Temperature probe suggestions

[cat007]

Hi all.
Wanting a simple temp probe to stick at the exit of the wort from my pump before it returns to mash.
Currently have a BBQ meat one and while it works really well it is only accurate to 1C. I'd really like it to be accurate to 0.1C.
I did think about a STC-1000 as they're cheap and reliable but a bit bulky and over the top for that purpose.

Any suggestions?

Cheers
Hunt

 

[Wobbly74]

I use an stc for mash temp control and just bought a cheap 8mm threaded probe off ebay and tapped it into a 1/2" bsp brass cap screwed in to a T on the kettle outlet (before the qd for the hose going to the pump).

 

[cat007]

Thanks for that.
What are the requirements for replacing the temp in an STC? Will any work or does it need to have certain specifications on resistance or something?

 

[MHB]

Absolutely has to be the right sensor, check carefully before you change sensors.
Another option is just to get a threaded thermowell, this one on eBay is only about $3

 

[koshari]

using a thermowell will introduce lag to your process, will be fine whilst the temp is constant but on change the sensor will lag, may i ask what you expect to gain out of .1 deg resolution?

What are the requirements for replacing the temp in an STC? Will any work or does it need to have certain specifications on resistance or something?

the STC is set up to use the specific thermistor, you will need one with the same resistance, the lead to the sensor can contribute to the reading, this can be off-set in the settings, more accurate instruments will use a 3 or 4 wire Wheatstone bridge circuit.

regarding temperature control accuracy you could look at a PID controller, these have a little bit smarter algorithms that can anticipate and pay on the control element earlier to keep tighter temp control but this comes at an expense of cycling the heating element which generally means a solid state relay is called for, you also lose the ability of using cooling to reduce the temp as well, so it wont actively control in the cooling direction.

for a simple .1 resolution display one of these $2 displays might be as good as anything,

 

[Wobbly74]

For an STC you'll need a 10K ohm NTC thermistor. Make sure and get one with the right beta value, 3435 from memory. You can find 8mm threaded ones on ebay pretty easily. I tapped one into a 1/2" bsp brass end cap to screw into a T on the tap outlet.

 

[MHB]

using a thermowell will introduce lag to your process, will be fine whilst the temp is constant but on change the sensor will lag, may i ask what you expect to gain out of .1 deg resolution?
Snip

I'm not going to disagree, thermowells will change the time it takes for the sensor to respond.
I would like to point out that most STC controllers sensors are in effect a small chip, imbedded in epoxy (or similar) encased in either metal or plastic. When I need sensors to put into a thermowell I buy ones that have just the epoxy insulation add a dab of heat transfer paste then imbed it into the thermowell. In effect the 8mm threaded sensor (I have a handful) isn't all that much different.
I cant see any difference in the response times between the ones I made up in a thermowell and the commercial variants.
But yes, take care to get the right sensor, be careful how you imbed and locate it...
Even if you shoved the existing sensor for an STC into a thermowell (a bit of heat transfer paste is good) it wont make a world of difference, but it might be an issue in some closely controlled situations.
More important is the distance between control (sensor) and action, larger gaps make for very rough control.
Mark

 

[koshari]

Good point about using a paste . A better conductor with less mass will also make less of an effect regarding lag. But again given temperature control loops are generally slow anyway the advantages are often negligable.

I didnt really understand what you were getting at regarding the distance from the sensor to controller slowing the process down? A loop of hundreds of meters for all practical reasons will be practically no different to being directly beside the controller?

 

[MHB]

Not the controller, the control point say an element that is being controlled by the sensor.
If there is a long pipe from the heat source to where you measure the temperature, there can be quite a lot liquid that is either too hot or cold... I'm sure you got it from there.
Mark

 

[koshari]

Not the controller, the control point say an element that is being controlled by the sensor.
If there is a long pipe from the heat source to where you measure the temperature, there can be quite a lot liquid that is either too hot or cold... I'm sure you got it from there.
Mark

Gotcha.
Yes You would really need an agitator to assist the process. A pump moving the fluid will help as well.
Your get pretty crappy control with an STC regardless compared to a PID controller. Even straight proportional control would be a lot nicer. i suspect the control algorithm with an STC is simply deadband control.

 

[Wobbly74]

While the stc is pretty basic, you can set one up to do a reasonable job. I've flashed mine with the ovbs firmware (it's an old one) which has programmable pulse widths during mash control. Combined with a sensor on the outlet tap and a recirculation pump (2 ports - mash basket and whirlpool) I get about a 0.3C temperature overshot while controlling mash temp. Given you can't get the old flash-able stc's any more I'll happily upgrade to something like the smartpid if/when it eventually dies.