We’ve got a Pi down … I repeat, a Pi down …

[This website served by the Raspberry Pi, purchased from element14. Stock is out at the moment 🙁 It would have been served by the Raspberry Pi from RS – if not for the problem]

Okay, so it’s been a week since I e-mailed the RS guys ([email protected]) in disgust at the defective Raspberry Pi they shipped me. For the record, this is what I sent them …

Dear Raspberry Pi Team,
I have received my Raspberry Pi as of 25th October. At long last, after 17 (!) weeks of waiting.
But all is not as sweet like raspberries.
I’m having many issues with stability – after 5 minutes of working in LXDE, the device would black-screen, the ethernet lights will go off and the SD Activity light will stay ON/OFF and remain stuck.
Three different SD cards (all from the known-good list – Sandisk 2Gb SD, Kingston 16Gb SDHC Class 4, Patriot LX 32Gb SDHC Class 10) were tried, three different power supplies (all from the known-good list – Apple 10W USB Charger for iPad, Asus Nexus 7 2A charger, 5000mAh emergency charger) were tried, the new kernel and boot.elf were tried, two different micro USB cables were tried, and the board was tried with and without the case with no change in behaviour. The board is only warm to the touch, and is not overheating. For the record, and for clarity – the GPIO pins have not been used, and overclocking was not enabled either, no additional peripherals except a monitor, keyboard and mouse were attached. Reimages were performed with every combination (to eliminate any carry-over corruption) with no changes – crashes happened before I could even use sudo apt-get install to download all the packages I require and as a result I have been able to perform zero work as yet.
Upon consulting this post (http://www.raspberrypi.org/phpBB3/viewtopic.php?p=200351#p200351), it appears that your boards DO have problems with their polyfuse. I have measured the voltage between TP1 and TP2 – with the USB adaptor giving 5.00v, in an X session, the voltage over TP1 and TP2 is only 4.68v. The voltage is dropped over the polyfuse – this is confirmed when I measured the voltage over the polyfuse which was hovering anywhere between 250-350mV. This was with NO additional USB peripherals except a USB keyboard and mouse.
I have spent a day tracking this down, and I am utterly disappointing that you would ship several devices to people with this kind of hardware fault which is intermittent and difficult to trace. You have inconvenienced many who have gone to try and find a USB adaptor which would be stable enough, when the fault is NOT with the USB adaptors. In fact, I implore you to check your stock right now. Test them, don’t ship rotten Pi’s out as it does no good for your brand, and it does no good for the Raspberry Pi Foundation either. I am an engineer myself, imagine what other non-engineer/non-technical types could/would do? They may encounter this problem and give up on the whole concept altogether over a lousy polyfuse!
I hate to say this, but most of my friends in Australia have chosen element14 as their supplier, and have not have any of the wait time issues (4 weeks for them) which I have had to endure, nor the intermittent hardware issue described above.
As such, I am asking you – how do you propose to remedy this “fault”? If I am to return this package to the UK, the cost to me is approximately 3/4 the cost of the RPi without any form of tracking, and exceeds the cost of the RPi if tracking is required. Likewise, returning it to the Australian arm of the RS Components chain would cost me 1/2 the cost of the RPi, all of which I find entirely unacceptable as the fault stems from your actions. In fact, the amount of time spent troubleshooting and tracking the problem, and the loss of one of my SDHC cards (due to plastic fatigue from the constant plugging/unplugging/reimaging and retrying) have already put me at a loss.
Your response is much anticipated.
Yours Sincerely,
Gough Lui.

It is clear that RS Components only has contempt for their customers by making them wait for a long time for their Raspberry Pis and then … well … shipping them defective ones. Since my post, three more people have posted with identical problems – a low TP1-TP2 voltage, with most of the voltage dropped over the polyfuse F3 (0.3v or more, the element14 one often has < 0.1v).

Not being the kind to be patient and wait any longer (one week is enough!), I went ahead and replaced the polyfuse with a wire.

And that solved the problem of having a low TP1-TP2 voltage. What it didn’t solve was the random black-screen with ethernet down.

Questioning my power supply, I used a 20A Lab Supply putting out 5.02v to the GPIO header directly. Stable, clean power. The supply metered a 0.5A draw when booting but as soon as it blackscreened, it was down to 0.2A. There was no anomalously high current draw which would suggest a static-damaged CPU, and there wasn’t any heat to speak of.

And so ends the life of this Pi – so tragically short, having done no work of any use. Its warranty voided, it’s parents unsupportive, its existence a mere frustration to its owner. A lost cause.

RS Components – you really failed me this time.

[For the visitors – you might be interested in: RPi element14 vs RS, How to set up an RPi (properly)]

About lui_gough

I'm a bit of a nut for electronics, computing, photography, radio, satellite and other technical hobbies. Click for more about me!
This entry was posted in Raspberry Pi and tagged , , . Bookmark the permalink.

3 Responses to We’ve got a Pi down … I repeat, a Pi down …

  1. Sadly I’m having the same problem after waiting 4 months to get mine delivered to Brazil. :'(

    Also wrote a post-mortem about it: http://nathancampos.me/post/36171845465/my-raspberry-pi-post-mortem

  2. viktor says:

    Well you can hack the polyfuse off or use a different charger. I had many stability issues. Now I use an apple ipad charger (puts out 5.2V) and no more problems. Try to find a higher voltage power source and it will fix your problems. (well don’t deviate from 5v too much) I’d rather keep the fuse, I already triggered it once by being clumsy. Once triggered its voltage drop is even higher for a couple days (it takes a couple days to recover) so the higher voltage usb supply really helps with that too.

    • lui_gough says:

      This particular batch of RPis DID have problems. Bypassing the polyfuse and even using a 20A Benchtop Lab Power Supply doesn’t help. I’ve tried 10W iPad chargers and Nexus 7 chargers with it too – it DOES NOT fix the problem. I repeat, it DOES NOT fix the problem.

      The same power supply and hardware works fine for my RPi I just got from element14 – with the same image and the same software. You can’t explain this with power issues because it wasn’t – it was more likely a hairline crack on a poor solder joint because it only fixed itself after it was baked in an oven.

Error: Comment is Missing!