Vinculum Vmusic2 just exploded!!!

Hello,

I'm currently working on a project using the Vinculum Vmusic2 module to play MP3 files to accompany a PICAXE controlled fountain display. I have recently finished building my circuit, and switched it on, only for the Vmusic2 to literally explode.

It was connected to a pair of 8R, 10W speakers (http://tinyurl.com/ll6zjlz), one on each line out, with a common ground, a 5V supply, regulated down from 12V, my 0V rail and an output from my 40X2. I believe I can discount a fault within the module, since I have tested it on its own, on a breadboard, and also a supply voltage above 5V, because my PICAXE and OLED, both 5V, still work perfectly. I also wasn't using the module at the time, there wasn't even a USB stick in it!

Does anyone have any ideas what might have caused this and how I can stop it happening to my replacement module?

Thanks in advance.

Welcome to the forum and we are sorry to hear it's under such unfortunate circumstances.

It could potentially be a power supply or other wiring issue. I am not sure the line out signals are designed to drive speakers directly and this may have overloaded the module.

Line out is meant for high impedance load 2 * 8ohm speakers could draw way too much current

Connecting 8ohm speakers to a lineout is not normally recommended see http://en.wikipedia.org/wiki/Line_level for details. However a decent circuit should tolerate this and would certainly not be expected to explode.

Did you connect the lineout GND to the logic GND?

Check whether the lineout GND is connected internally in the new unit to the logic ground. If not don't make an external connection. Many single supply chip amps float their output common ("GND") at VCC/2 and the datasheet will say to leave this floating.

Based on a quick scan, the datasheet for the Vmusic2 doesn't say anything but the circuit diagrams don't show the lineout ground connected to anything other than the speakers, if you open up the dead unit you may be able to see which amp chip they use and look up its datasheet.

Can't think of anything else obvious

matherp said:

Did you connect the lineout GND to the logic GND?

Check whether the lineout GND is connected internally in the new unit to the logic ground.

Click to expand...

There are three "GND" connections on the VMusic2 and all seem to be connected together when I check those with a resistance meter.

From the audio data http://www.vlsi.fi/fileadmin/modules/vs1000mod.pdf

Line outputs are AC-coupled, so they can be connected directly to an amplifier.
 Headphone outputs (including common buffer) are DC-biased to 1.2V. Do not connect the common
buffer output to ground! Do not connect Left/Right headphone outputs directly to an amplifier!

JB515 said:

I have recently finished building my circuit, and switched it on, only for the Vmusic2 to literally explode

Click to expand...

Perhaps you can tell us what exploded (Ie component). I presume you don't have total destruction and a crater on your workbench.

Thank-you all for the suggestions.

JimPerry said:

Line out is meant for high impedance load 2 * 8ohm speakers could draw way too much current

Click to expand...

Presumably this could be solved with a small resistor in series with each loudspeaker. To clarify, I don't have the line out ground connected to anything other than the two speakers.

inglewoodpete said:

Perhaps you can tell us what exploded (Ie component). I presume you don't have total destruction and a crater on your workbench.

Click to expand...

Sorry, I meant to upload a photo of the board.



As you can see, what I'm assuming to be a capacitor (C5) has opened up. There was also some suggestion that something black next to the 3.5mm jack was damaged, but I can see neither debris from this or a space where something should be. The other side of the board looks perfectly fine.

JB515 said:

As you can see, what I'm assuming to be a capacitor (C5) has opened up.

Click to expand...

There is a circuit diagram of the VMusic2 here ...

http://www.ftdichip.com/Documents/Schematics/VNC1/VMUSIC2 Schematic Prints.pdf

C5 is up at the top left near the USB socket, 47uF 10v. That seems to be confirmed to be across 5V0 and 0V when I continuity check on an actual VMusic2 PCB.

The explosion of C5 would suggest either a large over-voltage or perhaps reverse connection of voltage. It would definitely point to some sort of power supply issue.

In post #1 you wrote; "I believe I can discount a fault within the module, since I have tested it on its own, on a breadboard, and also a supply voltage above 5V". It could well be that testing it above 5V has caused the damage.

hippy said:

In post #1 you wrote; "I believe I can discount a fault within the module, since I have tested it on its own, on a breadboard, and also a supply voltage above 5V". It could well be that testing it above 5V has caused the damage.

Click to expand...

Ah, sorry. I meant that I had tested it on a breadboard, and that I could discount the issue of a >5V supply voltage since my PICAXE and OLED were functioning normally after the explosion. I was using a 5V bench supply during tests.

Difficult to tell from the picture but there doesn't appear to be a bar on what is left of the positive end of the capacitor. It could be that is was a manufacturing fault with the capacitor placed backwards, in which case simple replacement may correct the problem

It looks like the bar is there on C5 to me, and at the correct (left) end.

Yes, the photo isn't amazing but there is a bar on the left of the capacitor.

One other thought - I was using several large (I think 4700uF) capacitors for smoothing because of the high, and variable, current draw from my lighting (0-2.4A) and motors (1.2-4A). If this is looking like an issue with the supply circuit, could these have overloaded the module? They were connected between 5V and ground.

