EEVblog Electronics Community Forum
Electronics => Repair => Topic started by: junkpete on May 18, 2023, 01:20:22 pm
-
I am troubleshooting a Samsung TV (UE40D5700) which keeps rebooting, but it's not the usual "EPROM powercycle" problem which is typical for this model. If I turn the TV on, I can see for a brief moment "no signal" but after that the TV turns off. The background light does not even light up.
(edit: it seems it's not the power supply, since I got a replacement which behaves the same. Now troubleshooting the mainboard...)
I took out the power supply which is of course a switched-mode one and I when I plug it in (without anything else attached), I see 0.7W consumption which after a few seconds goes up to 15W and then immediately after that back to 0.7W for several seconds. This keeps looping...
I checked the output voltages on the cable which goes to the mainboard and I see that all of the supply voltages (5V, 12V, 13V) are most of the time close to 0V and every once and then they jump to 5V, 12V respectively 13V and then quickly go back to 0V.
The big 450V-rated electrolytic cap for the backlight oscillates slowly between 380V and 350V, in-sync with the other voltages.
I only found one small electrolytic cap which has steady 5.27V which is powered by the left MOSFET of the three which are below the heatsink on the upper right. The other two jump to around 12V and then quickly go back to 0V for several seconds. Those are powering the other caps which cycle in the same rhythm.
Any ideas where to continue? I am not really familiar with SMPS designs, so I thought I ask here if it's something obvious. Meanwhile I am studying how they are working...
-
I wanted to add that the the three components I mentioned are not MOSFETs but double Shottky diodes and all three are OK. I also checked all the other diodes and could not find any short.
The two MOSFETs under the heatsink in the middle together with a power diode are also fine. All other diodes I could find were fine as well. I desoldered all the electrolytic caps and measured their ESR values, which were also within specs.
I also applied via an external power supply 5V, 12V and 13V into the corresponding rails but could not detect any shorts. The current was below 20mA for each.
Any ideas are welcome, I am a bit clueless :-//
-
The big 450V-rated electrolytic cap for the backlight oscillates slowly between 380V and 350V, in-sync with the other voltages.
I expect that would be the bulk capacitor for the APFC circuit. Its voltage appears to be switching between the rectified 240V mains and the APFC boosted voltage.
What happens if you leave the PSU connected, but disconnect the backlight?
-
PFC could be the issue, however in standby mode it's usually disabled. So very likely PFC voltage change is normal. It easily could be non PSU issue at all. Also you cannot just test SMPS with nothing attached.
-
Oh I see, I thought that at least the 5V, 12V and 13V should be stable.
> I expect that would be the bulk capacitor for the APFC circuit. Its voltage appears to be switching between the rectified 240V mains and the APFC boosted voltage.
Yes that's what I also understood.
> What happens if you leave the PSU connected, but disconnect the backlight?
I tried only attaching the mainboard but I see the same power cycling behaviour, with a bit more current drawn.
-
This is what I get on the 12V rail when I attach the mainboard. The voltage stays longer at 12V but then drops again.
-
Just FYI, here is a service manual (no schematics):
https://elektrotanya.com/samsung_ue32d5500_ue37d5500_ue40d5500_ue46d5500_ue32d5700_ue37d5700_ue40d5700_ue46d5700_ue32_ue37_ue40_ue46d5000_ue19d4000_ue22_ue27d5000_chassis_u66a_u57a_u57b.pdf/download.html#dl (https://elektrotanya.com/samsung_ue32d5500_ue37d5500_ue40d5500_ue46d5500_ue32d5700_ue37d5700_ue40d5700_ue46d5700_ue32_ue37_ue40_ue46d5000_ue19d4000_ue22_ue27d5000_chassis_u66a_u57a_u57b.pdf/download.html#dl)
The 12V rail is the standby supply. Pin #2 of CN201 is SW_POWER, which I expect is an external power-on signal. I would think that, in the absence of SW_POWER, the B12VS rail should remain alive and the other rails should remain inactive.
