aluminium contact with water?

[bkiba]

is it ok to have aluminium touching my tank water?

I'm trying to figure out a good way to keep these elctronic temperature sensors in the form of ICs, from shorting out but at the same time getting a good temperature reading.

What I'm thinking is using some thin aluminium shim stock 0.005"-0.010" and pressing that up against the surface of the IC with some thermal compound, then dipping the electronic side of the sensor in liquid plastic.

So basically the aluminium will be touching the water always.... is this ok?

 

[fppf]

It will corrode fast.
They make thermal conductive epoxy or you could use Titanium.
You really don't need that good of thermal conductivity depending on the sensor. As long as it does not have a lot of self heating that needs to be dissapated.

 

[BeanAnimal]

Any epoxy or plastic will do. Seal em into a pice of pipe for that matter. As long as a good portion of the material is submerged, then you will be fine. Your temp readings may not be as fast with regards to temp swings of a few degrees over a minute or so, but should do just fine for our purposes.

 

[bkiba]

thanks for the advice. I thought about Ti too but it has crappy heat transfer. I'll probably just go with epoxy or plasticcoat then.

thanks again

 

[fppf]

Why do you think Titanium has crappy transfer?
It will still have much better conductivity than the epoxy will but not as much as aluminum.

 

[RicksReefs]

http://www.advancedaquarist.com/issues/july2003/chem.htm

 

[BeanAnimal]

It really does not matter in this application. If you tanks temp swings are to fast to be shown thourgh an epoxy interface, then youve got BIG problems.

 

[bkiba]

what I'm really wanting to monitor fast is my evaporative chiller. Fast isn't required but I'd like a fast reaponse to the controls I'm applying to it.

 

[stugray]

I am curious which sensor you are playing with. Is it the one-wire serial device?

I want to start playing with all of the one wire serial devices for a DIY controller project.

I would just dip the device in Red RTV ( automotive sealant ).

Stu

 

[BeanAnimal]

stu you may want to check out teh linksys 1 wire thread, there is a link in it to a WIKI that has some decent application specific reviews of some of the Dallas stuff.

 

[bkiba]

stu, I've used the one wires before. I had 4 of them in my evap column last year for monitoring and they worked great. My dad set me up with that system. I used 4 dallas DS1820s with that.

right now I'm working with a basic stamp using several DS1620s

I'm going to be building a monitoring and control system for an improved version of my evaporative tower.

the only problem with both of these is they are only precise to 1/2 degree C. I'd like a little more resolution but these work well.

I'd also be interested to see what you are building, please PM me with some details.

also I forgot, I'm just going to probably use this PLASTI dip stuff. There is a dipping solution, I'm trying out the spray paint verison first. That should be fine I guess.

 

[fppf]

Highest accuracy sensor I could find for a cheap price was +/- 0.33 degrees C which is not to bad.

 

[bkiba]

can you give me info on the sensor? thanks

 

[fppf]

LM92

 

[bkiba]

thanks fppf, stu which sensor are you thinking about?? is it the one wire DS1820 or something else?

thanks