PCB needs repairs

[partyanimallighting]

Red to blue ~ open. Red to green ~ continuity. Red to yellow ~ open. Here are the pix also. As you can see, I pulled a few parts for replacement. Besides the pcb being faulty, a large part of the problem has to do with my inexperience.

http://www.4shared.com/photo/1L7tzbsR/Follow_Spot_250_045.html?

http://www.4shared.com/photo/HYxnzyqB/Follow_Spot_250_046.html?

 

[KrisBlueNZ]

Good idea CocaCola. I have assumed that the positive (left) terminal of the left electrolytic is connected to the input of the 7805 by a fat track on the top side. I can see some of the track but not all of it.

 

[partyanimallighting]

So this helps? I'll pick up new replacement parts tomorrow and check in before I start to re-install so that I can benefit from your guidance and wisdom, O wise ones. See you tomorrow....

 

[KrisBlueNZ]

... And guess what, it doesn't! So how does the +18V get from the left terminal of the left electrolytic to the input of the 7805?

Apart from the bridge rectifier, the only place it goes is to the relay, where it comes through from the bottom side to the top side on the bottom right pin of the relay. Then it goes to a track that disappears off somewhere near the J12 marking, although that might just be going to the cathode of the back EMF diode.

The input of the 7805 goes to the big dropper resistor for the relay coil, and to J7 and J8 next to the bridge rectifier. Then that track looks like it runs off to the second electrolytic. That doesn't make sense yet.

And there's a fat track that runs under the relay on the top side. It seems to run towards C4, but C4 seems to be connected to the +5V rail with a track on the bottom layer.

I'm confused!

Do we have a pinout of the relay?

CocaCola, I like your approach of annotating an underside layout with points to measure and asking for voltages on all the important points on a good board and on the faulty board.

In the meantime I created an image of the top and bottom layers, with the bottom layer mirrored so you can match them up. I'll attach it to this post.

 

[CocaCola]

Yeah what Kris just said, are you sure this is a common 7805? Because nothing makes sense anymore... And I have come across other TO220 regulators with different pinouts...

 

[KrisBlueNZ]

CC, I think he said it was marked 7805. Also, the left pin, which should be the output, DOES connect to the two MLCC decouplers and to pin 40 of the micro (via a ferrite bead), so I think it is a normal 7805.

To the OP, I have some more questions. What do you plug into the boards when you test them? Obviously you wire 12VAC into the connector at top right, but do you plug anything else in?

Have you been testing this "faulty" board with some external component not plugged in? A component that you normally DO plug in when you're testing other boards?

Can you compare all of the part number and revision number markings on the faulty board to a known good board?

What is the history of the faulty board? Did it fail in service?

What are the colour stripes on the big resistor above the relay? It's marked R8, I think. Also can you measure the resistance across it.

My next step, I think, is to mark up a picture of the underside with lots of measurement points and ask him to measure the voltages on all of them, on the faulty board and on a good board. I'm not sure what else to do!

To the OP. Please check that you've answered all of my questions as best you can.

 

[CocaCola]

CC, I think he said it was marked 7805. Also, the left pin, which should be the output, DOES connect to the two MLCC decouplers and to pin 40 of the micro (via a ferrite bead), so I think it is a normal 7805.

If that is the case then correct me if I'm missing something obvious (as I am tired) but it appears the only input power source then is from the J7 take off terminal aka off board, as the other take off appear to be marked 'fan' and they share a common ground... That is unless the track loops back around and catches the back side of the bridge rectifier on the top edge of the board were we can't see...

 

[KrisBlueNZ]

I've just noticed something. Look at the pictures in post 44.

Start from the left hand electrolytic positive. Underneath, a track runs down to two terminals on the relay. On the top, a track runs from the bottom right relay pin over to the cathode of the back EMF diode. On the underside, a thin (easy to miss) track runs from there to pins 8 and 4 of the 555. That's how the 555 gets power. I thought it would be powered from 5V but it isn't.

Now, the bottom middle pin of the relay (which also comes from the left electrolytic) is probably the contact common, and the top right will be one of the others, probably the NO contact. From this pin, on the top there's a track going up to the big resistor, R8 I think, and the other end of R8 goes to the 7805 input. Maybe R8 is open...