-
Thanks for the manual. I also went through the (short) troubleshooting and already got stuck at the missing B5V and 13V. A5V is there with 5.27V (no other voltage is stable) but that's all, the remaining rails keep cycling as written above. Unfortunately the troubleshooting says "replace main assy" ::) so not much help from that. I also checked everything with a fully assembled device with and without the LED backlit, but no change.
Just to sum up: if everything is plugged in, I only see the screen lighting up for a few seconds, then "HDMI 4" (no signal) and then off again. The same power cycle happens with the isolated SMPS and also if I only plug in the mainboard.
I spent the whole afternoon, not sure if I manage to repair it in a reasonable amount of time, so I guess I'll give up for now. Feel bad for this 12 years old TV though.
Many thanks for the help so far, I'll definitely come back if I have any news! Of course I am happy to hear any other ideas, but I don't want to steal your valuable time.
-
I wonder if A5V is the "always on 5V" rail? I haven't read the manual so maybe I'm wrong about B12VS.
Edit: I just checked the manual. A5V is the standby rail.
-
You found the diagram on Elektrotanya SAMSUNG BN44-00423A or 00422A ? CM802 must be seen not to be shot and likewise the other blue capacitor in the PFC. CM813 must be measured and the very small resistance next to it. If everything is fine, the direction is towards the integrated circuit of the main source. Do you have 15V all the time?
-
Problems like this are almost always electrolytic capacitors. In a SMPS they have a hard life.
-
Problems like this are almost always electrolytic capacitors. In a SMPS they have a hard life.
It was true 10-15 years ago. With more modern equipment blindly recapping is a waste of time in most cases. Such TV more likely will have faulty LED strips rather than bad electrolytic caps.
-
It was true 10-15 years ago. With more modern equipment blindly recapping is a waste of time in most cases. Such TV more likely will have faulty LED strips rather than bad electrolytic caps.
I don't blindly recap, I test them, and even today the majority of problems I see with power supplies is bad capacitors. Yes LED backlights also commonly fail, but in this case he said power supply so I assumed it's the main power supply for the set, not the backlight driver.
-
Problems like this are almost always electrolytic capacitors. In a SMPS they have a hard life.
It was true 10-15 years ago. With more modern equipment blindly recapping is a waste of time in most cases. Such TV more likely will have faulty LED strips rather than bad electrolytic caps.
I desoldered all of them and measured them with a cheap ESR meter (like this https://www.elektor.de/lcr-t4-mega328-transistor-tester-esr-meter (https://www.elektor.de/lcr-t4-mega328-transistor-tester-esr-meter)) and they were all fine. Nevertheless, in my desperation I replaced a few of them which I had stocked.
You found the diagram on Elektrotanya SAMSUNG BN44-00423A or 00422A ? CM802 must be seen not to be shot and likewise the other blue capacitor in the PFC. CM813 must be measured and the very small resistance next to it. If everything is fine, the direction is towards the integrated circuit of the main source. Do you have 15V all the time?
Ah I got it thanks. I thought it's only the TV service manual which is up, did not think about searching for the SMPS model number. What I immediately noticed is that it's not the original SMPS board 00422A but the one for the bigger TV (with 46", I have the 40") named 00422B (found it here https://elektrotanya.com/samsung_bn44-00422b_sch.zip/download.html#dl (https://elektrotanya.com/samsung_bn44-00422b_sch.zip/download.html#dl)).
I will study the schematics.
As written above, the only signal which stays stable is A5V with 5.27V. Every other low voltage signal stays around 0V ad jumps every few seconds. The high voltages do not go to 0V but oscillate between e.g. 380V and 350V with the same period.
-
CM813 must be measured and the very small resistance next to it.
813 measures 22nF, that should be OK and also the small resistor (red black silver gold = 0.2 Ohm) measures spot on 0.199 Ohms with my Keithley 197 via 4-wire measurement.
-
Do you have clean 15V when the power to the main controller of the source turns on? Leave the PFC block alone because it works well.
-
I don't know where I should measure 15V. I only have the A5V, B5V, 12V and 13V. As written above, only the A5V (=5.27V) is stable. All others are 0V and jump every few seconds to 5V, 12V and 13V, then back to 0V. See the oscilloscope pictures.
