powering the Pi through the audio jack? NO!!!


 
22 posts
by toxibunny » Mon Jan 07, 2013 2:31 am
http://www.youtube.com/watch?v=LP-DZglWBIo WARNING, CONTAINS BAD LANGUAGE

wut?
[moderator: Updated title to avoid disaster.]
note: I may or may not know what I'm talking about...
Posts: 1084
Joined: Thu Aug 18, 2011 9:21 pm
by aTao » Mon Jan 07, 2013 2:45 am
Possible, but not a great idea...

Could wreck audio out, does not provide +5V and might well fry a protection diode or 2
>)))'><'(((<
User avatar
Posts: 439
Joined: Wed Dec 12, 2012 10:41 am
Location: Swine Town UK
by Burngate » Mon Jan 07, 2013 3:10 pm
It's a con.
Deliberate or accidental.

The connector he's using is a bog-standard barrel plug. When he pushes that in, only the outer connects to anything - the inner, be it + or -, doesn't connect to anything.
So putting that plug in does nothing.

Perhaps he's using his monitor to back-power the Pi via the HDMI, and his barrel connector is providing a ground return.

But it's still a con.
Wyszkowski's Second Law: Anything can be made to work if you fiddle with it long enough.
Brain surgery is easier than psychoanalysis
User avatar
Posts: 2727
Joined: Thu Sep 29, 2011 4:34 pm
Location: Berkshire UK
by tzj » Mon Jan 07, 2013 3:27 pm
He's just grounded the right output, which as far as I can tell shouldn't do anything...
http://www.themagpi.com <---- Checkout the MagPi
(The MagPi - Co-Founder)
User avatar
Posts: 300
Joined: Thu Jan 05, 2012 3:09 pm
Location: Barnsley, SY, UK
by rurwin » Mon Jan 07, 2013 4:10 pm
He'll ground everything, including Ground. All three wipers in the socket will bear against the barrel of the plug.

I'd agree; he's back-powering through HDMI and the power supply happens to earth it's ground output. Exactly how it all works when the plug is in and doesn't when it's out would take an electronics guru to unravel. But one thing is clear: he's not powering it with that power supply, because there is no central pin in that socket and therefore nothing connects to the supply positive.
User avatar
Forum Moderator
Forum Moderator
Posts: 2903
Joined: Mon Jan 09, 2012 3:16 pm
by mahjongg » Mon Jan 07, 2013 4:17 pm
Don't know how he does it, and I don't care, he probably just flips a switch simultaneous with plugging the mini-barrel plug in.

a total fake!
User avatar
Forum Moderator
Forum Moderator
Posts: 5181
Joined: Sun Mar 11, 2012 12:19 am
by redhawk » Mon Jan 07, 2013 4:49 pm
It's a bit early for April 1st videos :lol:

Inserting the power connector would only provide a ground path the +5v must be coming from somewhere else maybe a wire hiding under the HDMI cable to the circuit board.
While the video itself probably isn't fake his claims of powering the Pi using the headphone socket is pure nonsense.

Richard S.
User avatar
Posts: 3404
Joined: Sun Mar 04, 2012 2:13 pm
Location: ::1
by mahjongg » Tue Jan 08, 2013 12:38 am
why would he go to the trouble of wiring things so that the barrel connector closes a ground loop, as all that he, (or anyone assisting him) has to do is to turn on the real power supply simultaneous with him plugging in the barrel connector. IMHO the pi is powered through wires soldered to the bottom of the PI, and running through the holes of the raster that he placed the PI so conveniently on! Thats why he can say he can do it without HDMI plugged in!

Notice he strategically obscures the LED's on the PI with his hand in the video, while "plugging in", so there is no way to tell that the PI really turns on exactly the moment he "plugs in".

I'm not saying he may not have used the ground sleeve to close the loop, just that he didn't have to, and probably he isn't smart enough to think of this trick either!
You can tell he's not really technically minded, as he calls the 3.5mm audio port the "RCA port" in the video, but the RCA port is the video output coaxial connector, not the audio output connector! Stupid mistake!

he's just an attention seeking troll, that tries to trick people into doing something that may damage their PI's audio connector, while trying to accomplish something that cannot work.
User avatar
Forum Moderator
Forum Moderator
Posts: 5181
Joined: Sun Mar 11, 2012 12:19 am
by GChriss » Sun Jan 20, 2013 11:41 pm
There's some truth to this one -- I was able to partially power the Model B board using a 3.5mm stereo male <-> dual RCA male cable, no other cables (i.e., microUSB, Ethernet, HDMI) attached.

