LinkMeter v2 Homebrew BBQ Controller - Part 2


 
I have the ET-732 probe, and they are 6ft in length. I use the "pit" version which doesn't have the pointy end like the food probe, but they both have the same thermocouple and should work the same.

Since I can't attach anything on this forum, wish we could use a forum software that wasn't 20 years old, here is the table of temps vs resistance for the ET-732 probes:

F R(ohms)
-76 2.24E+08
-58 1.01E+08
-40 4.80E+07
-22 2.39E+07
-4 1.24E+07
14 6.68E+06
32 3.73E+06
50 2.16E+06
68 1.28E+06
77 1.00E+06
86 7.86E+05
104 4.93E+05
122 3.17E+05
140 2.09E+05
158 1.40E+05
176 9.59E+04
194 6.68E+04
212 4.74E+04
230 3.41E+04
248 2.50E+04
257 2.14E+04
266 1.85E+04
284 1.39E+04
302 1.06E+04
320 8.11E+03
356 4.95E+03
392 3.14E+03
428 2.06E+03
464 1.40E+03
500 9.71E+02
538 6.92E+02
572 5.05E+02

You can use this to see if the probe works at a couple known temperatures. I do room temp, and boiling water temp. If that is working, then the probe is fine. Let's double check your coefficients. For the ET-732 they should be:

A. 0.00043933992
B. 0.000208342
C. 1.2004001e-8

The Resist value should match the schematic which should be: 10kohms if you didn't make any changes here.

dave

Sorry to bring up a post over a year old but I just discovered this project a few days ago. I'm excited about it. Thanks to Bryan and all.

But my first question is for dave... Where did this table and/or these coefficients come from? Is there any documentation on the probe? Did you figure out it is some kind of standard probe that has this known info? If so, what kind is it?

I have been working on a similar project and I use thermocouples mainly because I couldn't find this type of info on some inexpensive commercial probes.

Roger
 
Roger,

I did find a data sheet for the thermocouple and that is where I got it. I can't remember right now where I found it. I'll look around when I get a chance next week if you would like.

dave
 
Roger,

I did find a data sheet for the thermocouple and that is where I got it. I can't remember right now where I found it. I'll look around when I get a chance next week if you would like.

dave

I feel like a google failure for asking, but yea, I'd love that. I can find tables for a bunch of other NTC thermistors but they seem to stop before they get to 1M ohm type.
 
I found them at Honeywell. I basically looked into the specs, and found one at honewell that matched. I was able to look at all the R/T data for their NTC thermistors and find the one that worked.

You can also measure the R/T values at three places and use the program on github to calculate the values for you. I did freezing, room temp and boiling water as those are easy to obtain.

dave
 
Thanks Dave.

Another question on the design, Bryan. The LCD data sheet says LCD pin 3 resistance should be 10-20k, but you use a 10k trim, which I think means *max* 10K. Since the higher the resistance the less power for the LCD, I would expect the 10k trim to make for a possibly over-contrasted display. Did I miss something? No issues with 10k trim? Or should you throw a 10k in series with the trim to make it 10-20k?

EDIT: just my inexperience with potentiometers, I think the data sheet means the potentiometer should be 10k-20k so a 10k is fine. Never mind.

Roger
 
Last edited:
I've used both and it doesn't seem to make any difference. I'm not sure of the amount of current the contrast pin pulls. I think it is just an input to some sort of high impedance comparator which means it doesn't have any current draw to speak of. Really what it is used for is as a voltage divider to set the voltage level at the contrast pin so 10k, 20k, maybe even 100k would be fine.
 
It depends on how far back you were updating from. There was a bug in LuCI that required an upstream fix but even if you update the appropriate packages, there's still a problem with a missing requirement. I've been trying to get the Attitude Adjustment RC running on the WRT54GL but haven't had a lot of time to dedicate to getting it working. You can try backing up your config and flashing the whole image, but the network configuration web pages won't work until I get a fix.
 
I don't suppose someone here happens to have a .bin laying around that works on WRT's that they could toss up on the web/email me? Unfortunately the posted builds with the latest builds of Attitude Adjustment are too mem hungry to run on the WRTs, and Bryan is understandably busy so rolling back to an earlier release of OpenWrt is not in the cards for a while.

If anyone has one that they know works (probably from some time before the Alarm code was rolled in), I'd love to get my hands on it so I can resurrect my LinkMeter.
 
Hmm, is there a trick to getting stock OpenWRT to take these .bins over the web interface? Every time I've tried to upgrade via the web interface with the LM images it always complains about it not being of the correct format. I tried this image over tftp as well, but the stock OpenWRT image I'm running is not tftp friendly (it sits for about 2 minutes before it responds to pings, and won't take an image).

I've had luck writing with mtd via the command line, but for whatever reason I never have luck over the web.

Edit: Success! I figured that a .trx was a stripped down .bin file, and determined that you can strip the bin header off of the file and flash via the command line as you would with a trx via:

root@OpenWrt:/tmp# dd bs=32 skip=1 if=openwrt-wrt54gs-squashfs.bin of=openwrt.trx
98432+0 records in
98432+0 records out

root@OpenWrt:/tmp# mtd -r write openwrt.trx linux

Bam! I've managed to resurrect my Linkmeter! It's not without bugs, but it has reminded me that it's time to order another rPi and build a new one. :)
 
Last edited:
Question: On my 3.2 build, the LCD doesn't always power on with the router. Functionally the board is fine, but sometimes it takes several replugs for the LCD to fire up. I'm using a 12v 1amp Linksys wall wart, and there's no obvious short or loose connection. Anyone run into this?
 