-
It is not clear to me what board you have. Is that the one in the photo with the FSFR-1700? What is the main source controller? If you have FSFR 1700, it is possible for the OVP protection to work and one of the series resistors to be devalued.
-
I've had a few experiences with Samsung PSU's based on the Fuji Electric IC like this one.
Most of the time they'll stop running when warm at first until it gets as bad as not working at all.
Considering the components you've already tested I'm confident enough ICM801 is the cause by now.
-
It is not clear to me what board you have. Is that the one in the photo with the FSFR-1700? What is the main source controller? If you have FSFR 1700, it is possible for the OVP protection to work and one of the series resistors to be devalued.
It's the BN4400423A, the photo in the original posting and it looks very similar to BN4400422A which is shown in the Samsung Service Manual for all the UEXXD5X TVs on page 4-97.
Yes, I think that below the bottom right heatsink it's an FSFR1700, I have not desoldered the heatsink yet.
I've had a few experiences with Samsung PSU's based on the Fuji Electric IC like this one.
Most of the time they'll stop running when warm at first until it gets as bad as not working at all.
Considering the components you've already tested I'm confident enough ICM801 is the cause by now.
OK I'll have a look, thanks for the hint!
-
@shakalnokturn many thanks, indeed, pins 5, 6, 7 and 8 are all shorted! Let's hope that it's the only issue.
edit: ...and let's hope I can find one (or an alternative...)
Hm maybe I am mixing up something. It's labeled labeled ICS801S on the board, is that what you mean @shakalnokturn? It's a D2011K but I doubt it's the same as this: https://html.alldatasheet.com/html-pdf/41976/SEME-LAB/D2011/319/1/D2011.html
I checked the pads on the board and the pads of pins 5, 6, 7, 8 of IC801S are also connected.
How can I test this IC? I could not find anything else about it.
-
Try to heat this FSFR1700 with a hairdryer and then power it up. If you don't get a result, try to cool this circuit with a Freon spray and then power it on. Sometimes it comes back to life for a short time. You have checked the series resistors R824-R828?
-
Sorry for the confusion I'm talking about the same one as nenea dani, I'm not sure I read the right position on the PCB.
Highlighted one on the photo.
-
I resoldered the D2011K but since then the board is behaving differently. Now it's using 0.7W and every few seconds something around 2W or so. The 390V on the big cap is stable and also the 5.27V standby voltage is still stable. However, now I don't see anything on the screen (before, I saw HDMI 4 for a second, every 10 seconds or so). Also the red standby LED is flashing periodically.
I also tried to heat up and cool down the FSFR1700 but no change.
Whatever happened (I only desoldered and resoldered stuff), something is different now :-//
I found a replacement board on eBay for 25 Euro, I'll try that in a few days when it arrives and report back.
-
The replacement power supply is there but either it has the same fault, or the behaviour I observed with the original SMPS is totally normal. Nothing has changed, I see HDMI 4 for a second, then idle and 0.7W and after a few second it repeats. The A5V is there with 5.3V and the others are oscillating as well.
I guess I have to search for the problem on the mainboard...
-
You must disconnect the SW_POWER(POW/on/off) wire from the motherboard and leave it in the air. You should have 12V and 114V(for inverter) all the time if the power source is good. The wire is found on the power connector that feeds the motherboard, I think CN201, no. 2. if I'm not mistaken.
-
Did you check whether the backlight including driver is working?
Is there a flashing pattern on the standby LED?
-
I managed to replace a NAND chip in a Samsung UE46D5700.
This unit also got stuck in the boot sequence.
This chip with (software installed) ordered in France and on mainboard
put. TV set works properly again.
Costs €26.95 (incl. shipping costs).
https://www.mjkelectronics.com/en/samsung-mjk-electronics/246-samsung- (https://www.mjkelectronics.com/en/samsung-mjk-electronics/246-samsung-)…
MJK Electronics
contact@mjkelectronics.com
+33 768652794