Version 2 of my homebrew controller


 
Hey john franks I found the problem that I was having with your code for the LCD that was keeping me from compiling. See red arrow and explanation bellow.

#define USE_WISHIELD
// LCD transmit pin
#define LCD_TX 3 <span class="ev_code_RED"><=======</span>
#define USE_KEEPALIVE

I deleted the space between the TX and the 3 and it compiled. I gutted the keep alive out and have something different I will try if need be.
 
Originally posted by S ROY:
Hey john franks I found the problem that I was having with your code for the LCD that was keeping me from compiling. See red arrow and explanation bellow.

#define USE_WISHIELD
// LCD transmit pin
#define LCD_TX 3 <span class="ev_code_RED"><=======</span>
#define USE_KEEPALIVE

I deleted the space between the TX and the 3 and it compiled. I gutted the keep alive out and have something different I will try if need be.
Click to expand...
strange... when i take the space out it breaks the code. maybe i made a change in the driver i forgot about???

let me know if you get something other than keepalive working! this is where im stuck atm. wishield connects to wpa2 after about a minute, responds to pings, doesn't serve webpages, then arduino reboots. i haven't dug into much yet, maybe tonight...

did your lcd come in yet? cheers!
-jc
 
That is strange if I leave the space it breaks it. We have the opposite problem. No My LCD was mailed yesterday usually takes 2 or three days to get to Texas from Colorado. Maybe Wed or Thursday. I have a quick Question dose your fan run ok. I noticed from your code you were running it on pin 4. I was under the perception that it should be on a PWM pin 3,5,6 and 9 (if you disable the indicator light on WiShield by removing jumper on pin 9.)
 
Originally posted by S ROY:
That is strange if I leave the space it breaks it. We have the opposite problem. No My LCD was mailed yesterday usually takes 2 or three days to get to Texas from Colorado. Maybe Wed or Thursday. I have a quick Question dose your fan run ok. I noticed from your code you were running it on pin 4. I was under the perception that it should be on a PWM pin 3,5,6 and 9 (if you disable the indicator light on WiShield by removing jumper on pin 9.)
Click to expand...
good catch! i didn't even notice... i hadn't connected the motor since i made that change, but it seems to spinning fine now. i didn't run up the probe temp to see what happens at less than 100% though. i'd kinda like to keep the led... i don't think 5 is used by anything. i'll move it there and see what happens.

care to share your possible keepalive alternative? i think i might need to reload the drivers and start over with the wishield.... i can barely remember the last time i made forward progress with it. =/
cheers!

-jc
 
I am going to use pin 6 for PWM of blower.

Click on this link.

http://www.asynclabs.com/forum...topic.php?f=10&t=290

For the keep alive go to the last post on the first page. There is a download in the post by GregEigisti. Download the file and follow the post instructions recompile your code; after all changes are made. Up load them back to the Arduino. This is the first thing that I tried so far and it appears to work. I have been running the web server example sketch since early afternoon and It has not disconnected. I power cycled the AP several times and the shield came right back up. I have my fingers crossed. Hope this helps.
 
@ john frank, I have bad news and some good news.

The bad news. If you lose connection for more than a 1 minuet this is not the solution for the situation. I set my AP on a timer killed it for 15 minuets while I was asleep last night. When I woke up this AM the Arduino had not recovered or reconnected. This confirms what another member in that thread stated.

The good news is that It runs pretty solid if your signal is strong and you do not lose connection.

Conclusion, I do think you can get through a long smoke with it patched this way if your WIFI signal is strong. I will try it as soon as I get it fully assembled.

If this dose not work I might buy an ethernet shield and go that route since I have ethernet access on my back yard Patio.
icon_eek.gif
(I know, I am a bit of a GEEK.)
icon_biggrin.gif
 
Wow my blowers just showed up I had no idea they were this small. I ordered the one that Bob suggests in his instructions which is 5cfm and a little larger 10cfm one.
 
Originally posted by S ROY:
If this dose not work I might buy an ethernet shield and go that route since I have ethernet access on my back yard Patio.
icon_eek.gif
(I know, I am a bit of a GEEK.)
icon_biggrin.gif
Click to expand...

