Be aware of F3 polyfuse

[tomexx]

Here's the sales pitch:

This adapter was specifically designed to provide 5.25V not 5V, but we still call it a 5V USB adapter. We did this on purpose to solve a problem that occurs often with USB-powered gadgets: they draw so much current than the resistance of the cable causes a voltage drop, so instead of 5V, the device sees 4.75V or so. To avoid this problem, we made the adapter 5.25V. This is because the USB power specification allows up to 5.25V, and its safe for all 5V electronics, and has the benefit of making up for any loss due to thin USB cables. Basically, you can use it where ever USB 5V power is needed, with no risk of damage, and it will happily work at the full current range, no matter what cable you use!

This adapter ( https://www.adafruit.com/products/501 ) was supposed to give me the right voltage for my PI. It's rated at 5.25V/1A. However, I'm only getting a 4.4V-4.6V on TP1/TP2, even if I disconnect mouse and keyboard. As we know that's below the 4.75V recommended minimum.

I replaced it with a larger 5V PS and I'm now getting 4.92V.

I still like Adafruit website as it has a nice selection of accessories for your PI, but this power supply should be avoided.

Tom

[jamesh]

It may be that particular adapter is faulty, so I would contact Ada fruit and get them to replace - they are pretty good like that I've heard.

[PeterO]

Check the voltage drop across the poly fuse. The volts could be being lost there if it has a higher resistance than expected.
PeterO

[RaTTuS]

what is the USB lead like - it could be excessively long or thin

[tomexx]

@PeterO
My board came with 0ohm resistors instead of polyfuses.

@RaTTus
The PS comes without the USB cable. I bought one separatelly. I also tried the P.S. with the cable that came with my Samsung phone. Samsung cable gave me 4.6V. The one I bought separately gave me 4.5-ish as it was longer.

@Jamess
Thought about returning it but it will probably cost me more to ship it back.

I've ordered a different one about a month ago from Newark and still waiting since it's on backorder. Should be getting it soon.

Tom

[Grumpy Mike]

How are you connecting it?
I have found that the micro USB lead can have a resistance of 2.5R per wire on some types, giving a 5R total resistance on the supply. So it could just be your cable rather than the supply.

[thsBavR10]

@PeterO
My board came with 0ohm resistors instead of polyfuses.
Tom

Hi Tom, not alle fuses are replaced by 0Ohm resistors.
There is another fuse under the board, labelled F3, between SD card and USB power connector.
I had a Pi, where the resistance of F3 was 3 times greater than average (0.9 Ohm vs. 0.-0.3Ohm) - I replaced F3.

[tomexx]

Hi Tom, not alle fuses are replaced by 0Ohm resistors.
There is another fuse under the board, labelled F3, between SD card and USB power connector.
I had a Pi, where the resistance of F3 was 3 times greater than average (0.9 Ohm vs. 0.-0.3Ohm) - I replaced F3

My F3 fuse has a resistance of 1ohm. What did you replaced yours with?

Tom

[thsBavR10]

Hi Tom,
I asked at a local electronic store with smd parts for a suitable replacement, and found a
SMD polyfuse with 1.1A instead of the 0.75A type.
Soldering was not a problem due to the relative big size of the component.
For testing it is possible (on your on risk) to bridge the fuse - if your power supply is rated about 1A, the risk is limited.
simply solder a litte bit of wire over the fuse.
Where did you buy the Pi? RS/Allied? (Mine with the faulty polyfuse came from RS)

[tomexx]

Thanks for the info, that's very interesting.
I got mine from newark canada.

Soldering won't be an issue, I just hope I can find the fuse somewhere.
I think it's a 6V/0.75A

Tom

[thsBavR10]

yes, it is - but 1.1A will do it the same, with a lower resistance.

[tomexx]

I think that's exactly the problem. I'm going to buy me some fuses

One of moderators should make your suggestion a sticky or add it to the troubleshooting section.

Tom

[thsBavR10]

My opinion is that the polyfuse(s) are another elephant for the Pi,
especially when the Pi will arive at the not-so-technical audience.

The real problem may not only be the power supply, it could just as well be the fuses.
The foundation and RS/Farnell should eliminate F3 like F1 and F2.
Or the suppliers should add a multimeter in the classroom packages

[tomexx]

Absolutly, I think the PS is fine.

