Sorry... I did try to search for earlier discussions... but couldn't find a good search string.
I have a simple TCP/IP server running in a Sparkfun Ethernet Pro... the product with "Pro Ethernet" in the silkscreen, sku: DEV-10536. (Nice little device!)
It works fine... when I have my ModernDevice "BUB" programming cable attached to the DTR/ RX/ TX/ 5v/ Gnd pin set. (The BUB uses the FT232R chip, the same as Arduinos, the FTDI cable ModernDevice sells, and the Spark Fun breakout boards.)
My program doesn't interact with the computer I use for programming it once the program in the Arduino starts running... at least I don't THINK it does! (I haven't started communications with the serial monitor, for instance.) I don't thing that anything is traveling via the BUB now except the power to operate the Ethernet Pro.
However... If I disconnect the BUB, and plug suitable a DC voltage into the connector on the Ethernet Pro, although LEDs wink merrily, the server no longer serves. I've re-booted the client software after the power swap. (Also pressed the Ethernet Pro reset button.) I didn't disconnect the device from the LAN during the power change over... those lines are pretty "hot swap" friendly, aren't they?
Anyone have any ideas? Why will it work if powered via programming cable, but not if powered from a wall-wart which powers a different Arduino just fine?
I've swapped the power provision a couple of times... same result each time. "5v" line at about 4.5 on both of my Arduinos this evening. Voltage from wallwart, when under load of Ethernet Pro, 7.5 volts.
If there is something specific in the code, rather than a simple newbie Gotcha, I'm in trouble, as I really can't expect anyone to pour through it all.... but the code is at...
... and it is explained at...
"It" is a SIMPLE server for TCP/IP... not a WEB server. It needs bespoke client software (provided), but the reward for that hassle is more room in the Arduino to add the features you want, and more "power" to make the whole system (server+client) do what you want, your way.... well... once the powering issue is overcome, unless you want to tie up a big PC and interface cable to make the thing run! <^_^>
tkbyd, I hate to say it, but this really sounds like the power is screwed up somehow. Have you checked the voltage on the 5V line with the FTDI plugged in and compared that to the same point with the wall wart? It could also be noise coming from a switching wall wart that isn't getting filtered and messing things up. Maybe a cap, .01uf or so somewhere. I totally agree with you that when it works with one power supply and not the other, it's just gotta be the power supply. The question is why.
If I remember correctly, the ethernet chip is 3V, maybe check that buss as well to be sure it didn't drop somehow when you changed power supplies. I once had a USB cable that caused this kind of failure. It seems those 5V wall warts that can supply up to two amps and have a USB output won't work through a cheap USB cable. That silly cable took me a couple of hours to isolate because I wouldn't believe a piece of wire could be causing my problem.
Thank you for that... you'e inspired me... I'll see what happens with a different wall wart. The other Arduino, the one it runs okay, uses a different ethernet interface... maybe less "fussy".
This mostly to post a bit of information for anyone reading this... The Sparkfun Ethernet Pro runs on 5v. (An output driving an LED reads 4.5v.)
Hmmm. As I type that, I see another thing I need to explore... thank you.
"It runs on 5v" at least as far as what you need to pump into it is concerned... if you supply power in the group of 6 pins from the programming cable. Now... when I try to run the thing via wall wart, I am feeding 7.5v to the onboard regulator... the regulator I'm ONLY using when the thing doesn't work. So maybe the fault is in the regulator? Although I do get the same value... 4.5 or so... on the "5v" which would go to a shield if I plugged one in... same value as I got when powering the thing via the programming cable. Does being so far from 5.000v surprize/ worry anyone? All I can say is that the voltage is (at least nearly) the same in the working (4.6v) and in the not working (4.5v... yes close, but maybe just below the threshold. Or noisy.) scenarios.
====
Ha! Happy ending!!!
With a different wall wart, I get a nice 5v... 4.99, according to my Fluke, and the device works fine! Hurrah! And I even, while hunting for the other wallwart found my jeweller's loupe which I've needed for days, but has been in hiding.
Nasty Mr Murphy!! Although I suppose I should be grateful that the wallwart IS able to supply power for ONE of my Arduinos, even if not for the other. (No, not both... I only ever asked it to power one Arduino at a time.) I suppose I should check the current demand of the two quite similar systems, for curiosity's sake.
Oh.. and as for house wasted by trusting a cable with a borderline fault... been THERE. Sigh.
===
Thank you, forum for being there so I could get down to asking the questions, and for the ideas which lead to the answer!