Heatermeter sous vide...


 
J

JoostB

New member
Hi all,

After getting a daugther and all the gadgets she needs to "survive" we also got a bottle heater. My creative mind started thinking out of the box to see what else I could do with that and soon the "perfect boiled (no not green) egg" came in sight. Problem however is that the bottle heater we got doesn't maintain a constant temperature, and even worse: you can only select 4 temperatures: way to cold, to cold, to hot and boiling.

As the idea didn't leave my head I started thinking about the Raspberry Pi and the Heatermeter. As the HM is already equipped to do a similar job (checking temperature and adjusting it if needed) I came to the conclusion that that would be the way forward. I did some research to find a 5V heating element that I could safely drop in the water but couldn't find one that fitted the idea. So now I was thinking about using the fan-out to control a relay switch that will start/stop a heating element. That way I can keep the HM hardware intact for my BBQ's.

Problem is that the fan mode is incrimental and I figure that a relay switch will only work at 100%. What I found in the documentation was that I could set a minimum fan speed. If I will put that on 100% will my idea work? Or am I missing something?

Any other way to use the HM to get a constant water temp would also be appreciated ;-)
 
Easy! Just get a SSR and hook the input to the "blower" output. Set the HeaterMeter for "Pulse" mode with 100% min and 100% max speeds. Run the output of the SSR to an electrical outlet, set the bottle warmer to "BOIL" and you're done. No hardware modifications and you can attach the SSR directly to a ethernet/blower cord. I'd put that inside an outlet box though because the other side is 120VAC!

This is a problem if the bottle warmer doesn't turn on the heater when it is plugged in because basically heatermeter will unplug and plug it in over and over to maintain the temperature. If that's the case you can buy immersion heaters that will work in any pot you want.
 
Great! Just as I expected, actually even simpeler ;-)

I am already looking at a good (i.e. waterproof) housing for the HM and the SSR as the current bottle warmer and 220V doesn't quite mix with water in my experience. And anyways I want something to protect the HM when it's controlling the Webers outside...

At the moment I'm thinking of using an electric stove as a heat source, the bottle warmer will stay bottle warmer until I get one that I can dissasemble completely to get the heating element free from the "control" unit. But again: as water is still involved I'd rather be safe than in the dark ;-)
 
Hi Bryan,

Got all the parts today but still have one thing I'm not sure of: my blower cord has two wires, when I look at the relay board I guess I need three: power, ground and a "trigger" wire. Do you (or anyone else) have a suggestion what to put where?

Tanx!
 
A relay board? Did you get a relay or a solid state relay? Relays are unsuited for this application because they wear out quickly being turned on and off every ten seconds. The power input would be whatever DC the relay was spec'ed for and you might have to step down the 12V if they are 5V relays? I'm not sure what you've got there.

If it is an SSR, then there will be two AC terminals and two input terminals. The input terminals you know what you got there, and the AC terminals you split the hot line going to your heater and insert the SSR in the middle of that one line. The neutral like stays intact. Actually I am not that familiar with international wiring. Do you have a hot 220V line and a neutral line?
 
Usually a picture says more than... So here we go ;-)

4-x-relais-5volt-no-nc.jpg


This is what I have now. On the left there are 4 free pins: VCC, GND, IN1 and IN2. The jumper sits on COM and the second GND.

Will this be enough to get it working or do I need something else?
 
What you want is something like this:

qN6lsu6m.jpg


You can pick them up on eBay for under $10, such as this one.
 
Last edited:
Bryan Mayland said:
Easy! Just get a SSR and hook the input to the "blower" output. Set the HeaterMeter for "Pulse" mode with 100% min and 100% max speeds.
Click to expand...

In my SSR experimentation today, I found both Pulse and Voltage mode worked fine with the SSR. Since min and max are 100%, is there really going to be a difference between pulse and voltage?
 
Hi Guys,

The mailman dropped the SSR in my mailbox this weekend so I've connected it all for my first experiment. The good news is that the basics work. But... The problem is that I somehow don't get the HM to blow 100% or 0%. It still keeps on trying to put out all the percentages between 0 and 100. I've put my PID Output parameters below. Am I missching something?

PID Output Parameters
Bias 4
Proportional 3
Integral 0.005
Derivative 5
Fan speed min 100%
max 100%
Invert output "off"
On at max only "on"
Servo pulse duration 600 us - 2500 us
Invert output "off"
Full open/close only "on"

Thanks for your help!
 
Do you have the "Startup Max" set to 100% as well? (you didnt mention that setting, so either you overlooked it or are running older software version that doesnt include that setting). Startup Max aside, with the fan min/max set to 100% it should be an on/off type of situation. Not sure how the HM will work with the fan min/max set to 100% and "on at max only" selected, never done that, but you shouldn't need the "max only" option with the fan min/max set to 100%, so I would deselect that option and see how it goes... the servo setting will have no effect on what is going on here if you are using the fan output to drive your SSR...
 
Then I guess I still have an old(er) version: "OpenWrt Attitude Adjustment 12.09 | Load: 0.41 0.14 0.11 "

All in all it doesn't seem to accept min 100% max 100% as it still uses less than 100%.

That aside I found out that not touching it also helps. After a couple of cycles it only drops to 88% befor going up to 100% again. The temperature drop is still 2 degrees so keepting it around 63 degrees seems doable...
 
Did you try removing the MAX ONLY option for the fan? The way that works, it makes the fan ignore anything under 100% as far as the blower goes, you do NOT want that... What you want is the HM to tally up the time spent under 100% and output that in bursts of 100%... So set all fan speeds at 100% and DESELECT "on at max only"
 
On at max only "on" < As Ralph mentioned, this is the cause of the problem.

When you set the Min to 100%, the HeaterMeter no longer uses PWM on the output, it uses a continuous 10 second loop and triggers the output based on the PID output divided by 100 (fan min speed) times 10 (seconds)

ie:

If the PID output is at 20%, then 20 / 100 * 10 = 2 seconds on, 8 seconds off
If the PID output is at 60%, then 60 / 100 * 10 = 6 seconds on, 4 seconds off
 
Hi all!

Good news: it all works now. Indeed on at max only off and the temperature stays very stable. +/- 0.5 degrees is realy nice so I'm happy with that. Time to start experimenting!
 
It's hard to say without knowing what you are controlling, but you PID settings look like the B & P settings could cause what your graph is showing.
B=0, rest to stock settings & see what it does.
 
It's a slow cooker (crock pot in US I think). It has 3 temp settings so I changed it to 'warm' and that made the fluctuations smaller but I'm seeking that flat line!

I'll try some different settings. Problem is I don't really know what they do so was hoping to copy someone else's!

Anyway, my small brisket that I smoked on bbq for a couple of hours then sous vide for 48 hours was delicious. Better than expected!
 
My suggestion is the stock settings. The PID shouldn't do a whole lot & you really don't have that much variation. Zoom out & the line will be smoother....I'm guessing your response time is too slow, so the PID thinks it needs to add power, until it starts to overshoot.
A simple explanation, not completely scientific.
b=voodoo :) I don't remember what it does. Make it 0, that's stock
P=Proportional - the farther away from setpoint you are, the bigger the response. I think yours is too big
I = Integral, or area under the curve (or over it if you are below setpoint). Kinda like a running average.
D = Derivative - how fast is the temp changing (look ahead & see if you need to cut output to reduce overshoot)
 
You must log in or register to reply here.

 

Back
Top