EEVblog #682 - Ness D16X Alarm Panel Repair

Part teardown, part repair, Dave looks at an Australian designed and manufactured Ness D16X alarm panel that has failed.
What's that smell?
Can it be fixed?
How do you repair solder mask on a PCB?, or add solder mask to your own home etched PCB's?
And another look at PCB spark gaps.

Datasheets:
Fujitsu H8/538
Holtek DTMF Receiver
Maxcap Capacitors:
Epcos MOVs

Support the EEVblog through Patreon!

EEVblog Amazon Store (Dave gets a cut):
Donations:
Projects:
Electronics Info Wiki:

[McCuneWindandSolar]

When I was in the Marines. we had to repair boards like this that had a blown hole in it. we had a drimel tool to dig out all the carbon until we got down to clean board. Then we take a piece of blank PBC board cut it to shape to fill the hole we cut out and we had a special resin we would put on the board and then rebuild the copper traces. then with a heat gun we heat up the area and it would cure the resin, with the copper traces that would then reconnect back to the original traces then with a tab of solder to bond the new traces to the old ones. most of our boards didn't have solder masks it was mostly tin plating. some did have solder masks all do is either tin the copper traces or put the solder mask on.  and it be like new.  and a stop like what you show could take a day to repair. 

[richfiles]

I did one just like this about 2 months ago. My boss snagged a huge lot of machine tools, including a mill with a power feed. The place where it was at had the mains wired incorrectly. To get the US 120 VAC we take half of a split phase 240 VAC feed. The neutral was on the wrong terminal of the outlet... and whoever installed that mill, decided it was EASIER to fix the MILL than have the outlet corrected. /)_-

Of course, this means when my boss plugs his newly acquired 240 VAC mill into his very much up to code and correctly wired shop... the 120 VAC power feed controller goes POP, cause it's wired wrong! This video was more or less everything I had to do. Dig out the carbon from PC boards, trace component damage, and figure out what kind of MOV used to be there! XD

Since it was a VERY old power feed (maybe late 70s or early 80s), and since we wanted to play it safe due to the fact that he planned to turn that mill around and make a little something off it (he already had gotten the equipment he needed... this was just an extra one), I decided to be very aggressive with component replacement. It was a simple enough design, that we could hit up every semiconductor aside from the drivers for just peanuts, so we figured, screw it. Why not! :)

[giumacgyver]

Just put on wires from the AC input to the bridge. And to finish the job in a goos way, replace the rectifier, just to be 100% sure! This is in case you're expected to fix that board, otherwise throw it away!! 
Great Video! Great Dave!! 

[MyLastSong]

Had the worst lightning and thunder back in February here in California like I've never heard in my life. Lightning stuck super close and fried our Modem/router, phone, phone cable, phone jack, the phone box outside that connects to the telephone line on the pole. Basically anything that was connected to the phone line was completely fried. Never seen wires get so hot they desolder themselves. Inside of our modem looked exactly like that in the thumbnail

[richfiles]

One thing I very much like to do when either building or repairing boards that will feature higher power components or MOVs, etc, is I like to fit the component away from the board, and have a gap between it and other parts. It slips my mind, often, but Dave pointed out a VERY significant factor regarding char contamination... Flame rises. Had that board been rotated so no components were above the part that burned, the flame out would have appeared less severe, due to fewer parts being contaminated by the carbon left behind.

I like to raise my MOVs and power resistors up at least a small amount. I will also sometimes place barriers beneath or beside them, between near by components, when I have appropriate materials on hand.

[necessaryevil]

You must have upgraded the DaveCad software to the full version, the logo has been removed!

[ece]

Sometimes MOVs just develop excessive leakage and self-destruct on their own with no outside provocation. Given the way the board was charred all the way through on this one, it wasn't a sudden failure; it was "cooking" for quite a while.

[redtails]

21:39 huh.. I don't understand this.. If the supply voltage is still connected to the device, how did the protection device protect anything at all? The SMD protection device simply blew up and didn't change anything in the actual circuit?

