Maker Pro
Maker Pro

K795 78 board repair.

Status
Not open for further replies.

JonathanAnon

Mar 22, 2012
50
Joined
Mar 22, 2012
Messages
50
Hi lads,

flytech k795

Just looking for a general piece of advice. I have this PC / Touch Monitor / all in one thing that is used in a factory that I do contract work for... It stopped coming on after an electrical storm and they asked me to have a look at it... out of warranty, and the company Flytech have not responded to any of my emails. So I pretty much have free reign to have a bash at fixing it, so I'm gonna try for the experience.

Now when I got this thing on my desk, I plugged it in and nothing happened.. It has a 19v 4.24A power supply (like a regular laptop power supply but with two sources for the PC and the scree), which I borrowed from a working K795 78.. so I know that this bit is working.

No signs of life when I hit the power button, no LEDs, no fan noise, no screen display, no POST... I went off to do something, came back 20 mins later and there it was booted up !! And this is the behaviour that it displays every time... Then it stays running.

- If you reboot it (warm boot), it comes back up straight away.
- But if you switch it off and leave it to cool down for about forty minutes, it again takes another twenty to boot. And once it boots it's fine.

I dont have any pictures yet, but just looking for what approach you would use to solving this. I have my own suspicions, but I would appreciate opinions from anybody who has had success overcoming a similar fault.

thanks,
 

GreenGiant

Feb 9, 2012
842
Joined
Feb 9, 2012
Messages
842
sounds like something in the power supply, bad cap(s) which is a typical problem when things get zapped

Bad caps dont always just stop working, they can sometimes just change in properties, so theres a chance that once the caps are nice and energized (and warm) they work close enough to properly in order for the computer to work.

In short, check the caps in the power supply, but be careful caps can be nasty little buggers
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
Another possibility: There could be an inrush protection using an NTC in the power supply (of the main board, not the external one - if that one is definitely working). In the cold state the NTC would limit the inrush current. The current, in turn, would heat the NTC which in turn would reduce its resistance and allow more current to flow.
The NTC could have been hit by a voltage surge and somehow be damaged, thus taking much longer to heat up and allow full current flow.
 

JonathanAnon

Mar 22, 2012
50
Joined
Mar 22, 2012
Messages
50
That was along the lines of what I was thinking alright.. though I never thought about the thermistor, I couldnt even find one on the board... Still toying with the idea of trying to having a go at fixing it. I've looked over the board for an hour and all the caps look fine (though I know that doesnt necessarily mean that they are all working). There is just one curious thing in the centre of the board.. It's marked X3 which is fuse holder on my chart..

30072012923.jpg
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
If the foil coating that crystal (looks like one, and X is often used for crystals) is flaking off, there is a possibility it has shorted something.

I think I can see some flakes of stuff on the board...
 

JonathanAnon

Mar 22, 2012
50
Joined
Mar 22, 2012
Messages
50
okay, I've tried to get a few more images from different angles... Please remember that the device is reflective and cylindrical in shape, so some of the colour on it's surface is reflection of it's surrounding objects..

http://omg.wthax.org/crystal_1.jpg

What is the yellow stuff, is it glue or coolant? I've measured across it with a multimeter and there is no continuity..
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
okay, I've tried to get a few more images from different angles... Please remember that the device is reflective and cylindrical in shape, so some of the colour on it's surface is reflection of it's surrounding objects..

http://omg.wthax.org/crystal_1.jpg

What is the yellow stuff, is it glue or coolant? I've measured across it with a multimeter and there is no continuity..

its just glue to hold it in position, it doesnt look physically damaged
for a crystal, you wont measure anything with a multimeter

most likely the fault is else where, crystals are relatively failure free

Dave
 

JonathanAnon

Mar 22, 2012
50
Joined
Mar 22, 2012
Messages
50
okay I'll leave that crystal alone and assume that it's working.. I realise at this stage that I could be looking for a needle in a haystack, but at the moment I thinking of two approaches...

1. Get a hair dryer and try heating different parts of the board (to initiate a boot) to see if I can identify the area with the problem.
2. Test the voltage across the all of the electrolytic caps.. See if I can work out where the blockage is.

Again, all opinions welcome.
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
OK one thing you havent confirmed/commented on is....

the plugpack type PSU ... is it working correctly from cold ? ie putting out the 19VDC
if this failure was related to the lightning storm, then the PSU would be first in line to be damaged and unless there was a near direct lightning strike, the PSU would probably have taken the brunt of the excess voltage.