As for the track that runs up to the two 2-pin connectors next to the bridge rectifier, I think it's a dead end. There doesn't seem to be any decoupling capacitor on the 7805's input.

So the 555 will need to be present to make the relay close. I don't know WHAT it's for. It doesn't seem to be a watchdog, because I can't see any connection to the micro. Maybe it detects fan failure. He said one of those 3-pin connectors is for the fan.

What do you think?

 

[CocaCola]

Start from the left hand electrolytic positive. Underneath, a track runs down to two terminals on the relay. On the top, a track runs from the bottom right relay pin over to the cathode of the back EMF diode. On the underside, a thin (easy to miss) track runs from there to pins 8 and 4 of the 555. That's how the 555 gets power. I thought it would be powered from 5V but it isn't.

A decent amount of JPEG artifact in that area but I believe you are correct...

Now, the bottom middle pin of the relay (which also comes from the left electrolytic) is probably the contact common, and the top right will be one of the others, probably the NO contact. From this pin, on the top there's a track going up to the big resistor, R8 I think, and the other end of R8 goes to the 7805 input. Maybe R8 is open...

Here is a picture of some HG 4130 relays for pin put, that matches the relay number as found in the first pictures of this thread...

What do you think?

I'm a little lost right now and burnt out but I do believe you are moving forward in the right direction... As for me it's been a long day with the children so I'm off to sleep I'll try to pick this up tomorrow if the kids give me 5 minutes of free time, but generally I can't get any 'thinking' time until late evenings when everyone is in bed already...

 

[KrisBlueNZ]

Cool, no problem.

I've just made a breakthrough on a SMPS from a flat screen TV that I'm trying to fix for my partner's brother. I've found a couple of zeners in the suppression circuit on the primary that are shorted. About time! I've been working on the stinking thing for hours! Now I just need to convince him to spend $40 on freight from America for two ****-arse diodes...

I'll post a diagram for the OP and ask for voltages at all the important points. TTYL

 

[partyanimallighting]

Hi guys, the linear regulator in the working pcb is marked L7805CV WCC14922 MOROCCO and the one in the faulty pcb is marked 7805A. But these were swapped out between pcb's and they both worked in the functional pcb. To Kris, to test, I just power up the faulty board next to the working board. Once the led lights, I know the working pcb is functional. Since I haven't gotten the LED to light on the faulty board, I figure it's not functional. I have tested it in the stage unit also but it doesn't work. On the part numbers, I've only changed out the 7805 and one relay and I've swapped these out between pcb's and they work in both. I pulled the big resistor also and the reading is 5.7 Ohms. I've attached pictures.

http://www.4shared.com/photo/hOx8L4wE/Follow_Spot_250_047.html?

http://www.4shared.com/photo/DbRyA9TR/Follow_Spot_250_048.html?

Off to get parts.

 

[partyanimallighting]

Hi Kris, Hi Cocacola, got all the parts that I removed last night. On checking the empty pcb, I am getting the required 12VAC into the bridge rectifier and about 14VDC out so that seems fine for now. I also see that the - from the bridge rectifier feeds one cap and the + feeds the other. Should I start by replacing these caps first?

 

[CocaCola]

Sorry not much time for me to actually look into this (kids and all) but yeah make sure you have some clean pictures with the parts off for reference, and start repopulating the board... At this point if we don't pop the cause soon you might just consider swapping all the parts (less things like the inline resistors on the ULNs that are almost certainly not the issue), might cost a few dollars but there isn't that much on the board that it's going to get stupid expensive...

 

[KrisBlueNZ]

Hi again, sorry for the slow reply.

Ideally I would like to see some good clear photos of the top and bottom side of the board with the large components removed. Then I think you should refit all the components, INCLUDING the 555 but NOT the micro and NOT the ULN2803s, and measure the voltages on all the marked points on this picture.

Attach your multimeter negative to the 0V point, which is on pin 1 of the 555 or anything connected to it. Measure voltage on each point, on the underside. Have the board powered up from 12VAC but with nothing else plugged into it.