[thsBavR10]

[joke mode on]
maybe it could be a business idea:
selling tuning kits for the Pi,
consisting of
multimeters,
soldering irons, solder,
replacement fuses,
pavement for burn
[joke mode off]

[pluggy]

You can isolate whether its a power supply / lead problem or a polyfuse problem by measuring the voltage drop across the polyfuse. On the back of the board you can measure the voltage between The back of TP2 and the end of the polyfuse nearest the side of the board. Which is what is coming in from the supply. The other end of the polyfuse should be the same as the voltage across the test points. On mine with Just a keyboard plugged into the USB there is a 0.1 V difference. It doesn't change unplugging the keyboard which suggests my keyboard doesn't consume much juice.

[tomexx]

UPDATE.............
Well I was wrong about the adapter, turns out it's the F3 fuse... .

Thanks to the tip from thsBavR10 I ordered new F3 fuse from newark. What a difference.
My old fuse had a resistance of 1 ohm and my TP1-TP2 voltage was 4.5-4.6V
The new fuse has a resistance of 0.1 ohm and my voltage is now 5.03V (the adapter provides 5.25VDC)

So, if you have a lower than normal voltage on TP1-TP2 (less than 4.8V) or are having weird problems with your PI, check your F3 fuse with an ohmeter (with power off).

Note: this fuse is small and if you don't have much experience with soldering I would suggest to ask someone experienced to desolder the bad one and solder a new one for you.

BTW, here are the details:
NEWARK
- LITTELFUSE - 1812L110PR - FUSE, PTC RESET, 6V, 1.1A, 1812, $0.29 each
Newark Part Number: 18M4788, Manufacturer Part No: 1812L110PR

Is there a way to change my subject line? Right now it's a bit misleading...
Tom

[dom]

Is there a way to change my subject line? Right now it's a bit misleading...
Tom

Done.

[tomexx]

Thanks dom.
In the future, is there a way for thread owner to edit a subject line to add something like "RESOLVED" at the end of it to let people know that the solution has been found?

Tom

[mahjongg]

Just send any of the moderator a personal message if you want (your own) post header to be changed and he will do it for you. I sometimes will add (solved) to a thread myself if its clearly solved, but will also remove it again upon request from the original poster.

[dasankir]

Thanks for the useful information... first I read about F1 & F2, and in the absence of them in my R2 (rev. RB1244) board, I soldered some extra condensers to the board (that solved most of the resets when hot-plugging usb devices), I posted my mod here: http://therandomlab.blogspot.com.es/201 ... wn-on.html.

I have a question:

It will be hard for me to find the 1.1A polyfuse to replace the 0.75A one. According to my low knowledge in electronics, I think I could solder a resistance in parallel to F3, to reduce the resulting resistance. These are my calculations:

If the actual fuse is R1=0.9ohm and the maximum current is 0.75A, the voltage across the fuse is: V=I*R1=0.75*0.9=0.675V
If the new maximum current will be 1.1A, the resulting resistance is R=V/I=0.675/1.1=0,6136ohm
To achieve this, I can add a R2 resistance in parallel to the fuse's R1 resistance, so the resulting resistance is R= (R1*R2)/(R1+R2).
Being R1=0.9ohm and R=0,6136ohm, the formula is 0.6136=(0.9*R2)/(0.9+R2).
Extracting R2 would be: R2=(0.9*0.6136)/(0.9-0.6136) = 1.928ohm

That is the same as the easier calculation: saying that 0.35A of the total 1.1A should go through R2 and 0.75 through R1, R2=0.675/0.35=1.928ohm

The nearest resistance is 2 ohm, and it should be at least 0.6136*0.35 = 0.214 Watt, that is to say 1/4 Watt. I wonder if such small resistances are easier to find that the fuse...

But, are my assumptions/calculations right?

And, would it be safe to make the resulting resistance 0.45ohm rising the maximum resulting current to 1.5A?

[Mortimer]

If you going to put a resistor in parallel with the fuse, you might as well remove the fuse, as it will no longer be able to function in the way intended.

[dasankir]

If you going to put a resistor in parallel with the fuse, you might as well remove the fuse, as it will no longer be able to function in the way intended.

I see, didn't think about that but is quite obvious, if the fuse triggers, all the current will still flow through the parallel resistor... so my suggestion is a bit stupid

[aTao]

It will be hard for me to find the 1.1A polyfuse to replace the 0.75A one.

why difficult? Where did you buy your RPi? RS and Farnell (Element 14) sell the polyfuses.

[dasankir]

Well, I don't want to spend 6eur in delivery expenses for a 0.50eur polyfuse...

I'll look after it next week in my local electronics stores.