'Partially' powered means the cables were sparked against a 5V USB power source resulting in partially-lit LED power lights. Some real soldered connections and we might be in business.
Posts: 5
Joined: Sun Jan 20, 2013 11:35 pm
by mahjongg » Mon Jan 21, 2013 12:18 am
GChriss wrote:There's some truth to this one -- I was able to partially power the Model B board using a 3.5mm stereo male <-> dual RCA male cable, no other cables (i.e., microUSB, Ethernet, HDMI) attached.

'Partially' powered means the cables were sparked against a 5V USB power source resulting in partially-lit LED power lights. Some real soldered connections and we might be in business.

Stop promoting this stupid and dangerous idea!

There are diodes to clamp the output signal to 3,3V so what you are doing is injecting 5V into the 3V3 supply! If you are not very careful you will put 5V on the 3V3 supply and blow up your PI!
Don't you get that this is a a very dangerous thing to do, and was started as a hoax by someone who didn't know what he was doing? :evil:

Even you he manage to put some power into the PI, you still don't end up with a really functioning PI, as that power can never reach the USB ports, so you can never control the PI this way. And prolonged exposure to 5V will destroy the Broadcom and ethernet chip eventually. If the protection diodes don't overheat and blow first.
User avatar
Forum Moderator
Forum Moderator
Posts: 5181
Joined: Sun Mar 11, 2012 12:19 am
by GChriss » Mon Jan 21, 2013 4:54 am
OK, I'll grab a 5V barrel connector and report back.
Posts: 5
Joined: Sun Jan 20, 2013 11:35 pm
by rurwin » Mon Jan 21, 2013 7:46 am
A fool and his money are quickly parted. In this case $35 plus tax and postage.

Let me save you a job. It may work... for a while. You probably wont have any USB devices because you don't have any 5V power, just over-driven 3.3V. And then something will over-heat and it will all stop working. Permanently. If you're very, very lucky you may see the magic smoke as it escapes.
User avatar
Forum Moderator
Forum Moderator
Posts: 2903
Joined: Mon Jan 09, 2012 3:16 pm
by GChriss » Tue Jan 22, 2013 3:47 am
Using a generic 2-Amp USB battery pack wired to a TRS 3.5mm audio jack as follows I can confirm the Pi powers just fine without any discernible damage. This is particularly helpful when you need to "walk" the device from one Micro-USB power source to another without rebooting the device.

Tip <-> +5V USB
Ring <-> Not used
Sleeve <-> USB Ground

Notes:
-rurwin is correct in that Pi USB power isn't maintained when running via 3.5mm jack power. Thus, any USB devices attached to the board will disconnect/re-initialize according to the mains rail power status.
-When running via 3.5mm jack power attaching the Micro-USB cable to a free-floating power adapter (e.g., an Apple USB Power Adapter) will cause the device to reboot due to temporary undervolting.
-Even after playing audio into the reverse-powered 3.5mm jack analogue audio-out works just fine afterwords.
-There's no dropout in HDMI during power supply changes.
-The Pi can do all of this while in the middle of compiling the hello_pi examples -- no problem there.

I haven't tested GPU performance, overclocking, or long-term stability. Trying any of this may void your warranty.
Posts: 5
Joined: Sun Jan 20, 2013 11:35 pm
by GChriss » Tue Jan 22, 2013 4:27 am
Ethernet works without issue as well.

While on the topic, it may be possible to power the Pi over PoE using something akin to the following:
PoE to USB / IT-CHRG-P2U/IT-WPCHRG-P2U
http://www.fsrinc.com/products/it-chrg-p2u.htm
Posts: 5
Joined: Sun Jan 20, 2013 11:35 pm
by rurwin » Tue Jan 22, 2013 7:51 am
GChriss wrote: -When running via 3.5mm jack power attaching the Micro-USB cable to a free-floating power adapter (e.g., an Apple USB Power Adapter) will cause the device to reboot due to temporary undervolting.

It isn't undervolting, not unless it managed to draw all 2A from your supply through the 3.3V regulator and the audio protection diodes. I wonder if the 3.3V regulator can signal an over-voltage shutdown.

PoE
Will the madness never end?

Why does this thread remind me of street drag-racing and doing handstands on the rails of hotel balconies?
User avatar
Forum Moderator
Forum Moderator
Posts: 2903
Joined: Mon Jan 09, 2012 3:16 pm
by GChriss » Tue Jan 22, 2013 9:14 pm
It isn't undervolting

I'm pretty sure it's undervolting.

Will the madness never end?

This is creative powering, that's all.
Posts: 5
Joined: Sun Jan 20, 2013 11:35 pm
by daveake » Tue Jan 22, 2013 11:11 pm
"Creative" as in likely to create a broken Pi.