There is of course the possibility that the failure coincidentally occured at the same time and the PSU or monitor was on the verge of failing anyway.

Those inline PSU fail with monotonous regularity, I have repaired a number over the years, its generally always the electrolytic caps in the DC output side bulging and going leaky.

SO ... firstly confirm the PSU, find another one lying around to try on the monitor to see if the monitor then fires up straight away.
Do this BEFORE you condem the monitor as the culprit and report back to us :)

Dave
 

JonathanAnon

Mar 22, 2012
50
Joined
Mar 22, 2012
Messages
50
I've tried two different power supplies.

1. The power supply that came with it.
2. When the faulty PC failed originally, it was replaced by a brand new machine of the same model (which had it's own brand new power supply). I borrowed the new power supply from that PC two weekends ago, tried it with the faulty PC, and it displayed the exact same behaviour. That power supply is back in the factory working away on the new machine.

I'm confident it is the PC itself and not the power supply.
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
I've tried two different power supplies.

I'm confident it is the PC itself and not the power supply.

ok, .... Then you need to do some checking with the multimeter see if you are getting any of the correct voltages from the sections of the internal PSU circuitry
That 19V power pack will go into the unit and then very likely to several regulator circuits say 5V 12V etc etc
you need to check to see if any of them are there or missing during the fault condition and during normal operation


Dave
 
Last edited:

JonathanAnon

Mar 22, 2012
50
Joined
Mar 22, 2012
Messages
50
I was on the way back from band practice, and passing through the 24 hour shop, saw a hair dryer for six quid and said to myself wtf... I'll give the hairdryer trick a try... Whipped the back off the PC (quietly cos it's 3AM here).....

Plugged in the adapter and heard the usual power noises with no boot... Switched on the hairdryer and focussed it on the area where the power comes in (which has a few transistors, and small caps)... nothing happened...

After about 60 seconds I moved the hairdryer to the following section, just above the processor.

caps_1.jpg


And after about 5 seconds, IT BOOTS... Tried this twice more and same thing happened.. So I have a solution to the problem, I'm gonna put the PC back in the factory and buy them a hairdryer.. :)

Does anybody have any clever way of isolating the caps to test them individually.. I suppose I could use cooler. I'll take a look at the back of the board, and see how difficult a solder job it would be to replace them.
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
you can put a cap of the same value across each of those caps in turn and see if that finds the faulty one
I can only read one value in that pic cuz of the angle the pic was taken.... the right hand cap is 150uF 16V

none of them look bulging or otherwise physically bad but maybe one or two of them are starting to dry out

Dave
 

JonathanAnon

Mar 22, 2012
50
Joined
Mar 22, 2012
Messages
50
Oh I agree lads, I did the same yesterday when I had a TV with two dodgy rectifying diodes.. Replaced all four... However, with these caps, I'm not sure what will greet me on the other side of the board. They may not be easily accessible to my soldering iron. And I REALLY dont want to make the situation worse.. The board is working to some extent at the moment, and I would rather not leave it in the position that it doesnt not boot at all. It's a tricky call.. :(
 

JonathanAnon

Mar 22, 2012
50
Joined
Mar 22, 2012
Messages
50
I pulled out some of those capacitors.. They are actually 820uF 2.5V by the look of it..

820uF_Capacitor.jpg


On closer inspection it seems that there is a bit of deposit on the top of the 100uF 25V Capacitor. So this could be the one causing the problem..

100uF_25v_Capacitor.jpg
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
yep, looks like 2.5V

Looks something like this

There are several different types with different ESR figures, so this may not be an exact replacement
 

JonathanAnon

Mar 22, 2012
50
Joined
Mar 22, 2012
Messages
50
Hmmmm, replaced the brown capacitor above and still the same characteristics.. (I'm waiting on my new DMM to test all five caps before I go ahead and buy replacements.. Those silver ones are quite expensive, so I want to test them as much as I can before buying them).

http://www.alldatasheet.com/datasheet-pdf/pdf/123525/VISHAY/SI4431BDY.html

There is also a chip similar to above, beside the caps, I'm wondering could it be this that is actually affected by the thermal change... Then there is a load of SMD caps as well.

Would it be a good idea to go through the components one by one with coolant, and see if I can isolate the problem? Or is there any tricks of the trade to isolate the heating to individual components?
 
Status
Not open for further replies.
Top