How accurate is a hot water cylinder thermostat?

[Chris123]

I am looking for a way to keep a tank of fluid at an reasonably constant temperature.
How accurate are the hot water cylinder thermostats?
I'm thinking of wiring two up to to control a narrow heating range.
The field I'm heating needs to be at around 96c but if it goes over 100c it will destroy the fluid!

Thanks,

 

[lame plumber]

need something more accurate and responsive than 2 cyl stats!

 

[Chalked]

How big is the tank? And what is it for?
the main problem will keeping it all the same temp as water will be hotter at the top.

 

[Worcester]

I am looking for a way to keep a tank of fluid at an reasonably constant temperature.
How accurate are the hot water cylinder thermostats?

They're not accurate at all.

I'm thinking of wiring two up to to control a narrow heating range.
The field I'm heating needs to be at around 96c but if it goes over 100c it will destroy the fluid!

Sounds to me like someone's brewing hooch

You'll need a proper monitoring control circuit, a little bit of arduino programming should see you able to knock one up easily Home | OpenEnergyMonitor or JeeLabs and JeeLabs . net with a few encapsulated DS18B20 temperature sensors And you'll have to manage the hysteresis of the heating system as well as the monitoring system, as well as the volume and stratification of the fluid.

Simples! as Aleksandr would say.

 

[Ermintrude]

Set up a few baffels and an ah stator and some pocket probes. Easier to control it with a bulky jacket as it won't be dynamic but very sluggish - u will need a decent pid to control lag and overshoot. Quite a tight range what size of fluid and is it exothermic or endothermic? What's the viscosity like as low viscosity more turbulent convection and better distribution . What u doing?

 

[zzzjim]

.......................... if it goes over 100c it will destroy the fluid!

Thanks,

The steam produced could spoil your day as well ! Stay safe


( Lab.. thoughts , magnetic stirrer )

 

[village idiot]

Set up a few baffels and an ah stator and some pocket probes. Easier to control it with a bulky jacket as it won't be dynamic but very sluggish - u will need a decent pid to control lag and overshoot. Quite a tight range what size of fluid and is it exothermic or endothermic? What's the viscosity like as low viscosity more turbulent convection and better distribution . What u doing?

Note to self: ''Must go back to skool, must go bark to skool, must go bark 2 skool''

 

[Ermintrude]

Something this close must be reaction? Any half whit would simply boil it's **** off and condense the required fraction off?

 

[oz-plumber]

Something this close must be reaction? Any half whit would simply boil it's **** off and condense the required fraction off?

Pure Gold.

 

[Chris123]

Thanks for the replies.
It's for a tank on a small anodising line. Using a nickel acetate solution. pH is 5.5 so not too bad.
Ideal temperature is 96c, but could go as low as 90.
If it boils then it destroys the anti-smut additive and parts would need to be polished after and this is time consuming.

 

[Worcester]

Definitely DON'T use normal stats then nowhere near accurate enough.

You'll need a proper thermostatic control system, PT1000 sensors, and a controllable heating system - if you also use a pwm circuit to control the heating elements, you'll get much finer temperature control (as opposed to on/off) giving you exactly what you need.

Pretty standard stuff really.

 

[oz-plumber]

Something this close must be reaction? Any half whit would simply boil it's **** off and condense the required fraction off?

Ermi,

What were your initial thoughts on the OP?

 

[Ermintrude]

Thought was trying to keep a reaction vessel at a CT on the cheap. Problem is if it's exothermic you need to cool it and endo heat it but rate of reaction can be a function of temp so it gets incredibly tetchy!

Oh I was thinking of using a distillation column or a condensing reactor to lift reagents or improve reaction. Hence boiling .


What's the weather like down under? It's raining cats and dogs up ere.

 

[oz-plumber]

I was thinking of a meth lab.
You don't want to overheat the pseudoephidrene - particularly with the other chemicals involved.

The weather is a chilly 6C overnight and 16C during the day.
Not much rain - if any today, but the sun was shining.

 

[happyflyer]

I am looking for a way to keep a tank of fluid at an reasonably constant temperature.
How accurate are the hot water cylinder thermostats?
I'm thinking of wiring two up to to control a narrow heating range.
The field I'm heating needs to be at around 96c but if it goes over 100c it will destroy the fluid!

Thanks,

Your average cylinder thermostat is about as crude as you can get, I never tested one but I would bet if it was plus or minus 5 Deg C I might be about right, its a crude
bimetallic strip from the 1960 and has never changed, its a bit like company car allowance 45 p miles from the 1960.

If you want to keep water in a vessel at 96 plus or minus 2 Deg C you are on the wrong website, these guys work to the nearest brick, (oh what have I said) you will also need some for of de-stratification too for the vessel, I work in process and BMS control, what is your heat source, lets have some more details, sound a bit top secret to me, or a Uni project, I
may be able to help you.

 

[Chris123]

Hi, it's a final sealant tank on an anodising line. It's a Nickel Acetate based sealer, if it boils it destroys the anti-smut additive. Although this is really only a problem on black colours, it would be nice to stop it happening!

 

[Chris123]

Currently using a deep fat drier from a takeaway, which isn't accurate at all!

 

[Ermintrude]

If you can jacket tank and have some sort of stirrer or de stratification device. Ransome orbital sander plate under tank? Use a wet pocket probe or two?

Heat the jacket with water from a thermal buffer (20 times the volume of the jacket) ?

 

[happyflyer]

Currently using a deep fat drier from a takeaway, which isn't accurate at all!

The thermostat on a deep fat fryer will have an even more wild hysteresis, +- 20 Deg C no reason for it to be accurate with chip oil....they used a long rod type thermostat

 

[Ermintrude]

No he is using the fryer element and stat in the frying bath. That's what he meant to say. HSAWA and HAZOP really ring a bell!

 

[happyflyer]

Ermi,

" put mi to fish in please"

 

[Ermintrude]

K? You want your fish cooked in Ni solution? Make your fillings go funny?

 

[Bronze_tap]

Ideal temperature is 96c, but could go as low as 90.

It looks an ideal case to be heated by steam at 100C at atmosferic pressure, passing through the coil (you can rip the top of the suitable unvented stainless steel cylinder to be used as a bath.) and connect the coil pipe to another boiling tank/steam boiler, so that the condensate can run back into it. (Provided yours process i OK with stainless steel walls).

 

[Ermintrude]

If you take the lid of an unvented cylinder and boil the water in it it will splay open like a yogurt left on a stove top. The lid acts as a structural part of the cylinder . Plus open top will create a thermal gradient through bath. Why go to the hassle of raising steam when he could just use an element?

Best to keep cylinder closed but vented and have a small closed jacket. Vented essential as no one likes flashing to steam.

 

[Dotty]

Bronze tap.

I'm watching you now.....