I have had success with this powerline type ethernet interface:
ethernet over powerlines

Haven't had any dropped connection problems so far.

Curt
 
Originally posted by S ROY:
@ john frank, I have bad news and some good news.

The bad news. If you lose connection for more than a 1 minuet this is not the solution for the situation. I set my AP on a timer killed it for 15 minuets while I was asleep last night. When I woke up this AM the Arduino had not recovered or reconnected. This confirms what another member in that thread stated.

The good news is that It runs pretty solid if your signal is strong and you do not lose connection.

Conclusion, I do think you can get through a long smoke with it patched this way if your WIFI signal is strong. I will try it as soon as I get it fully assembled.

If this dose not work I might buy an ethernet shield and go that route since I have ethernet access on my back yard Patio.
icon_eek.gif
(I know, I am a bit of a GEEK.)
icon_biggrin.gif
Click to expand...
sroy, i had read that thread before, but didn't feel my wifi link would be reliable enough for the location of the smoker. on the other hand, i do have a 1 watt wifi amp just sitting here collecting dust
icon_biggrin.gif


the ethershield is a option.... but more $$$! ethernet would be easy though as there is a drop right next to the porch door. not quite nerd enough to have one outside
icon_wink.gif


i tried a few blowers i had laying around, but i didn't care for the way the bldc motors handled the pwm from the arduino. i instead, opted for a brushed motor for the mechanical commutation. ( hair dryer
icon_biggrin.gif
) i think a more elegant approach to the blower fan setup might be to use a BLDC fan with integrated PWM control. ie: cpu fan, 3 wire case fan, etc. this would eliminate the need for the transistor & diode and still allow for a variable fan speed. startup issues should be eliminated as well since the fan's internal controller handles the switching.

the OTHER blower option i've been thinking about is to use ducted fan made for rc airplanes. these are cheap, easy to adapt to the smoker, and really easy to control! (servo pwm example sketch)
some examples:
speed controller

fan & motor

that controller uses anywhere from 5 to 12v in and even provides a 5v 1a regulated output you could use to power the arduino and/or a servo. anywho, just some thoughts... cheers!
-jc
 
I figured it out.
icon_biggrin.gif
99.9 percent of my problems were user error and failure to read the instructions carefully the other .1 the (Keepalive) was info that was not included in this post or on Bob's web page. I found the solution posted by Bob on another forum by sure coincidence. I am on my way to completing this project.

LCD - Done
Wishield setup - Done
Code and Libraries - Done
Proto shield - Working on
mount for motor -
Final assembly -
Tinker and tweaking to come

john frank I have a handle on the keepalive. If you need help let me know.
 
s roy, i haven't tried them yet... been working on a few other projects for a bit.

are you referring to this thread?? i followed those instructions and still had problems... i think it might have been my server though. what did you do to get yours up & running?
cheers!
-jc
 
Once I finally reread the comments at the top of Bobs code; I realized that I did not edit the apps-config file to make WISERVER active and deselect WEBSERVER. Also go into the WISHIELD LIBRARY AND EDIT spi.h file comment out reference to #define USE_DIG0_INTR and un-comment USE_DIG8_INTR. Move INTR jumper on the wishield to the D8 pin if you have not done so already. My code broke the same place yours did when I tried to test it. I recognized the error immediately. In that post you referenced Bob gives examples of the edited library files. The first edit to the WiServer.h file is correct and complete. The edit to the WiServer.cpp file is correct but not complete. There is more code than what is listed in that post. You have to go to the Asynclabs thread that he referenced to earlier in that post to get the second edits full code. Insert the edits as per instruction. add a good Ip address to #define KEEPALIVE_ADDRESS. KEEPALIVE code makes a call to the Library files specifically for the edits he made to them. If they are not there then it won't work.
 
