the 0.11 resistance would not show as I was probably in the 200k setting.
Hope I have that right this time.
When I asked you about testing Vcc with the shorted part in I was responding to your question earlier today of " How about Vcc pin after installing the new IC? You should have 6V on that pin" but at that point there was no IC in the circuit as I had said so thought you wanted me to report the Vcc when new IC put in.

OK, when you test the board, was the board fully mounted to the metal chassis of the TV? Most TV will ground the cold side of the circuit to the chassis using the board mounting screws.

Pin 4 (GND) of the U500 is connected to the cold side circuit ground but it may not be connected to the chassis ground, most TV COLD SIDE circuit GND is also connected to the chassis GND, so you can verify if pin 4 is connected to chassis GND or not by checking the resistance between pin 4 and chassis, it should show the same reading as when you touch your two meter probes together.

So at this point if that cap is showing 00 Ohms when tested on the board, then remove the cap and re-test the pads where the cap was to see if the 000 Ohm is now gone.

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

Further update.
You commented "also pay attention as to what the value of R522 is installed on your board". I had mentioned that in my post #67 the schematic said it was 330K but mine was marked 100 or only 10 ohms. So the Resister I put in was 11ohms ( nearest I had)(could that be the reading of 0.11 obtained earlier)
In terms of the grounding you were spot on. The board is not attached to the chassis and I was using ground on the hot side. When tested with the right ground point Pin 13 shows 000
If I disconnected Cap 519 ( torn trace when burnt resistor removed that had been soldered over to retain correction reading at Pin 13 remained at 000.
I started to unsolder cap Cap 519 then thought I would try a reading of it with the Resistor 522 disconected from it as well and this gave a cap reading of 1 on 200k and 1.45 rising on 2M (ie cap isolated except to earth)
Resolder cap, resistor and Pin 13 together reading on cap gave 000 on 200k and 000 on 2M.
This seems as though the cap if isolated is not shorted and is charging (unless my thinking has again gone wrong).

[budm]

Re: 17IPS20 No Led Voltage Solved With Video

Further update.
You commented "also pay attention as to what the value of R522 is installed on your board". I had mentioned that in my post #67 the schematic said it was 330K but mine was marked 100 or only 10 ohms. So the Resister I put in was 11ohms ( nearest I had)(could that be the reading of 0.11 obtained earlier)
In terms of the grounding you were spot on. The board is not attached to the chassis and I was using ground on the hot side. When tested with the right ground point Pin 13 shows 000
If I disconnected Cap 519 ( torn trace when burnt resistor removed that had been soldered over to retain correction reading at Pin 13 remained at 000.
I started to unsolder cap Cap 519 then thought I would try a reading of it with the Resistor 522 disconected from it as well and this gave a cap reading of 1 on 200k and 1.45 rising on 2M (ie cap isolated except to earth)
Resolder cap, resistor and Pin 13 together reading on cap gave 000 on 200k and 000 on 2M.
This seems as though the cap if isolated is not shorted and is charging (unless my thinking has again gone wrong).

OK, so R522 is 10 Ohms on your board, one end of that resistor is connected to pin 13 (I SENSE) of the IC, another end of the resistor is connected to the very low Ohm (<1 Ohm) Source resistors to ground. Since the resistance is very low so to be able to read low resistance using the meter that can show resolution of 199.9 Ohm, you have to use 200 Ohms scale to be able to read low resistance and able to show down to < 1 Ohm resolution. Using 200K or 2M scale you meter will show 000 dues to the resolution of high Ohm scales when testing low resistance, you just to learn how to use proper scale of the meter, using wrong scale will give you wrong conclusion.
Basically you need to set the meter scale accordingly when performing test, in this case for checking pin 13 of the IC and GDN with R522 (10 or 11 Ohms) in place, the meter should show about 10 ~ 12 Ohms on 200 Ohms scale depends on test probes resistance.
Since we have low Ohm connected to pin 13 and Ground, the only way to find out if Pin 13 has internal shorts (or low resistance) to ground or if the C519 is shorted (or low resistance) to GND is to not have R522 in place, and test C519 or IC off the board..
Regarding to pin 13 of the IC, the internal circuit may have now blown open circuit when over current flows through R522 to pin 13 to GND, so now when you test pin 13 and GND without R522 in place it does not show low resistance.
At this point, put new IC and C519 in place, no R522 in place, then check the resistance between pin 13 and GND, if no low Ohm reading then put R522 in place. Also do good visual inspection on all the IC pins to make sure there is no solder bridge or bad solder joints, no broken traces, etc.

[johnboy183]

Re: 17IPS20 No Led Voltage Solved With Video

I had already said in post # 73 that the new IC was in place. So the figures I gave above were with the new IC . Given the poor pad and traces serving the IC I had checked for solder bridging and bad joints at the time I installed the IC and all appeared fine. The only possible bridge was where a trace goes directly under C519 so the bulky solder joint made to connect Pin13 trace, R522 missing pad and the trace to C519 could have bridged that but again before the measurements above were taken the joint was cleaned, reduced and can no longer bridge the trace under the capacitor.
I will follow your above instructions tomorrow and report back using the meter scales you suggest. Thanks again

[budm]

Re: 17IPS20 No Led Voltage Solved With Video

If all the bad part are replaced and properly inspected, first thing you need to check are the Vin (~12V), and the VCC 6V of the IC.