I have an Arduino Mega 2560 board (R2) and Ethernet Shield R3. My problem is, the Wiznet W5100 chip on the Ethernet Shield gets insanely hot, it actually burns after touching for 3 seconds.. And that is running from USB power supply.
So my question is, what the hell?? Is that supposed to be normal?.. I don't want it to burn down my house.
Any suggestions would be greatly appreciated. Ow and btw, i bought the mega board from dealextreme, so is it possible the problem is with the arduino board instead of the shield? Or maybe because the shield is meant for R3 instead of my R2 board?
You might want to check the underside of the Ethernet Shield and the Mega to see if you have anything on the Shield touching something it shouldn't. The only other thing I a think of is the R3 SDA and SCL pins shorting on the DFU ICSP header of the Mega. Also, try spacing the Shield a bit (since the pins on the shield are VERY long).
@spcomputing
I've tried your suggestions to no avail. Nothing seems to be touching, SCL and SDA pins don't seem to be shorting with DFU header. Thank your for your advice!
@SirRich
I have not tried my R3 shield with a R3 mega board. That's not really an option, since i would have to buy another board, which in my banana republic costs 60€ And there's no guarantees that it would work.. That was my first thought too tho, maybe i'll get a chance to try it out.
The chip gets hot very quickly, half a minute or so, after a minute it's pretty much untouchable..
I'm guessing i can't just wire up the IOREF pin and SCL and SDA pins with appropriate pins on my board? I'm guessing those SDA and SCL on mega board aren't the ones as the additional ones on R3?..
As far as i've been reading up on this issues, the eth shield seems to be drawing A LOT of current, 150 mA or so, maybe there's a way to limit that current with a resistor or something so the shield would still work? I'm not sure how i would go about doing that tho..
valekovski:
As far as i've been reading up on this issues, the eth shield seems to be drawing A LOT of current, 150 mA or so, maybe there's a way to limit that current with a resistor or something so the shield would still work? I'm not sure how i would go about doing that tho..
Attempting to limit supply current is asking for trouble. Indeed the W5100 datasheet indicates 138-146mA typical operating current, 183mA max. It could therefore be dissipating nearly 600mW, which means that a package the size of the W5100 will get quite hot. As James has observed, this in itself is not a issue. Suggest concentrating on other aspects of the project that may be issues, as any attempt to address this perceived "problem" will likely just cause more.
Hm, ok.. Thanks for the advice. Let's say burning down my house isn't a likely outcome
I know this isn't the right thread for this, but still the problem exists with using a temperature sensor, the heat from eth shield will set it off.. I'm thinking of isolating the temp sensor as much as possible on the pcb, but if i wrap everything up in a box, the heat will accumulate over time. Also, i just want to point out that for example ARM's ethernet controller gives off little to no heat, neither does the MCU and the prices are comparable now. This points to bad design of Arduino. Sorry, i don't wanna be a jerk and i know it's noone's fault here, but i just wanted to point it out..
valekovski:
Hm, ok.. Thanks for the advice. Let's say burning down my house isn't a likely outcome
I know this isn't the right thread for this, but still the problem exists with using a temperature sensor, the heat from eth shield will set it off.. I'm thinking of isolating the temp sensor as much as possible on the pcb, but if i wrap everything up in a box, the heat will accumulate over time. Also, i just want to point out that for example ARM's ethernet controller gives off little to no heat, neither does the MCU and the prices are comparable now. This points to bad design of Arduino. Sorry, i don't wanna be a jerk and i know it's noone's fault here, but i just wanted to point it out..
Thanks again for your help. Cheers, Val
Back in the day our electronics gadgets were relatively large (often wooden) boxes filled with red-hot elements glowing inside glass envelopes. These ran just a tad warmer than a W5100 chip but I don't remember an unduly large number of house fires attributed to them. The point about the temperature sensor is a good one, obviously appropriate steps to mitigate heating from other circuitry are warranted. As an electrical engineer, I'd be the last to call the design of Arduino bad.
I haven't looked at the library yet, but this pin4, pin10 pulled high while idle, is in the Ethernet library, or does one have put these in the sketch?
Hmm ok, thanks Jack, i'll think about it! For starters i can at least put the temp sensor on one of the edges of a pcb, and mount it on a socket to lift it a bit, i'll see what that does, tho if i put a box around the whole thing, i might have to lift the sensor so that it sticks out from the box
I'll see what happens when the project is finished (my BSc thesis actually lol) and i'll let everyone know if this is an issue or not
We do try to learn from our mistakes, and others' as well. I will confess to recently putting a temperature sensor too close to a voltage regulator on a board, made the sensor pretty much useless Coincidentally, the board also contained a W5100, but as part of a WIZ811MJ module that was socketed to the board. But the regulator was the main culprit.
Ah yes, the voltage regulator I'm having issues with that too, I actually have another thread opened about that. Took some effort to diagnose the problem, since i'm a computer engineer and i'm a bit new to electronics. I used a 12V adapter and with the eth shield and a reed relay drawing a lot of current, i almost melted the damn thing hehe. Have to switch to a 7V adapter now, hopefully it'll be enough and i won't have to bypass the regulator altogether, since i don't know exactly how to go about doing that.
Thanks for the heads up, i really do have to watch out for that regulator
And there's no guarantees that it would work.. That was my first thought too tho, maybe i'll get a chance to try it out.
Coincidentally, the board also contained a W5100, but as part of a WIZ811MJ module that was socketed to the board. But the regulator was the main culprit.