Web Controlled All-in-One Temp Controller

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
I purchased a board that allows me to read 8 digital temperature sensors and output 8 logic signals that will run solid state relays. The board has an ethernet port and is programmed and controlled via a web page. It can turn on and off the logic outputs based on timers, temp sensors, humidity sensor, digital inputs, and analog inputs or a combination. Currently I use it to control my keezer and also monitor fermentation temps and run a heater if needed to maintain those temps.

My plan is to mount this all up nicely in a box. Here are some things I'm considering:

1) Have several (up to 8) 110 V wall outlets all ran by the board via Solid State Relays mounted so that you can plug power cords into the outside of the box.
2) I'm currently using telephone wire for the temp senors. I'm planning on using 3.5mm headphone jacks to disconnect the sensors easily and use panel mount females in the box. (I'm successfully using one temp sensor through 30 feet of telephone cord) The temp sensors are digital and all run to the same place on the board so I can always use a 3.5mm splitter to add another temp sensor.
3) Mine is hooked up to a spare wireless router so it can connect to my network without running an ethernet cable, but one could easily use a ethernet to wireless converter and mount that in the box.

What does everyone think? I'll add some pictures and more information as I go.
 
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS

Thanks for that. Either thats new or I somehow missed it when I was looking before. I've been running this board for quite a while inside of a zip lock container. I really need to move on to the next stage of mounting this up in a box before I break something.

I was originally going to use a toolbox, but I'm beginning to lean towards a computer case. First, I could use the power supply for DC power. Currently I use a cell phone charger. Also, I could use fans to cool heat sink mounted SSR's.

The beauty of this thing is you can control eight different things pretty easily, but you are limited to where you can run an extension cord and phone cord (or stereo cord) from the unit. And you can monitor and control all these things remotely.
 

md20_20

Supporting Member
HBT Supporter
Joined
Aug 15, 2008
Messages
44
Reaction score
1
Location
Alabama
2) I'm currently using telephone wire for the temp senors. I'm planning on using 3.5mm headphone jacks to disconnect the sensors easily and use panel mount females in the box. (I'm successfully using one temp sensor through 30 feet of telephone cord) The temp sensors are digital and all run to the same place on the board so I can always use a 3.5mm splitter to add another temp sensor.

Can you elaborate on the wiring for this? Is it sufficient to have each DS18B20's pins tied to the corresponding pins of the others (i.e. in your example of using 3.5mm plug/jack connections, all the tips are common DQ, all the rings are common Vdd, all the shields are common GND)?
 
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
Can you elaborate on the wiring for this? Is it sufficient to have each DS18B20's pins tied to the corresponding pins of the others (i.e. in your example of using 3.5mm plug/jack connections, all the tips are common DQ, all the rings are common Vdd, all the shields are common GND)?

Yes. They can all be common. Each sensor has its own ROM code and the board identifies what signal comes from which sensor by that. I'm planning on having multiple sensor leads. But the corresponding conductors on the jacks will be all tied to the same terminal on the board (I'll probably look at terminal rails to keep it organized).
 

jcdillin

Well-Known Member
Joined
Mar 7, 2008
Messages
641
Reaction score
3
Location
Miami, FL
Watch out on using those headphone jacks as they do short out when plugged/unplugged.

We had considered the in the beginning of our project but ditched them for this reason.
 

jcdillin

Well-Known Member
Joined
Mar 7, 2008
Messages
641
Reaction score
3
Location
Miami, FL
Right, but it was mentioned in the thread that they were using stereo jacks which will do this.

The other option would be to run the DS18 sensors in one wire mode which would mean you would only need 2 wires instead of the normal 3. That way you could use RCA's or mono headphone jacks.
 

leboeuf

Well-Known Member
Joined
Feb 24, 2009
Messages
319
Reaction score
3
Location
Albuquerque, NM
Right, but it was mentioned in the thread that they were using stereo jacks which will do this.

Ahhh correct, shot down on safari.

As for providing useful information:
I would probably use a mini xlr connector for this setup...
 
