Must-haves for EASY DIY controller?
- Thread starter der_wille_zur_macht
- Start date
der_wille_zur_macht
Team RC
I've been thinking about that, but am not sure if that would be the best route to go - because then every time we changed something, we'd have to get a mod to update that post.
Instead, I'm thinking about doing "news updates" of some kind, but the google code site is awkward for that too, IMHO. . . so I'm kind of stalled.
Really what I'd like to do is get the Ethernet working, then concentrate on a "public release" version complete with a user's guide of some kind, or at least an FAQ. But with my time going elsewhere right now I'm not gonna get to that point unless others continue to help out. Once we hit that stage, I think it would be fair to update the original post, and/or just start a new thread, so we can keep "how do I build/use this?" discussions separate from "let's engage in further development" discussions.
Instead, I'm thinking about doing "news updates" of some kind, but the google code site is awkward for that too, IMHO. . . so I'm kind of stalled.
Really what I'd like to do is get the Ethernet working, then concentrate on a "public release" version complete with a user's guide of some kind, or at least an FAQ. But with my time going elsewhere right now I'm not gonna get to that point unless others continue to help out. Once we hit that stage, I think it would be fair to update the original post, and/or just start a new thread, so we can keep "how do I build/use this?" discussions separate from "let's engage in further development" discussions.
DustinB
New member
Yeah, that's probably the thing to do. Really once the ethernet part is fixed we are pretty much ready for a "final/almost final" board huh? I haven't encountered any other problems, other than maybe allowing the user to select which AVR he wants the ENC on due to size constraints.
DustinB
New member
I don't know, the circuits seem the same. The only thing I can think of is something in the circuit is causing some interference or something. The shield I ordered is the same nuelectronics circuit, which appears to be the same as the hydra circuit.
I'm ordering a scope tonight to see if that will help me sort it out. Been meaning to get one anyway.
I'm ordering a scope tonight to see if that will help me sort it out. Been meaning to get one anyway.
der_wille_zur_macht
Team RC
Is that the hydra running the shield, or RBBB running the shield?
DustinB
New member
Alright, so we are almost there. I finally found the biggest problem.
ENC Pins
Pin 22: "Ground reference for oscillator" - This was set to +3.3v
Pin 25: "Positive 3.3v supply for oscillator" - This was set to ground.
I pulled the chip and bent the two pins upward, connected them to ground and 3.3 respectively. I now have both LED's on the magjack flashing twice and the green LED stays on.
I cannot get a reply ping or any activity on the LEDs indicating they are receiving anything. TPOUT+ & TPOUT- are both staying at 1.1v instead of 3.3v for them being active. MISO is also staying low. Everything seems to be indicating that the AVR is communicating and the ENC is accepting input from the AVR, yet the ENC is not sending anything back to the AVR for some reason.
I'm tired and I'm going to bed, I'll play with it some more tomorrow evening.
ENC Pins
Pin 22: "Ground reference for oscillator" - This was set to +3.3v
Pin 25: "Positive 3.3v supply for oscillator" - This was set to ground.
I pulled the chip and bent the two pins upward, connected them to ground and 3.3 respectively. I now have both LED's on the magjack flashing twice and the green LED stays on.
I cannot get a reply ping or any activity on the LEDs indicating they are receiving anything. TPOUT+ & TPOUT- are both staying at 1.1v instead of 3.3v for them being active. MISO is also staying low. Everything seems to be indicating that the AVR is communicating and the ENC is accepting input from the AVR, yet the ENC is not sending anything back to the AVR for some reason.
I'm tired and I'm going to bed, I'll play with it some more tomorrow evening.
mammut89
New member
damn that is a good find dustinb. how did we miss that ?
have you tried using wireshark to see what exactly is going to the network ? might help you debug it a bit easier.
http://www.wireshark.org/
have you tried using wireshark to see what exactly is going to the network ? might help you debug it a bit easier.
http://www.wireshark.org/
Last edited:
der_wille_zur_macht
Team RC
Nice find Dustin. It's amazing that I've spent at least a dozen hours checking the schematic, and I've posted it on numerous electronics forums, and NO ONE noticed that! Yikes.
I'm wondering if your ENC is fried thanks to the reversed polarity on that pair of pins. Or if there's some other issue with it. Did you socket that IC? If so have you tried replacing it?
I'm going to go ahead and make that change on the Eagle files. Getting blinking lights on the magjack is a huge success and I'm hoping we're close to solving this now. Eagerly awaiting any other findings as your troubleshoot.
I'm wondering if your ENC is fried thanks to the reversed polarity on that pair of pins. Or if there's some other issue with it. Did you socket that IC? If so have you tried replacing it?
