Power supply circuitry not working

[jh112323]

Hi everyone,

This is my first problem I've posted, so I'm going to try and follow the suggestions about how to make it comprehensible. I'm pretty much brand new to electronics, so bear with me.

The item is a (cheap, Argos) portable DAB radio, that can either use 4x 1.5v AA batteries or a 5VDC 1A input (from adapter, UK 240VAC 0.2A). Unfortunately I have no circuit schematic, and its too cheap for anything to be online.

The problem is that the radio works fine using battery power, but not using the AC adapter. On the AC adapter, I get nothing, no lights, no sound, not a flicker of life.

The history of the problem: no specific incident, but has fallen in the past, no obvious damage to circuitry. The radio previously worked fine with the mains adapter.

What I have done so far:

Tested the adapter, I get 5.2VDC from the connecting jack, as well as the solder points on the circuit (see annotated photo), so the problem doesn't appear to be there.

There is a 4-wire cable connecting this smaller PCB to a larger PCB (see photo). From testing with my multimeter, the two wires nearest the edge carry power (labelled). When the radio is working on batteries, there is approx. 5V across these two, however when using the mains adapter, I only measure about 2.4V - I assume this is part of the fault?

I have tried to study the traces on the PCB between points, but can't really make sense of it. There are a whole host of tiny resistors, and there seem to be two separate paths to the power outputs to the next board.

Please excuse my inexperience, and my thumbs.

Let me know if I can provide any other info. Shame pics are low res!

Thanks in advance,

JH

 

[KrisBlueNZ]

Hi there JH and welcome to Electronics Point

Thanks for the thorough problem description and the annotated photos. It's great to see someone making an effort right from the start!

It looks like the batteries are switched through the adapter socket. On your first picture, the top connection from the batteries goes to a terminal on the adapter socket. This will make contact with another terminal in the socket until you put the adapter plug into the socket, which disconnects it, so the batteries are disconnected when the adapter is plugged in.

So the rest of the circuit is the same in either case - batteries, or adapter power. If the radio works on batteries, it should work on adapter power too.

My guess is that there is actually a problem with the adapter. You may measure 5.2V across the terminals with a multimeter, but that test doesn't put any load on the adapter, i.e. it doesn't draw any (significant) current from it. It's like a car engine that's running on three cylinders - it will rev high, but as soon as you let the clutch out, it will stall.

It's possible that there's a break, or a near break, in the adapter wire. These often happen right next to the plug, because of poor strain relief and bend radius control.

You can check for this problem by measuring voltage between the adapter socket pins with the adapter plugged in. If the voltage is only 2.4V when you plug the adapter in, or drops to 2.4V when you switch the radio ON, it's most likely that this is the problem.

You may be able to confirm this, and locate the break, by gently bending the adapter lead and pushing the cable in towards the plug. This can make the circuit temporarily. Once you've found the problem, cut off the plug and put a new one on, making sure that the polarity is right.

It's also possible that the adapter has failed internally; in that case, it's unlikely to be worth fixing - just buy another.

 

[shrtrnd]

I'd do what KrisBlueNZ said. If you've still got problems, I'd touch-up the solder contacts at the points in question, just to make sure they're all conducting.
A lot of the wall adapter plug-in stuff I work on experiences two primary problems because of the constant plug-in, remove plug, repetition:
1) the solder joints at the power connector (on the board) crack, and needs to be resoldered because of physical pressure on the jack.
2) the power jack (on the board) leaf contacts are pushed out of alignment from continuous plug-in, pull-out action, and need to be bent a little to realign them.
Good luck with the project.
One more thing that popped into my mind:
Is the adapter plug pin size the correct one for the jack? If this wasn't the original power adapter, know that there are different sized pins in some connectors.

 

[jh112323]

Hi Kris and shrtrnd,

Many thanks for your prompt responses. I understand your point Kris about how the fault likely lies in the adapter if the circuit is shared. What I am confused about is testing the adapter under load. I think I have done this already? Perhaps I'm getting the terminology wrong.