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
The sensors only require one wire (data) to go back to the board. I believe I could still use stereo jacks to power the sensors as well, but use a separate power source so as not to harm the board by shorting it. Then make sure the sensor data line only gets crossed to ground on the plugs when plugged or unplugged.
 

md20_20

Supporting Member
HBT Supporter
Joined
Aug 15, 2008
Messages
44
Reaction score
1
Location
Alabama
The other option would be to run the DS18 sensors in one wire mode which would mean you would only need 2 wires instead of the normal 3. That way you could use RCA's or mono headphone jacks.

Looking at the manual for the board it seems that the board expects to run the temp sensors in powered mode, not parasite. I'm not sure if there's a programming issue with using parasite mode but I'd sure like to be able to, as I already have three DS18B20s wired for parasite mode potted into stainless probe ends I ordered from Derrin at http://www.stirplates.com/probeends.htm .
 
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
Ahhh correct, shot down on safari.

As for providing useful information:
I would probably use a mini xlr connector for this setup...

Ya, I looked at that. Then I wouldn't have to worry about this shorting at all, but it looks like they are pricey compared to these mini stereo jacks.
 
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
Looking at the manual for the board it seems that the board expects to run the temp sensors in powered mode, not parasite. I'm not sure if there's a programming issue with using parasite mode but I'd sure like to be able to, as I already have three DS18B20s wired for parasite mode potted into stainless probe ends I ordered from Derrin at http://www.stirplates.com/probeends.htm .

I haven't tried parasite mode, but I was under the impression it would work. It just seemed easier to run an extra conductor. The telephone wire I'm using has 4 anyways. I guess I should look into it, bc mono plugs are even cheaper than stereo plugs. I thought there was a downfall to parasite mode.?.?

I hadn't seen those. For the two sensors I'm using now. I sealed the sensor connections up with many layers of liquid black tape and believe it or not they seem to be waterproof. I don't know how long that that would hold up though (I don't submerge them anyways). Definitely would/will go with the stainless probe ends for some applications
 

md20_20

Supporting Member
HBT Supporter
Joined
Aug 15, 2008
Messages
44
Reaction score
1
Location
Alabama
I thought there was a downfall to parasite mode.?.?

The downfall is that it probably takes longer to request and receive a reading in parasite mode. Also I had to fiddle with the value of the pull-up resistor for long cable runs, I ended up with something like 1.5 Kohm (with multiple resistors paralleled on the breadboard) before it would work reliably with a 50' extension cable. (You can get a 50' stereo headphone extension on Amazon for a ridiculously low price, BTW, but the ones I got had the tip and ring cross-wired so I had to cut and redo them anyway, plus drop a conductor out because I have mono cable and plug on the probe.)

I ordered one of these, I will give it a parasite-mode try when it gets here.
 
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
Ya, I might have to pick one of the 50' extension cables up just so I can see if it still works over 80' (when I add my current 30' cord).

I'm going to do some playing with the stereo plugs in a few days and see which pins get shorted/crossed during plugging. I'm hoping that it doesn't cross all three because if not I think I can make them work.

Here's female panel mounts for 30-45 cents a piece. I don't know if I'll use 100 though......

http://www.allelectronics.com/make-a-store/item/SMJ-2/STEREO-3.5MM-MINI-PHONE-JACK/1.html

I ordered one of these, I will give it a parasite-mode try when it gets here.

Keep me informed.
 

leboeuf

Well-Known Member
Joined
Feb 24, 2009
Messages
319
Reaction score
3
Location
Albuquerque, NM
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
I really need to sit down and test this out, but I've been short on time (or motivation).

It shouldn't cross all three ever. The connection sequence would be: pin1>gnd pin2>nc pin3>nc, then pin1>jackpin2 pin2>gnd pin3>nc, and finally pin1>jackpin1 pin2>jackpin2 pin3>gnd....

I see what your saying, but here is how I think it could go down: The ground on the jack would be shorting pin2 with the ground on the plug. The next conductor on the jack would simultaneously be shorting pin1 and pin2 on the plug. Therefore all three pin connectors and are connected and 2 of the jack conductors. It may even be possible that pin1 of the jack is touching the last conductor in the jack at the same time. Whether this happens in practice is yet to be seen (by me anyway). I believe your sequence is missing the states at which things will be shorted (ie in between your steps).

I like those xlr's, but I'm looking for cheaper first. I may look at using RJ-11 (or similar jacks), if the stereo plugs flop... They are cheep and I'm already using phone cord.
 