OK, here we go....I'm going to go down the road and try to build this thing. :D

From the wiki assembly / parts list:

Mouser is out of stock on the 10k trimmer resistor / pot. (http://www.mouser.com/Search/Produc...3KT20virtualkey61330000virtualkey72-T70YE-10K)

Digikey is out of stock on the header socket (http://www.digikey.com/product-detail/en/PPPC132LFBN-RC/S7116-ND/810252).

And of course the wifi dongle at Amazon that was in stock all weekend is out of stock now. (That'll teach me.)

For alternatives....

Mouser has 13,000 of the 20k version of the pot in stock. Would that work for a substitution? (It's for contrast control on the display, right?)
Newark shows the pot in stock, but doesn't return the header in their search. Same with Allied Electronics. Is there a substitution for the header connector?

Futureelectronics.com turns up in a Google search with the header in stock, but of course is OOS on the pot.

Are there other sources I should be checking? Thanks!!

Jarrod

(Yes, I hate paying $6 shipping on a $0.50 part.....)
 
Mouser has 13,000 of the 20k version of the pot in stock. Would that work for a substitution? (It's for contrast control on the display, right?)
Newark shows the pot in stock, but doesn't return the header in their search. Same with Allied Electronics. Is there a substitution for the header connector?
You probably should have started a new thread for this so other people can find the answers, but

Potentiometer: Yeah the 20k is an ok replacement, I've got a 20k in a couple of my builds. There's also the Bourns equivalent potentiometer.
Header: Pin headers/socket are a real nightmare in general to find. But here's an alternative part I considered when looking for the part I settled on Digikey. It is a little larger because it has some extra plastic on the ends, but I think it will still work.

The wifi adapter, all you need is something that uses the 8192cu driver. I have this Rosewill wifi adapter that uses the same chip and works identically. I'm willing to bet that 90% of the usb wifi adapters that are similarly sized use the same realtek chip.
 
I bought the wifi dongle at newegg. I looked yesterday about getting another one, for a linux box of mine, and I think they were $9.99 with free shiping.

http://www.newegg.com/Product/Product.aspx?Item=N82E16833315091

dave

OK, here we go....I'm going to go down the road and try to build this thing. :D

From the wiki assembly / parts list:

And of course the wifi dongle at Amazon that was in stock all weekend is out of stock now. (That'll teach me.)

For alternatives....

Are there other sources I should be checking? Thanks!!

Jarrod

(Yes, I hate paying $6 shipping on a $0.50 part.....)
 
You probably should have started a new thread for this so other people can find the answers, but

Potentiometer: Yeah the 20k is an ok replacement, I've got a 20k in a couple of my builds. There's also the Bourns equivalent potentiometer.
Header: Pin headers/socket are a real nightmare in general to find. But here's an alternative part I considered when looking for the part I settled on Digikey. It is a little larger because it has some extra plastic on the ends, but I think it will still work.

I almost started a new thread, but then I saw some similar requests in this one. I'll get a new one started with the pertinent info from this one.

Once DigiKey's ordering site decides to respond I'll complete that order.


Jarrod, try this pot instead (Bryan suggested it to me last week).

Edit: Too slow, Ninja'd by Bryan!

He's on it! Swapped in the 10k Bourns pot, Mouser order is now placed.

I bought the wifi dongle at newegg. I looked yesterday about getting another one, for a linux box of mine, and I think they were $9.99 with free shiping.

http://www.newegg.com/Product/Product.aspx?Item=N82E16833315091

dave

Ahh, thank you! Indeed $9.99 with free shipping. Cancelled the OOS one at Amazon, ordered from Newegg. It'll get here fast enough for my needs....and after doing the "Ah, why not" and ordering a Hakko soldering station, I should probably try to save a couple dollars on this vs Bryan's option. :)

PCB and RPi were ordered last week, SD card is in an Amazon order along with the rest of my new soldering stuff. Gotta order probes yet, and will dig through my assortment of wall warts to see if I have anything that will work there.

Thanks, guys!!

Jarrod
 
Assembled my 4.0 build tonight, and everything went well. Sorta.

When I first setup this build, everything was fine (probe temps were showing up in LuCi, etc). I did something (who knows), and now everything is showing up as "off" in LuCi, I can't change the setpoint, etc. The LCD readout shows probe temps correctly for each jack, and I can flash the AVR via LuCi, but something else seems wonky. Any thoughts? Is there a way to reset the HeaterMeter related settings?

Here's a screenshot if it helps: http://i.imgur.com/hPKvm.png

Typing anything in the fields that are blank and tabbing or changing focus does not update the "Update URL" at all.

Edit: It appears there is some type of communication issue with the rPi. I can plug in, watch the LCD update the pit probe temp, recognize additional probes, etc. At some point during the rPi boot, the LCD stops updating. Removing probes does nothing, changes in temp are not reflected, etc.

It's as though my board is crashing after the Pi boots for some reason..

Edit: It's definitely the board crashing, even if I power on the Pi without a flashcard, the LCD eventually goes unresponsive.

Speaking of the LCD, I really need to order a case, trying to wedge crap behind the LCD so that it makes contact and works without one is a pain.
 
Last edited:
That would require that I've got my button soldered on. :)

Is there some method via the command line that I can use? As I said, I can flash the AVR firmware off of the net just fine, OpenWRT is fully functional, etc.

Edit: Snapped my button into place, here goes!
 
Last edited:

 

Back
Top