With the adapter plugged into mains and connected to the radio (switched ON), the voltage across the solder points on the board (the connector plug jack?) is ~5.2VDC. Isn't this under load, as the radio should be on? However at the points labelled in blue text, the voltage reads ~2.4VDC. This reading is the same if the radio is switched ON or OFF.

Under battery power, the reading at the points labelled blue text, the reading is ~5.2VDC when switched ON, and ~0.2VDC when OFF.

You are right in that if the radio is working with battery power, and you plug the AC adapter in, it cuts out.

I have wiggled all adapter joints whilst plugged in and switched ON, and haven't gotten any life out of it. I guess the next step is to open up the adapter plug body?

Are there other ways to test the adapter under load?

I know new AC adapters are cheap, even the radio isn't worth much, but I'm determined!

Thanks,

JH

 

[KrisBlueNZ]

Hmm, the plot thickens!

You're right, if the adapter is plugged into the radio and the radio is turned ON, that should be a test of the adapter under load.

I'm not sure you're right about which wires are carrying the power to the main board. Can you take a photo of both sides of the main board in the area where the four wires from the socket board come in please.

Also can you take two more photos of the socket board. One from directly above the non-solder side, and one at an angle from slightly behind, so I can see the top and back of the adapter socket.

Also, what is the thing behind your finger at the top left of the first picture in post #1? It looks like it has six connection pins and two mounting tags. Is it just the ON/OFF switch? Is that what it's called?

 

[jh112323]

Hi Kris,

Thanks for your reply, and sorry for the delay on my part. Busy couple of weeks!

I have taken a few more photos, hopefully that show you what you need. The thing behind by finger is indeed the ON/OFF switch.

In terms of carrying power to the board, of the 4 wires in that cable, I can only measure any voltage (2.4 VDC on adapter) on the two outer pins on the small board, or the corresponding two bottom connections on the main board.

What I can't understand is that even when switched off, the points to the main board still read 2.4 VDC? Is that a switch fault? On battery power it is only 0.2 VDC

Not sure where to go from here, hope you can suggest something else!

Let me know if you need any more info.

Thanks,

JH

 

[KrisBlueNZ]

OK, I think this may be a lot simpler than I realised. Can you make some continuity measurements please.



First I want to confirm that there's continuity between the points marked A and B, and continuity between the points marked C and D. Then check between B and C. If there's no continuity between B and C, L7 is open circuit. You can temporarily bypass it with a bit of wire or a solder blob (it's only used for interference rejection), and later tack on a through-hole (wire leaded) part: http://www.maplin.co.uk/p/rf-choke-470uh-wh39n

 

[jh112323]

There's continuity at all the points you suggested to measure, A-B, C-D and B-C.

 

[KrisBlueNZ]

Oh. Damn. I thought I was onto something there.

OK, I'm pretty sure that the two end pins are the power supply from that board to the main board. On the main board, they connect to the electrolytic capacitor marked C1 that you can see in the last three pictures in post #6, and from there, they go further into the circuitry on that board.



The negative side of the adapter comes in at A, and connects to B, through L7 to C, and to D, at the right hand end of that connector. And you say that path has continuity.

So we need to look into the positive path from E to N. The track that runs upwards from E then across to P is only used to detect the presence of adapter power, so forget that one. The track that runs downwards and left goes to F, which I think is the anode of the diode that's mounted on the other side. Current will flow through that diode to the cathode at G, then up to H and over to J, on the ON/OFF switch. This point also connects to K, the positive wire from the battery.

When the switch is ON, J is connected to M, and to N on the four-wire connector. This should complete the circuit. So to test it, turn the switch ON, plug in the adapter, set your multimeter to volts, connect the black probe to D, and follow the positive voltage from E --> F --> G --> H --> J --> M --> N. If what you said before was all correct, I think the voltage must disappear at some point in that path.

I'm having some trouble understanding the battery switching on that board. It looks like the negative wire from the battery, on the unlabelled pin, goes to the unlabelled connection on the adapter socket, which must be switched to A when the adapter is removed. And A does seem to be connected to D, the negative (0V) rail. But there seems to be copper connecting K to A, which doesn't make sense.

