News:

Printed Amstrad Addict magazine announced, check it out here!

Main Menu
avatar_khaz

My 12V wire melted

Started by khaz, 23:34, 19 June 16

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

khaz

While the computer was off.

I'm using an external power adapter that outputs both +12V and +5V. It's forever connected to the mains and to the CPC. While I was busy not doing anything with the computer, I smelled burning and saw smoke coming out of the CPC. I immediately disconnected the power adapter from the wall, and that stopped the smoke, but it took me a while to locate the origin of it. It was at the interface between the +12V CPC cable, and the new cable, the two wires inside the new connector melted.

I redid the connector, but it will most likely melt again at some point. I noticed the +12V ground cable wasn't connected to the ground in the power adapter, only the one along +5V is. I supposed it was on purpose so that the circuit is closed only through the +5V ground an the drive is only powered when the computer is powered. Its a bit curious though, why even have a ground wire in the cable at all? Can I remove it safely to prevent any future fire? Or is this cable terribly made?

Bryce

If the wire melted it means that there was a short circuit somewhere in the wire or connector. Whether the 2 GND wires were used or not doesn't change anything. You need to find out where the +12V wire managed to touch the GND wire and make sure this can't happen again.

Bryce.

khaz

So it melted where the short was? Or can there be a short in some other distant place, that could in turn affect this weak spot?

If the first, then I would remove the extra ground. It's just a wire that isn't connected to anything on one end and just asking for trouble.

1024MAK

Typically, when a short circuit occurs (two conductors with a voltage difference touching or a conductive item bridging across two conductors), a high current will flow. Everywhere that this high current flows will heat up more than normal. But if there is a weak spot, where the resistance of the conductor(s) is higher (thinner wire, plug/socket/connector), this will get hotter much quicker.

So to answer your question, the short could be at the point where you found the melted insulation (as I doubt the copper actually melted), as if the two conductors (wires) were just touching, it may have been an imperfect connection with some resistance, hence it got hot. Or the problem is between the melted area and the computer. Carefully check the cable and connectors for any other damage.

Any unused wires should be insulated. I don't understand why a ground / GND / 0V wire has been left loose and unconnected.

Was your mains adaptor / power supply built / modified by someone, or is this how it was supplied when bought from a shop / supplier?

In a similar mains adaptor / power supply that I have for my CPC, I'm sure that the ground wire is connected in both leads.

Mark
Looking forward to summer in Somerset :-)

Audronic

HI Kaz


It gets a little bit more complicated as the tip for the 5 Volts is +(Positive) and the Tip for the 12 Volts is - (Negative).
and the reference to the Ground wire does not apply Photo (Rough sketch)




Ray
Procrastinators Unite,
If it Ain't Broke PLEASE Don't Fix it.
I keep telling you I am Not Pedantic.
As I Live " Down Under " I Take my Gravity Tablets and Wear my Magnetic Boots to Keep me from Falling off.

1024MAK

Yes, the polarity of the power connectors is different between the 5V and the 12V. But internally in the computer, the two supplies have the negative/ground/GND/0V sides connected to a common ground/0V network on the circuit board.

Mark

Looking forward to summer in Somerset :-)

chinnyhill10

Is this a PSU sold as CPC compatible on the likes of Ebay per chance? Usually repurposed generic netbook power supplies where somebody has rewired the end so you get the 5v and 9v for the CPC?
--
ChinnyVision - Reviews Of Classic Games Using Original Hardware
chinnyhill10 - YouTube

khaz

#7
I got it from retrocables.es. The brick directly outputs +5V and +12V, with a 4pin DIN connector. Then the +5V is routed to the tip of one connector, the +12V is routed to the sleeve of the other, and one GND runs along the +5V. There are always four wires, just at the DIN interface the other GND wire isn't connected.

I did some mods on the cables and CPC though: I modified the Scart cable so that it takes its fast switch signal from normalised sync instead of external +5V and removed the extra connector; I took +5V and GND directly from the board (one of the RAM chip legs iirc) to power an external drive, so that it switches on with the computer. That's all.

[edit] It's this one exactly. The NIMO ALM035:
Electrónica Molgar - ALM035

chinnyhill10

Quote from: khaz on 11:15, 20 June 16
I got it from retrocables.es. The brick directly outputs +5V and +12V, with a 4pin DIN connector. Then the +5V is routed to the tip of one connector, the +12V is routed to the sleeve of the other, and one GND runs along the +5V. There are always four wires, just at the DIN interface the other GND wire isn't connected.

I did some mods on the cables and CPC though: I modified the Scart cable so that it takes its fast switch signal from normalised sync instead of external +5V and removed the extra connector; I took +5V and GND directly from the board (one of the RAM chip legs iirc) to power an external drive, so that it switches on with the computer. That's all.

[edit] It's this one exactly. The NIMO ALM035:
Electrónica Molgar - ALM035


Have you tested the continuity on the adaptor side of it and then on the PSU? To determine where the fault lies?

--
ChinnyVision - Reviews Of Classic Games Using Original Hardware
chinnyhill10 - YouTube

Bryce

That's sound like it was wired ok, however, I would guess that the +12V wire somehow got to touch GND. This can happen very easily if the 12V socket has some of the outer metal ring exposed. It only has to touch off one of the metal centronics connectors or clips on the back of the CPC to cause a direct short.

Bryce.

khaz

Yes, I think his is what happened too. the two wires go through sort of a ring that you have to close with pliers to keep them from moving. Said ring must have been tightened too much and cut through the sheath, resulting in melting after prolonged use. I'll just remove the ground wire, it's useless anyway.

Bryce

That crimp ring should be connected to the outer ring of the plug/socket. So a short between the ring and the +12V shouldn't be a problem. On the 5V line the ring would be connected to GND and this would cause a short.

Bryce.

Gryzor

Jesus, thank god you were there when that happened...

Bryce

More worrying is the fact that it didn't trip anything inside the PSU, which means that the PSU has absolutely no output protection. Not something I'd want to be using on any retro computer.

Bryce.

khaz

Well it didn't completely melt, I suppose the resistance wasn't low enough to trigger any potential safety shutdown. The cable was still working, as I used it for a while while investigating what burnt (I thought the smoke came from within the CPC). I could hear some crackling noise, but only when the machine was off.

Bryce

That shouldn't happen either though. If the wire is 0.5mm copper it can take about 10A before it gets hot enough to smell or start melting. The PSU is rated at 2A, which means there should be a fuse (or equivalent shut-off circuit) that triggers at about 3A. It was obviously well above 3A and nothing happened = Not good.

If there really was any burning inside the CPC, then check the traces from the 12V header (where the wire connects to the PCB) and the tracks coming and going from this header. There is also a small pass-through transistor on the 12V line as far as I can remember. Check that this hasn't fried too.

Bryce.

Powered by SMFPacks Menu Editor Mod