Mark 2 power failure?

Yesterday I tried to use my spare Empeg (Riocar Mark 2A) again after a year of non-usage. It does not power up at all (no display messages, no hard drive starting up). The network light on the back is flashing blue very fast, same for the lights of the lighting buttons on the front (very dim though).
I checked the power supply, it works OK with the other Empeg so the supply seems OK to me.
I connected it via serial to Hyperterminal, nothing at all displayed.

Any suggestions what to check and/or how to fix?

Jelle

Unplug the drive and try again.

Forgot to tell, already tried that, with same result.

The power supply is a known-good power supply of the correct type?

There isn't a physical problem inside the unit like a loose screw that got stuck against a critical power system component?

Display board cable is plugged in correctly (i.e., not off-by-one-pin)?

Thanks for all the input. Just tried to check a few points mentioned, however with 2 kids around that didn't work out that easy - it will have to wait until the weekend:)

I think I've found a reasonable explanation, see picture below. Cause unknown, might be (as Tony indicted) caused by some loose metal part inside?

Any suggestions to fix, or is replacement Empeg a better alternative?



Jelle

Wow.. looks like you have/had a wiring issue with the in-car docking sled.

I don't have a Mk2a unit opened up here right now, but I do have a Mk2 open --> the black/burned component at far left is a 1.2Kohm resistor in my Mk2, and the blown-away item was a 0-ohm resistor.

EDIT: MMmm.. in my Mk2, those four contacts (the two for each burnt part) are all earthed/grounded. Which, to me, suggests that the in-car wiring may have been wired backwards, trying to feed +12V through the ground wiring.. BOOM!

Cheers

If the damage (burning/melting) extends inside the multilayer circuit board, then it will be very tricky to repair.

You could try the simple thing of replacing the two obviously blown components, and then powering it up using the indoor AC-adapter. Most likely, they will blow again, but you could get lucky, and have almost nothing to lose.

If that doesn't fix it, then I recommend putting the unit aside for use as "spare parts", and trying to purchase a replacement unit. This should be cheaper than paying for a difficult repair job, and the unit may be "beyond repair" already regardless.

But first I'd try the suggestion of replacing the two blown parts, just for fun.

-ml

Something else to do first --> get some mostly-pure alcohol, or similar cleaning fluid, and swap away all of the burn marks in the area. If the circuit board cleans up really well that way, then the damage may not be very bad.

But if it stays very black in colour, that's an indication of possible major damage.

Cheers

Heh.. after examining my own Mk2 even more closely, I now believe it was assembled incorrectly --> the 1.2Kohm resistor should probably have been another zero-ohm resistor.

OK, I have to ask. Isn't a 0 ohm resistor like, a wire?

Yup. Its handy to have it in a regular resistor form factor for when you need to bridge two points when you're doing automated assembly.

In this case, more like a fuse.

Hi,

I checked one of my Mk2a boards and it has a 1.2k (marked with a 122), and a zero-Ohm (marked 000), in the locations that Mark originally had indicated. The 1.2k is closest to the screw hole, the Zero-Ohm is next to it on the right.


In case you were wondering, the 122 marking on the resistor is 12 with 2 zeros. So that would be 1200 Ohms. There are 1000 Ohms in a k-Ohm. So the resistor has a value of 1.2k Ohms.

The same can be said for the larger yellow "block" looking solid Tantalum capacitors. A marking of 226 is 22 with 6 zeros. That is in pico-Farad (pF). So it would be 22,000,000 pF. There are 1,000,000 pF in a micro-Farad. So the capacitor has a value of 22uF. I know, geek stuff, get back on topic...


Mark is right, clean it up, snap another picture and maybe there is hope.

After the board is cleaned up, some measurements can be made with an Digital multi-meter to see some of the extent of the damage.

Since there isn't information on the layer stack-up and trace routing of the Printed Circuit Board, if it has caused excessive heat and has gone through the top layer to internal layers, it may not be fixable. Not that anything in that area is likely to be controlled impedance, but if the resin/glass weave separated (exceeding the Glass Transition (Tg) of 120 to 160 Deg C.), and allows adjacent layer routes to touch, it could be flaky, intermittent, or just a complete short. If the layer information was known, sometimes you can drill or open the traces (depends if there are planes - it looks like there is, you can't see through the board - in the same area though), and use wires to correct it.

Too bad that kind of data isn't available, it could be useful in these situations.

Keep hope, after you clean it, we'll know more, there's still a good chance it might be minor.


Ross

And good board designers add a liberal dose of them on most I/O circuits, so that individual parts of the circuitry can be disabled/corrected easily in the event of a design/layout error.

Cheers

Or possibly caused by a problem in the car wiring... a loose or chafed wire, a crimp that came undone and got grounded, or perhaps even the sled connector is having trouble. If you get the player repaired, make sure to check for these possible issues before you plug it back into the sled.

Interesting. I couldn't quite get my head around why you would want to interrupt a trace with a more expensive component that doesn't DO anything. But that makes perfect sense now. Modular hardware design. Neat.