So if you don't get anywhere with the above test, can you clean the board by rubbing it firmly with a cotton swab dipped in isopropyl alcohol or some other mild solvent (methylated spirit is OK), rub it dry, and post a photo of the underside of that board under a very strong light, to really clarify where the copper is, and isn't. Try to avoid capturing the reflection of the light itself in the photo by putting the light or the camera at a slight angle.

 

[jh112323]

Thanks for your continued efforts Kris.

I (think) have made some progress, and believe the problem is related to the diode.

Testing the continuity from the +ve adapter input: E-F and F-G has no continuity. Each step between G-N however is fine. Testing the diode, theres only a voltage drop ~0.16V in forward bias (testing the wires, not the solder points). Having googled it, this should be higher at 0.6-0.7V? The marking on the diode reads 819 C.

I don't think that explains non-continuity between E-F though..

Soldering iron came in the post today - maybe just in time!

 

[KrisBlueNZ]

F-G is the diode, so I wouldn't expect continuity, just a voltage drop, as you measured. The diode's voltage drop would normally be around 0.6~0.7V but it's a lot lower for a Schottky diode, a special kind of diode. The "819" marking is probably the last three digits of the full part number, 1N5819, which is a very common Schottky diode. So that's all probably normal.

I was expecting continuity from E-F but it's quite possible I've misread the tracks on the board. I really need to see a good photo under strong light, after the board has been cleaned.

 

[jh112323]

Ah I see, got a bit excited there. Thanks for the info.

So perhaps E-F non-continuity is the problem, if it's taken a fall then could there be a hairline crack in that conductor? I could solder a bypass between the two?

In terms of testing voltage from D, E shows a potential difference but F doesn't - proving the same point.

I know you want a photo and cleaned board. I have attached the best I can do with a mobile phone and a magnifying glass - will try and sort some proper photos over the weekend. Sorry I don't know why they're rotated, the files on my computer aren't. From inspecting the circuitry closely, the connections all appear to be those you mentioned previously.

Out of interest, what is the purpose of the circuit between R48 and Q4?

Let me know if you think of anything else in the meantime.

Thanks

 

[KrisBlueNZ]

OK, those pictures make it pretty clear that the positive pin of the adapter socket IS supposed to be connected to the anode of the diode. If you're not getting continuity, my guess is that the adapter socket's copper pad has detached from the circuit board. This can happen if the socket is not fully flush with the board when it is soldered; a bump to the socket pushes it against the board, and the pin pushes through the board, lifting the pad slightly and breaking the copper.

If that's what happened, here's how to fix it. Desolder all three pins of the adapter socket from their pads. See whether the damaged pad is still strongly attached to the thick track that runs around to R45. If so, leave the pad there; if not, cut it off with a scalpel or razor blade. Check that the other pads aren't damaged. While pushing the socket firmly against the board, resolder the other two pads. This will hold the connector hard against the board while you repair the broken track.

Get a piece of insulated, stranded hookup wire, and cut through the insulation carefully with side cutters or a scalpel about 5 cm from the end. Pull the insulation to expose about a 4 mm length of copper strands. Hold both pieces of insulation and twist them, to twist those strands. Wrap the twisted strands once around the pin from the adapter socket and push the wire loop against the board using the side of a scalpel or side cutters. Quickly solder the connection.

Connect the two ends of the wire to the diode and the pad on R45. Cut them to length, strip the ends, and twist the wires. Tin the wires and the pads (remove the huge blob of solder from the diode's pad first) then solder the wire ends. You may want to tack the wires to the board with small blobs of glue, to keep it tidy.

I don't know what Q4 does. I haven't investigated it.

 

[jh112323]

Hi Kris,

Apologies for no recent reply, have been busy with studies.

You were right! I think the anode pad of adapter socket had fractured somewhere. Anyway I used an arduino breadboard wire (all I had) to bridge from adapter pins to their next pins - and it works! Messy job, I'm new to soldering, but it works.

Just wanted to say thanks so much for all your help and I'll be sure to keep going on the forums as and when.

Best,

JH