You're applying 5V to a pin that has a protection diode to the 3.3V rail. A diode with an Imax of 125mA, and a Vf of around 1V at the sort of current it'll be passing. So you're going to over-voltage the 3.3V rail, meaning that exceeding the diode's Imax is going to be the least of your problems. Good luck with that.

Dave
User avatar
Posts: 62
Joined: Thu Jul 12, 2012 12:07 am
by mahjongg » Tue Jan 22, 2013 11:46 pm
Also, lets not forget that the initial suggestion is still a hoax, as there is no way that a barrel connector can do anything when stuck into a 3.5mm jack, as there isn't any way to make contact with the inside of the barrel. Which is lucky as he is proposing putting 5V in, and if you succeed somehow to put 5V on the 3V3 supply its will blow eventually The video is just a lie!

Also, yes its very possible to damage your PI, even when you succeed in putting exactly 3V3 on the 3V3 line (and you would need to put a diode drop (0,5V) more in to do that! When you do, and you do not connect a 5V supply then the NCP1117-3V3 voltage regulator (RG2) has no input voltage, and a 3,3V forced output voltage, which isn't a good scenario for a low drop regulator, you will damage it eventually. In fact in some schematics where this could happen you always see a diode between the output an input of the regulator so that if the input voltage is somehow lower than the input voltage the diode makes sure that the input is at least at (almost) the same voltage as its output. See figure 25 of the datasheet of the device, which can be found here:
http://www.onsemi.com/pub_link/Collateral/NCP1117-D.PDF for an example of this diode used.
Lastly its possible that when you do this you are trying to put in excess of 500mA through a very small (double) diode, the BAV99, which is normally rated at just 100mA or so!


If you are desperate for an alternative way to power the PI, (when transporting it, yeah, right... :twisted: ) then I suggest putting in 5V via the GPIO pins!
User avatar
Forum Moderator
Forum Moderator
Posts: 5181
Joined: Sun Mar 11, 2012 12:19 am
by MattHawkinsUK » Wed Jan 23, 2013 10:51 am
Lets see him do his trick with no HDMI connected and a clear shot of the underside of the board.

We won't see this because it would show that there is no current flowing from his 5v supply into the audio jack. It's a lie/hoax/joke/scam. Take your pick.
My Raspberry Pi blog and home of the BerryClip Add-on board : http://www.raspberrypi-spy.co.uk/
Follow me on Google+, Facebook and Twitter (@RPiSpy)
User avatar
Posts: 483
Joined: Tue Jan 10, 2012 8:48 pm
Location: UK
by jamesh » Wed Jan 23, 2013 11:17 am
OK, enough is, as they say, enough.

The original OP was a hoax, subsequent posts are trying to do something that will, in all likelihood, destroy your device.

So, as a final warning DO NOT TRY ANYTHING IN THIS THREAD AT HOME UNLESS YOU WANT TO THROW AWAY YOUR LONG WAITED FOR DEVICE AFTER BLOWING IT UP.

Locking thread.
Soon to be unemployed software engineer currently specialising in camera drivers and frameworks, but can put mind to most embedded tasks. Got a job in N.Cambridge or surroundings? I'm interested!
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 11617
Joined: Sat Jul 30, 2011 7:41 pm
by jamesh » Thu Jan 24, 2013 10:58 am
Just a quick addition to the thread. After speaking to Gert, he has confirmed this is a very silly thing to attempt, and will more than likely destroy your Raspberry Pi.

DOING THIS WILL BREAK YOUR RASPBERRY PI AND WILL VOID YOUR WARRANTY.
Soon to be unemployed software engineer currently specialising in camera drivers and frameworks, but can put mind to most embedded tasks. Got a job in N.Cambridge or surroundings? I'm interested!
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 11617
Joined: Sat Jul 30, 2011 7:41 pm
by mahjongg » Sat Jan 26, 2013 11:16 am
Also to quickly comment on the OP's "latest youtube video".
He seems to be happy enough to destroy a PI to continue his hoax!

What you see him doing is putting 5V on the 3,3V supply, (probably he modified the tip of the plug, otherwise it normally wouldn't make any contact). And so he is putting 5V on the 3,3V supply which will obviously light up the power LED which is directly powered from the 3,3V line, now carrying 5V. Actually therefore the LED will shine more brightly than normal, as it will get about twice as much power.

But at the same time he may manage to damage the rest of that PI, probably including the SD-card which is also 3,3V powered, and certainly isn't designed to withstand prolonged exposure to 5V!

thats probably why it doesn't boot anymore!
User avatar
Forum Moderator
Forum Moderator
Posts: 5181
Joined: Sun Mar 11, 2012 12:19 am