I'm afraid all the letters are mirrored, because I annotated a picture that I had mirrored so that it would correspond to the top view picture. I'm sure you can handle it. The points are marked A~Q excluding I, L and O.

Measure the voltages on a good board and on the faulty board.

I'm sorry to put so much work on you. Normally a fault like this would be easy to find but there seems to be something strange about this board. CC and I are having a lot of trouble figuring out just how it's supposed to work and what's gone wrong.

Edit: You might want to clean all the crusty flux off the board before you take the photos. Use a cotton swab (a stick with cotton wool wound around each end - we call them 'cotton buds' over here) with isopropyl alcohol or any other solvent, and rub the board hard to get it really clean.

 

[QuantumCheese]

Hi again, apologies for delay been a bit busy here.

Looks like you've picked up some extra help too

It looks to me like the supply to the 7805 (and right hand cap) goes via the relay, hence the non working board has no 18V on the input to the 7805?
I did ask if the relays continuity had been checked as well as just hearing a 'click'.
Now for a bit of guess work - the relay is being triggered probably by a power on RC and then held on by either the 555 of the uProc - I can't make out the fine detailing of the tracks. One one picture there looks like a track shadow from the component side going from q1 to presumably the relay coil, yet on the component-side-up picture i can see no such track?? also one image looks like a mirror of what it ought to be which is a bit confusing - i'm used to flipping the image in my head! (not the one you intentionally mirrored KrisBlueNZ)

 

[KrisBlueNZ]

Hi QuantumCheese,
I believe he's already swapped the relay between another board, so we can assume it's OK.
The relay is controlled by a transistor driven from pin 3 of the 555. I thought the 555 was a watchdog for the micro, but it seems that the 5V rail won't appear until the relay closes, so either it's a watchdog for something else (I was thinking it might be monitoring a signal from a fan) or it does something else. I can't find any signal from the micro to the 555, though the track might be hidden.

I think the best way forward is to get voltages on all the critical points on a good board and the faulty board and try to figure out what the 555 is doing and why the 5V doesn't appear when the relay closes.

To the OP: You could also try bridging the +18V from the electrolytic that it's present on, to the input (right pin) of the 7805 and see what happens. I would get those measurements first.

 

[partyanimallighting]

Hi guys,

I replaced all the components and tested and .... nothing. Same problem. I'll run all the recommended tests and get back to you with the results. I did find an open through-hole on the relay side and repaired this but that did not help. I know it's a simple problem but because of my lack of experience, it's difficult for me to find and I am very grateful for all the assistance.

 

[partyanimallighting]

Hi guys,

Here are the pictures of the pcb with some of the components removed and afterwards loaded. The yellow arrows show the components that were replaced. I'll run those tests and get back to you in a bit.

http://www.4shared.com/account/home...&fau=1&ausk=weuCVfQPYzj9ePAq&hlfid=2911148342

 

[partyanimallighting]

Hi Kris and CC, here are the readouts as you requested.

A - 17.7VDC
B - 17.7VDC
C - 17.7VDC
D - 17.7VDC
E - 0.02VDC
F - 0.03VDC
G - 0.03VDC
H - (-)0.002VDC
J - (-)0.002VDC
K - 0.044VDC
M - 15.9VDC
N - 17.7VDC
P -12.36VDC

The relay is warm to the touch, the diac (bottom left) is hot and the resistors to the right of the diac are really really hot. Hope this helps. A link to a picture of what was done is attached. A Sharpie is a really great tool!!!

http://www.4shared.com/account/home...&fau=1&ausk=UBDLaXEt7HT07aVe&hlfid=2911907809

What next??

 

[QuantumCheese]

Hi KrisBlueNZ,
My mistake then i thought he'd 'heard it click' but not tested the continuity of the contacts. it does look like the supply to the rest of the board is controlled by the relay.

Which through-hole pin on the relay did you fix?
Anyone else here a little concerned about a hot relay/driver circuit? especially the relay.
Can you put your meter on the current range and measure how much the board is drawing (whilst unpopulated)?