The Development Log


 
Bryan, I too vote for the RJ45 jack.... First of all, most people wont have an ethernet cable out by the grill for internet, they will be using wireless, so the opportunity for this mishap will not be frequent.... and if you built the HM and still plug your ethernet cable into the servo/fan jack then maybe you need your router blown up! LOL

That said, I am wondering if you have decided on the pin-out for the RJ11 jack? I would like to know so I can plan the pin-out of the CAT5 jack on my "Roto-Damper" to match so it can be used without a special cable (cause you CAN plug a phone cable into a CAT5 jack and it will work for 4 or 6 wires)

Also, I have been doing some experimenting with running the probes over the CAT5 cable and have found that the servo and/or fan has effect on the temp readings from the probes. when the blower/servo sits at 0% the temp graphs are nice and flat, when the servo is moving they get a little rocky. Not sure if it is because the servo and probes are sharing the same ground wire in the cable or what. I am wondering if you have done any testing with a phone cable or ethernet cable running the fan, servo and probes?

BTW, I've been experimenting with the ThermoWorks probes and I am pretty impressed. They seem better constructed than the Maverick probes, specially the seal where the cable meets the probe. They also seem to read the temp more steady (graph is nice and flat while the Maverick line is jittery, though the Maverick probes I have right now are used and abused, waiting on new ones to compare and test), and the ThermoWorks probes seem to react to temp changes faster than the Maverick probes, specially the thin needle probe. (I got two of the regular Thermoworks Pro series probes and one of the needle probes).
 
Last edited:
It would take up more space. but a DB9 female 9-pin port on the HM board/case could be used for the servo/blower/probes and there would be no way of mistakenly placing a cat5 cable into it. You can still wire up an 8pin rj45 jack on the other end of the cable to connect at the grill.

Just a thought.
 
That said, I am wondering if you have decided on the pin-out for the RJ11 jack?
Right now it is (left to right looking into the jack hole)
1- NC
2- 12V blower
3- GND
4- 5V
5- Servo
6- NC

Can't say if that's finalized until the prototypes actually work but that's the way the first set of boards has it.

I am wondering if you have done any testing with a phone cable or ethernet cable running the fan, servo and probes?
I have not. There's a good possibility that the current draw of the servo pulls the power low all across the board. I mean my servo can brown out my RaspberryPi if I plug it in when the Pi is already booted. You could try putting 0.1uF caps across the probe/ground connections to see if it stabilizes them.

BTW, I've been experimenting with the ThermoWorks probes and I am pretty impressed.
Any idea if the coefficients are ok or not?
 
It would take up more space. but a DB9 female 9-pin port on the HM board/case could be used for the servo/blower/probes and there would be no way of mistakenly placing a cat5 cable into it. You can still wire up an 8pin rj45 jack on the other end of the cable to connect at the grill.
Except 99% of America either has an old phone cord in their house, or can get one via quick trip to the grocery store, CVS, Home Depot, etc. I don't know who still sells DB9 cables. Also a 25ft serial cable probably costs like $50 if you tried to buy it retail.
 
I mean my servo can brown out my RaspberryPi if I plug it in when the Pi is already booted. You could try putting 0.1uF caps across the probe/ground connections to see if it stabilizes them.


Any idea if the coefficients are ok or not?

Thanks for the suggestion to add a cap across the probes, I should have thought of that on my own... I will give that a try since I am experimenting with the CAT5 wiring right now trying to get that laid out in the best possible way. I will rearrange my wiring to match the RJ11 wiring on your prototype and go from there...

I just got the ThermoWorks probes last night so I haven't tested them yet for accuracy. At room temp they seem to be in the right ballpark... I will try to get them into some boiling water and ice water soon and report back on this and the effect of the caps across the probes.

Thanks for the reply and all your hard work on the HM....
 
Except 99% of America either has an old phone cord in their house, or can get one via quick trip to the grocery store, CVS, Home Depot, etc. I don't know who still sells DB9 cables. Also a 25ft serial cable probably costs like $50 if you tried to buy it retail.

You can source a DB9 <> RJ45 adapter for under $5, then all you need is the cat5 cable for the rest of the way.
 
I'm all for cat 5 jacks and cables. I have neither a crimper to make my cables, nor the store in our little podunk town to purchase anything other than a 2 connection phone cord, or 8 connection cat 5 cable.
 
I'm with Ralph. If you are clever enough to solder up everything and don't know the difference between the rPi ethernet and the output rj45 maybe this isn't the project for you.
 
I think it's also a good idea to use a rj45 connector, but would be ideal to differentiate it somehow. Do they make male pcb rj45 connectors? If not can you find a connector that is upside down or a different color than the pi connector?
 
RJ45. That would allow you to use the green pair for the motor, keeping the noise level off the temp probes (green has a tighter twist, great for noise reduction).
If you happen to plug in stuff backwards, I don't think any damage will be done. We use POE systems everywhere, and customers constantly plug those in backwards, pushing 24V into their computer jacks, without any damage.
 
This probably should not be discussed in this thread, but I'm still 80% sure it's gonna be RJ11 in the final version. Start a new thread if you think you can change my mind.
 
Keep individual jacks at the case. I think it would be tough to make a clean breakout cable, and I wouldn't want to be cutting off probe plugs. Plus no one seems to have picked up on Ralph's issue with interference.
 
Last edited:
Keep individual jacks at the case. I think it would be tough to make a clean breakout cable, and I wouldn't want to be cutting off probe plugs. Plus no one seems to have picked up on Ralph's issue with interference.

There's no need to cut and splice any probes. The cat5 cable would extend out to a remote box sitting close to the cooker. The remote box would have the 2.5mm jacks that you plug the probes into. Ralph chose to incorporate that into the server/blower unit, but there's no reason it can't be on its own.
 
There's no need to cut and splice any probes. The cat5 cable would extend out to a remote box sitting close to the cooker. The remote box would have the 2.5mm jacks that you plug the probes into. Ralph chose to incorporate that into the server/blower unit, but there's no reason it can't be on its own.

That's pretty much what I meant by the breakout cable comment. You either need to make up a breakout cable, a remote box, or cut off probe jacks. None seem too desirable for me, but that's just one guys' opinion. What's the payback, that I get to keep my heatermeter in the breezeway? Would you really trust your breakout cable / remote box to be weatherproof? It seems like you'd be back to square one - still having to make a weather proof container.
 
That's pretty much what I meant by the breakout cable comment. You either need to make up a breakout cable, a remote box, or cut off probe jacks. None seem too desirable for me, but that's just one guys' opinion. What's the payback, that I get to keep my heatermeter in the breezeway? Would you really trust your breakout cable / remote box to be weatherproof? It seems like you'd be back to square one - still having to make a weather proof container.

I am thinking the same as you. I have had a heatermeter(linksys router and Pi version) now for about 4 years. I have used it during heavy rain, snow, wind and you name it, a cheap ziplock bag or tubberware container works just fine. Also, not everybody will want or needs to have to get a 3d printed box for the HM.

Im all for one main unit
 
I am thinking the same as you. I have had a heatermeter(linksys router and Pi version) now for about 4 years. I have used it during heavy rain, snow, wind and you name it, a cheap ziplock bag or tubberware container works just fine. Also, not everybody will want or needs to have to get a 3d printed box for the HM.

Im all for one main unit

Me too. I like the RJ11 to have the option of combining the servo and fan, but having the probes extended over cat5 seems silly to me. I think it rains every time I bbq and I've never lost a HM.
 

 

Back
Top