Plan9

Well-Known Member
Joined
Jan 13, 2008
Messages
273
Reaction score
4
Location
Apopka, FL
Does this thing have to have a browser running at all times, or does it operate on it's own?
 
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
Does this thing have to have a browser running at all times, or does it operate on it's own?

It operates on its own... It has an internal web server for you to access to monitor or modify settings. Once you have it setup (to do what you want) you don't even have to have it hooked to a network.

If you don't like the way the built in web gui handles things, you can set the outputs via an external script....... but then you would be relying on it being on the network.
 

leboeuf

Well-Known Member
Joined
Feb 24, 2009
Messages
319
Reaction score
3
Location
Albuquerque, NM
I see what your saying, but here is how I think it could go down: The ground on the jack would be shorting pin2 with the ground on the plug. The next conductor on the jack would simultaneously be shorting pin1 and pin2 on the plug. Therefore all three pin connectors and are connected and 2 of the jack conductors. It may even be possible that pin1 of the jack is touching the last conductor in the jack at the same time. Whether this happens in practice is yet to be seen (by me anyway). I believe your sequence is missing the states at which things will be shorted (ie in between your steps).

Hmmmm that's a good point...

I think the rj11 connectors would be pretty slick. Why do you want to move away from them?
 
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
I bought some stereo plugs already (not many so NBD) and it seemed like they'd be cool when I thought of it. I'll probably go with RJ-11 plugs to avoid problems though.
 
Joined
Apr 23, 2009
Messages
35,612
Reaction score
15,438
Location
☀️ Clearwater, FL ☀️
The downfall is that it probably takes longer to request and receive a reading in parasite mode. Also I had to fiddle with the value of the pull-up resistor for long cable runs, I ended up with something like 1.5 Kohm (with multiple resistors paralleled on the breadboard) before it would work reliably with a 50' extension cable. (You can get a 50' stereo headphone extension on Amazon for a ridiculously low price, BTW, but the ones I got had the tip and ring cross-wired so I had to cut and redo them anyway, plus drop a conductor out because I have mono cable and plug on the probe.)

I ordered one of these, I will give it a parasite-mode try when it gets here.

I've designed hardware and firmware using Dallas' one-wire comm. It isn't hard, but you do have to mind the timing closely because the power is being derived from the comm line. Once it works, it always works. Nevertheless, I'd still run the power to it if there isn't some obstacle to doing that.
 

md20_20

Supporting Member
HBT Supporter
Joined
Aug 15, 2008
Messages
44
Reaction score
1
Location
Alabama
Nevertheless, I'd still run the power to it if there isn't some obstacle to doing that.

Yeah, well, the obstacle is I'm cheap enough not to want to buy new probe ends if these will work in parasite mode. :)

After futzing around with the pull-ups, the next set I build will be run in powered mode.
 

wormfishin

Well-Known Member
Joined
Jan 8, 2009
Messages
65
Reaction score
0
Location
Dallas, GA
I just grabbed one of these on ebay today, can't wait to start messing around with it, looking like the wiring might be a bit over my head, but that's how you learn right?

Anyone have advice on the probes?
 

wormfishin

Well-Known Member
Joined
Jan 8, 2009
Messages
65
Reaction score
0
Location
Dallas, GA
OK, so I found the sensor pretty damn cheap, is the wiring really as simple as connecting the 3 pins to the board over some cat 3(or 5) cable?
 
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
OK, so I found the sensor pretty damn cheap, is the wiring really as simple as connecting the 3 pins to the board over some cat 3(or 5) cable?

Yes, thats pretty much it. Are you planning on using solid state relays with the board output to control something (like a freezer)?
 

md20_20

Supporting Member
HBT Supporter
Joined
Aug 15, 2008
Messages
44
Reaction score
1
Location
Alabama

wormfishin

Well-Known Member
Joined
Jan 8, 2009
Messages
65
Reaction score
0
Location
Dallas, GA
Thanks for the link md20_20.

physfarm, I do plan on controlling relays, thinking I will replace my to love controllers and add a heat stick to my HLT. OF course that is just the start of it also considering for a pump and valve to fill the hlt and a few other random things
 

md20_20

Supporting Member
HBT Supporter
Joined
Aug 15, 2008
Messages
44
Reaction score
1
Location
Alabama
I ordered one of these, I will give it a parasite-mode try when it gets here.