I'm going to go ahead and make that change on the Eagle files. Getting blinking lights on the magjack is a huge success and I'm hoping we're close to solving this now. Eagerly awaiting any other findings as your troubleshoot.
DustinB
New member
No kidding, I've spent many hours myself. It's always something so simple. The problem finding the issue was the board didn't totally match the schematic.
It did fry the ENC, I had to replace the ENC to get the blinking lights after the fix. So yes we are very close. :beer:
Just gotta think, if everything is initializing and the ENC is receiving/executing commands, then why would it not communicate to the AVR? Hmm......
It did fry the ENC, I had to replace the ENC to get the blinking lights after the fix. So yes we are very close. :beer:
Just gotta think, if everything is initializing and the ENC is receiving/executing commands, then why would it not communicate to the AVR? Hmm......
der_wille_zur_macht
Team RC
Time to check the board to make sure the traces from the ENC back to the AVR are correct I guess, including the level shifting chip. . .
Sadly I totally hosed my main prototype Hydra trying to replace some of the Ethernet-related components so you're on your own right now unless others with the prototype board jump in. I only have one PCB left but I don't want to build it unless it seems inevitable, since I'd rather wait for rev2 and not waste components. . .
Sadly I totally hosed my main prototype Hydra trying to replace some of the Ethernet-related components so you're on your own right now unless others with the prototype board jump in. I only have one PCB left but I don't want to build it unless it seems inevitable, since I'd rather wait for rev2 and not waste components. . .
DustinB
New member
No I haven't tried anything like that, it was late last night when I had my eureka moment. I'll be sure to check it out, see if the jack is even receiving anything. You know, I didn't even think to see if my router had a connection light on that port. I have a feeling it may be something in the wiring of the jack, but due to the wiring of the jack it makes it hard to check continuity of the pins.
DWZM: My scope won't be here til next week. Maybe if you get a chance could you put a scope on TPOUT and some of the other communication pins to the magjack/AVR. See if anything is even being attempted? Simplest fix to get to my stage without modifying anything is to pull the chip, bend the pens out, and lightly solder 2 wires to the pins to plug into the board.
DustinB
New member
Ouch, that sucks man. I almost did that myself trying to get the magjack out. Started freaking when I saw solder pads lifting. :hmm2:
No worries, we'll find the answer one way or another.
Actually, if someone else has a good prototype and hasn't jacked with anything. Maybe they could try my fix and try it out just for the heck of it. Maybe I screwed something up on the board removing/replacing stuff?
terahz
1x10^12 Hz
Hey guys, sorry for being out for so long but I've been busy...
That relay strip from musiciansfriend looks really nice. Makes for a neater package and less wiring, I assume.
As for the ethernet, I can give the fixes a try on my board. I don't know if I have a spare ECN, but I'll try tonight and report back.
That relay strip from musiciansfriend looks really nice. Makes for a neater package and less wiring, I assume.
As for the ethernet, I can give the fixes a try on my board. I don't know if I have a spare ECN, but I'll try tonight and report back.
DustinB
New member
That would be great. Your ENC may be fine, definitely worth a shot. Shoot me a PM if you can when you find out, I don't get thread reply notifications for some reason. I'll be working more on this tonight too.
TheFishMan65
New member
Click the "MY RC" button on the top right. Click "View all Subscribed threads" bottom of the list on the right. Check the box for the one you want to change, change the subscription type and click go. Sorry for the off topic response.
PS Good find on the power problem.
[EDIT]
OK so I should have PMed him (done now).
PS Good find on the power problem.
[EDIT]
OK so I should have PMed him (done now).
der_wille_zur_macht
Team RC
I just have to say, SON OF A B#*!#
The schematic and board are "correct." The problem is, the ENC28J60 part I used in Eagle, which I grabbed from some third party, had those two pins matched between the schematic and the part incorrectly. They were in the "right" places on the schematic, but the pins were not correct on the physical part. So according to Eagle, the design was correct. Matching it to the datasheet, I see the problem you're describing. That's why no one caught it to date, we've all been assuming the Eagle part was correct.
Now I get why people say you should never rely on others' Eagle parts without checking them.
The schematic and board are "correct." The problem is, the ENC28J60 part I used in Eagle, which I grabbed from some third party, had those two pins matched between the schematic and the part incorrectly. They were in the "right" places on the schematic, but the pins were not correct on the physical part. So according to Eagle, the design was correct. Matching it to the datasheet, I see the problem you're describing. That's why no one caught it to date, we've all been assuming the Eagle part was correct.
Now I get why people say you should never rely on others' Eagle parts without checking them.