I would not have thought high value reservoir capacitors would have had an adverse effect. C5 is between 5V and 0V rather than in-line on the 5V rail, and it is difficult to see how any large amount of current could have been dumped through C5 or the reservoir caps would charge to substantially more than 5V. That's not to say it isn't possible.

FB1 seems to be a fuse between the 5V supply and C5 +Ve which would be expected to blow or trip if substantial current were drawn but there are no details of what that fuse type or rating is in the schematic.

I am not however an electronics engineer and perhaps someone who is, or better understands capacitor failure modes, may be able to shed some light on what the outcome suggests could have been the cause.

Most of the VMusic2 circuity is powered from regulators from 5V so, if only C5 were damaged it might still be possible to operate the module standalone from a 5V supply. FB1 may need shorting out or replacing if blown open circuit. It depends if what destroyed C5 also damaged anything else.

hippy said:

The explosion of C5 would suggest either a large over-voltage or perhaps reverse connection of voltage. It would definitely point to some sort of power supply issue.

Click to expand...

C5 looks like a SMD tantalum electrolytic capacitor - in my experiences, the normal way to explode them is to reverse the supply voltage

From my experience, the most likely cause of an electrolytic capacitor exploding is being connected to a reversed power supply. I know from experience going back 40 years. Yep: I was a spotty faced teenager once, who needed to take more care when assembling projects.

The capacitor heats fairly slowly (usually 10 seconds to a minute) until the electrolyte boils. Then, ahem, the pressure gets released. The fuse may not blow unless the plates short together after the explosion.

I'd clean up the board with a cotton bud and alcohol and replace the capacitor. At least it's fairly big if you're not familiar with surface mount components. Of course, you don't have to replace it with a surface mount component: the pads should be big enough to take a pigtailed electro. Of course, other components, esp. chips, may have been damaged too. It won't take much to find out.

Rats - beaten by Martin.

hippy said:

Most of the VMusic2 circuity is powered from regulators from 5V so, if only C5 were damaged it might still be possible to operate the module standalone from a 5V supply. FB1 may need shorting out or replacing if blown open circuit. It depends if what destroyed C5 also damaged anything else.

Click to expand...

Yes, the module will power on, but it doesn't play anything. Unfortunately, I won't be able to replace the capacitor until I can get back into school next week.

JB515 said:

Yes, the module will power on, but it doesn't play anything.

Click to expand...

That could be because something in the module has been damaged and possibly even the PICAXE which is trying to control it. It could simply be that some required connection has been left out after trying to figure out what happened.

You may have to double check everything is as it should be, the connections are correct and the PICAXE is sending commands to determine what the state of the module is.

Don't connect the speakers; use headphones to check if audio is working.

Unfortunately, I breadboarded the module on its own with a new PICAXE 20X2 and the same code (I changed the pin number) that worked when I first obtained the module. I was using headphones to test, and did get a low level 'ticking' sound before and after sending the play command.

hippy said:

I would not have thought high value reservoir capacitors would have had an adverse effect. C5 is between 5V and 0V rather than in-line on the 5V rail, and it is difficult to see how any large amount of current could have been dumped through C5 ...

Click to expand...

Actually, it IS possible that large reservoir caps may have contributed, though I would suspect that reverse polarity is the main suspect.

Tants are a very common decoupling device due to their low ESR.
In a circuit powered via a regulator or long leads the charge current is limited.
If the cap is charged (possibly intermittently) via a large capacitive source, the charge current can be huge.
Boards which are hot-swapped in large rack systems often pop their decouplers due to the unlimited current available from the other boards.
That is why I suggest tants are NOT used for decoupling on boards that may be subjected to hot swapping or power-up via relays.

I'm currently working on a project using the Vinculum Vmusic2 module to play MP3 files to accompany a PICAXE controlled fountain display.

Click to expand...

I'm late to this thread, but just thought I would mention that you would just love the Jerudong Park Musical Fountain. It is a bit like what you are doing, but a little bigger. Where unlimited money meets high tech. Jerudong Park is in Brunei and the fountain was paid for by Jefri, the brother of the Sultan. The Park is a cheap entry Alton Towers-style setup, though rather faded of late. The Fountain is an area of water roughly the size of a football pitch, and stuffed with nozzles and lights. Somebody (I think a UK software house) had a ball developing it. Perhaps it will give you some inspiration!

http://www.youtube.com/watch?v=5fjRFW9x0j0

geoff07 said:

Where unlimited money meets high tech.
http://www.youtube.com/watch?v=5fjRFW9x0j0

Click to expand...

The Bellagio in Vegas has a show like that too. It would be a blast (pun intended) to build.



These are pretty neat too. https://www.youtube.com/watch?v=f5WliWhBFHo

Wow, those are impressive! I actually started thinking about bulding a 'home version' when I saw this slightly more modest setup on the Isle of Wight:https://www.youtube.com/watch?v=1EsEk7vT3b0.

Back to the issue of the Vmusic2, I replaced the capacitor and after a bit of fiddling around (I disconnected the smoothing caps and put a 100R resistor in series with each speaker), did get it playing music again.

Thanks for the help.