Initial tests of my parasite mode DS18B20 probe with this board are not promising. (That's not unexpected, I was skeptical given the way the documentation reads.)

First off, it only enumerates the OneWire bus once, at power on. (This is a documented feature, see page 5 of the manual.) So the discussion about whether the phone plug/jack connection strategy is prone to shorting becomes somewhat academic. You have to cycle the power on the thing to get it to read any probe you connect, so you might as well power it down to plug them in.

I took a mono headphone jack and wired it to the board's temperature sensor input, using only the GND and Data(in) connections. This matches the parasite mode wiring of the DS18B20 where the DQ (center pin) is the data line and the other two pins are wired together and attached to ground. After cycling the power on the board and attaching the probe, I saw that it identified the probe's ROM (maybe) but the temperature reading was totally off. (The probe was at ambient but the reading was 85 C.) Immersing the probe in a cold beverage did not change the reading. I am not sure if the ROM reading value is correct or not as I did not note the ROM values of the chips I have. I will try to verify this using my Arduino.

Looks like I'm going to be building some more probes.
 

wormfishin

Well-Known Member
Joined
Jan 8, 2009
Messages
65
Reaction score
0
Location
Dallas, GA
yeah, I think I'm going to get those probe ends, they look pretty good, just have to figure how many and what length I need.

Can't wait on my board to show up. Anyone posted any pics of how they have mounted theirs?
 
OP
OP
physfarm

physfarm

Well-Known Member
Joined
Oct 24, 2008
Messages
53
Reaction score
0
Location
Wichita, KS
You have to cycle the power on the thing to get it to read any probe you connect, so you might as well power it down to plug them in.

I believe I have added a sensor without powering the board down and up, but I'm not running in parasite mode.

(The probe was at ambient but the reading was 85 C.) Immersing the probe in a cold beverage did not change the reading. I am not sure if the ROM reading value is correct or not as I did not note the ROM values of the chips I have. I will try to verify this using my Arduino.

I'm am familiar with the 85C reading. Several of the sensors I got exhibited this. However, I found that I was getting a higher voltage on the board from the power terminal associated with the humidity sensor than the power terminal associated with my temp sensor. When I started using that higher voltage for the probes they all started working (where only 1 worked on the lower voltage). I'm sorry I don't remember the specific voltages (wanting to say I was getting 3ish from the temp one and 5ish from the humidity one). And this may not be an issue for you in parasite mode.
 

Cede

Well-Known Member
Joined
May 24, 2007
Messages
72
Reaction score
2
Location
Thetford Mines
I had some trouble too with the d18b20 when I was playing with them and a PIC18F2550.
The power was not really 5 volts and was too noisy.
For the probes, you can get a look on ebay and search on thermowell.
I saw some that fit 1/2NPT.
The best thing I found using thermowells is to put a bit of thermal paste (I use silver paste for computers) at the bottom, insert the d18b20 all the way, they stick it at the bottom with some epoxy glue.
I have some that are a few years old now and they're still working great.
 

md20_20

Supporting Member
HBT Supporter
Joined
Aug 15, 2008
Messages
44
Reaction score
1
Location
Alabama
I'm am familiar with the 85C reading. Several of the sensors I got exhibited this. However, I found that I was getting a higher voltage on the board from the power terminal associated with the humidity sensor than the power terminal associated with my temp sensor. When I started using that higher voltage for the probes they all started working (where only 1 worked on the lower voltage). I'm sorry I don't remember the specific voltages (wanting to say I was getting 3ish from the temp one and 5ish from the humidity one). And this may not be an issue for you in parasite mode.

I'm pretty sure the power for parasite mode is the issue. I have ordered some more DS18B20s and will make up a set of probes to run in powered mode. How many temperature sensors are you running using that 5V power on the humidity sensor power terminal?
 

wormfishin

Well-Known Member
Joined
Jan 8, 2009
Messages
65
Reaction score
0
Location
Dallas, GA
Got my board yesterday, hooked up a power supply, 9v 300mA. Board powers up, but I can't ping it. What do the 2 LEDs mean, D2 and D3, D3 is flashing green, D2 is solid red.

Guessing it shouldn't be red...
 
Top