finally, some clarity!
icon_biggrin.gif
this makes sense as to why it worked for me at one point and then stopped... i followed the code here which got me serving webpages, but broke after a few minutes. after tinkering around with no forward progress, i reloaded the drivers and followed the instructions here which got me nowhere. i swear, getting this wishield information is like a digital scavenger hunt. async's documentation is weak at best. they should disclose that wiserver is experimental or a work in progress or something...
icon_rolleyes.gif


anywho, so yea... im back where i was sometime last week i guess... arduino boots up, wifi connects to the AP, led lights up, i can retrieve webpages intermittently.

after this point, one of two things happens... within a few minutes, either the arduino reboots and its another 40sec wait for it to reconnect OR the arduino locks up completely.

not too sure where to go from here... it would be nice if i could monitor what the arduino/wishield is doing when this happens. bob's code looks like it has this function, but i can't figure how to display it. ie:
<pre class="ip-ubbcode-code-pre"> // It has been one minute and no response - reset WiServer
Serial.println("No response - resetting WiServer");</pre>
<pre class="ip-ubbcode-code-pre"> // Send request to the keepalive ip and make sure we can still talk
Serial.println("Sending keepalive GET"); </pre>

i think i need to re-read this thread and go from there.

yours still up & stable? what AP/router are you using? cheers!
-sj
 
Yes it worked without a hitch. I am using a Linksys WRT54G version 2.2 router with the latest WRT54GL firmware Ver.4.30.14. Check Here to see if there are any compatibility issues with your router / AP. Can you borrow another router / AP to try to rule this out. Also try installing the most current firmware update on your router.
 
doh! ok, im a idiot... i commented out #define USE_KEEPALIVE at some point during testing and never re-enabled it. stoopid!

so atm, im using hruska's + greg's fix and im no longer getting the lockups, but still getting the intermittent reboot. i'll revert to just hruska's patch since its know to work for this application and then go from there. i suspect UDP packet blasts might be the cause of the restarts. mcaffe (which is a known culprit) is running on 3 windows boxes. when i unplug the hardline connections to the router so that its just my mac & wiserver connected via wifi, the wiserver stays up. when the other boxes are reconnected, wiserver goes down. sooo.... someone is killing it, just gotta figure out who. wonder if i can block udp to the wiserver???

i can't believe you aren't running dd-wrt on that wrt54gl! you do know the "L" is for Linux right??
icon_wink.gif
cheers!
-jc
 
I have it completely up and running. I just need to install it in the hobby box to finish it up. Yes I know about dd-wrt just have not had the need for any more than the oem firmware.
 
Originally posted by S ROY:
john franks I have a question for you. Two of my probes are not working. Any guess to why they are not working. Yes I know about dd-wrt just have not got around to messing with it.

Thanks
Click to expand...
i had a lot of difficulty getting my probes to work consistently at first. the main problem was with the maverick plug; it is about 2mm longer than a standard 2.5mm phone plug. if used with radio shack 2.5mm jacks, pushing the maverick plugs completely into the jack resulted in the tip contact of the jack touching the sleeve of the plug. pulling the plug back out about 2mm allowed the tip contact to seat properly.

the other issue i had was with the 2.5mm jacks from radio shack. they're garbage! if you have those or similar plastic jacks i would recommend swapping them out for something more reliable. switchcraft TR2A for example. switching the jacks out completely solved my wacky intermittent probe issues.

there might be a standard jack for those long plugs, but to my knowledge, none exist. i just took a closer look at a maverick wireless monitor and it looks as if they have the jack recessed from the case so they seat properly.

anywho, i would check the connections first. i'd bet thats where your prob is... cheers!
-jc
 
I went back and edited my post I got all four of the probes working with a little editing in the code. It is working very well. I will have to wait a little longer because I decided to order some locking power plugs for the power supply and the blower. So that they can be solidly plugged in without fear of the being accidentally unplugged. All I have left to do now is decide the lay out inside the box, shorten,clean up the wiring and mount it in the box. FYI for anyone that cares the probes from the Oregon Scientific BBQ/ Oven Meat thermometer Model AW131 (the talking one) will work with this just fine.
 
You must log in or register to reply here.

 

Back
Top