Steve, before you embark on adding Peltier cooling to your system, why not use this opportunity to collect some data based on actual use of your existing system? You’ve got the ability to accurately measure your water temperature, and you could run jobs for a while that are typical of how you use the machine. You could also characterize power loss at temperatures above the 22°C “optimal” you mention.
My crude calculation indicates that if you ran your tube wide open (a full 40W, continuous operation) for five minutes, that you’d only raise the temperature of 5-gallons of water by 1°C. (That why the 5-gallon bucket is so popular!)
Hi Nate, The last engraving I did the room temp was about 20 C and two hours later when I finished the job the room temperature was just about 22C.
I was running the laser at 10ma and the parts cycle time was about 45 seconds with I’d say 20 or 30 seconds to remove the finished part and put in a new one then turn back on the laser. The material was black anodized aluminum. Soooo… what… about 65% duty cycle at the 10ma I had ran them at?
The water temperature when I finished was 25.6C and had been stable for that last half hour. That suggest the cooling efficiency was 3.4C above ambient. At least for that time period.
One calculation I read claimed it would take one hour to cool 5 gal of water 1 degree C if two chips (120 watts rated) Peltiers were efficient to 78%, then you would get about 94 watts of actual cooling performance.
If that is true, then based on that one run and other things being equal, suggest to me that I need around 250 watts of functional cooling with proper heat dissipation. i.e. Five, sixty watt chips. That’s how I decided on five or six in my post above.
Also, don’t forget that we are also fighting a significant outside influence …the ambient temperature and what it is doing. I suspect that may have as much effect on the temperature as the tube firing. I maybe should be considering foam insulation on the tank, lines and even the housing the tube sits in.
I’m not a Thermal Engineer so I’m just sort of winging it here…
On your comment about the pressure switch. I agree that a flow sensor is also needed electrically in the interrupt circuit. I thought of it later after I had finished the system. Grin, just haven’t found one I like yet.
Just an update… this posting and system has been update in a newer post. Don’s blog has this information also.