[DanielVidz]

9:50 Yes that's what those resistors are for, *EXCEPT!* They're installed wrong just like every alarm installation I've ever seen.
They're called End-Of-Line (EOL) resistors and you guess it! *They're meant to be on the END OF THE LINE!* Otherwise anyone "could" just cut & join them like Dave mentioned.

[babybearkill]

Hey Dave, wanna do a fundamental Friday on MOVs and transorbs? Good education on HV protection is required for the General Public

[Graham_Langley]

I've been a long time lurker but you've now moved me to comment.

Of course the two VDRs/MOVs Y10 & Y11 will limit the voltage across the blown cap - they're connected in series across it via their common OV/ground connection.

And of course the 7805 (which looks like it's jacked up to give 13.8V or so to charge the battery - there's a ceramic pot next to it for adjustment) doesn't get hot when you power it up at 24' 30" as it's not charging a battery or supplying any external kit like PIRs. 

[KrisBlueNZ]

20:55 "you can have a huge differential voltage [...] 240V or 1000V and these MOVs [to ground] won't blow because it's not relative to the mains earth". Surely that's not true! Those two MOVs are in series across the input voltage.

[friggetyfuck]

Had to deactivate one of these today. Opened up the case that has a tamper switch, unplugged the battery, turned off the mains power, then plugged it back in with the case open and boom you're in the "installers" programming mode. 

[electronicsNmore]

Always nice to find a highly visible area like that when checking a faulty board.

[power-max]

It is an 85*C Cap. I can kind of sort of see it @ 3:50

[theLuigiFanProductions]

EEVblog
Please finish fixing it, I want to see that done and see it working again. :)
I would cut it completely out, make a small PCB that fits there (maybe the type that works like a solderable breadbord), and glue its sides to the existing board with epoxy. I believe that would eb the most "clean" looking way to fix this, or you could run wires to a seperate tiny board.

I would definitely replace the capacitor, knowing how a innocent looking cap can cause a PC (or anything really) to be unbelieveably buggy. I replaced about 18-25 filter caps and a few regulators on my motherboard a few years back. I think the caps are what cooked the regulators, because there were a few MOSFETs in parallel and it caused a cascading failure as caps shorted and overheated them one by one. The board in question was recalled in 2005 for faulty caps, and was a miracle to have ran untill 2011/2012. I got it a few months before it died, used from someone who had upgraded. It didnt burn the board, but it definitely discolored the solder mask, its still a brownish/wilted leaf color haha.  The PSU had gone out too, a few weeks before, which probabally stressed the poor counterfit caps even worse then their stolen electrolyte formula haha. Im just shocked the CPU survived 4 of the 12V to 1.45V mosfets regulators going open and drawing most of the power through its other power pins (CPUs generally have a mono rail design or every so many pins gets its own regulator). The GPU was being buggy before it failed as well, not knowing one of the two power regulators for the AGP slot had blown clean in half. Cannot believe the traces (and other regulator) survived the full power going through it when usually both idle aroung 150F-170F.

Anyways, interesting repair video, I always like trying to figure out what is wrong before you say it. The second the video began playing. I noticed the marks in the lower left corner. It was interesting to learn how you can figure out what caused failure, because generally Im not too good at that. I just replace everything I can in a PSU when it fails and cross my fingers. I can figure out what has all failed, but usually have no idea to the start of it.

[dooronron]

Did you ever do a video of you routing this out and repairing it, would love to have seen it up and running again

[power-max]

If I was repairing the board, Id just bypass the entire charred area with a separate perf board with the AC to DC conversion stuff and protection devices, and have a few mod wires going to it! Not sure where I'd mount it, if there is room in the case, probably just directly above the charred area, then I would cut the traces to ground zero area! 

[pvc]

I don't know if it is worth the hassle but you could just cut out whole corner and put small custom PCB (with a protection components and diode bridge) instead. Then solder it directly to the filter cap and input terminals. Even copper stripe board would be ok.

[markusfuller]

Thanks Dave. burnt Carbonisation on the board has caught me out before and kept me measuring round for ages trying to figure out whats going on even after taking away components and still reading a resistance. bloody annoying when its staring you in the face. face and palm eventually meets. Thanks again from Markus