PDA

View Full Version : "Turbo timer" for LC?



SXRguyinMA
11-17-2008, 09:28 AM
I was up late last night and this popped into my head. on air cooled machines its no big deal as they still have heat sinks to shed the heat after you shut down. But what about on LC machines? Say you just got done with a hardcore gamins session or rendering session and you temps are a little high from the CPU, RAM, VGA, etc. When you shut down the pump shuts down as well, so you have hot procs touching hot waterblocks with no circulation.

what if you made up a turbo timer of sorts. a lot of performace cars and diesels have them so when you shut the key off the engine actually stays running for a set length of time to let the turbo spool down and cool properly. Anyone thought of this? Is it even needed?

Luke122
11-19-2008, 05:49 PM
I'm not sure that it's needed, but you could use something like a delayed turn off timer to keep the pump circulating and rad fans going for a set period after shut down.

We sell them here in the store for use in automotive installations, typically with car phones and/or boosters. We set them to run for 4-6hrs after the vehicle is shut down, since most of them will be restarted and moved within that time period. If not, then the operator is likely away from the vehicle, and there's no point in leaving the devices powered up. :)

Installation is pretty easy too.. constant power in, switched power in, and a ground, then you just connect the output power to the device you want to keep running for the set period, and that's it. Set the timer and forget about it. (Set it and forget it! Yay for Ronco!)

SXRguyinMA
11-19-2008, 06:10 PM
I was just thinkin one night - where does all that heat in the waterblock(s) go if the pumps not circulating? granted they're not producing any more heat, but its not like a finned heatsink where it will shed heat whether a fan is on it or not :think:

halcyonforever
11-19-2008, 07:30 PM
Once you cut the forced circulation basically all we have left to look at would be fundamental thermodynamics.

Warmer fluid will rise, cooler sink, the water block will continue transfering heat (albiet poorly) to the fluid in it. This process continues until the entire system settles to a happy entropy level at about room temperature.

Now this could actualy be very problematic if for say your processor is mounted higher than your coolant storage or radiators. Suddenly you have hot fluid flowing towards the processor. Not enough to raise the temperature but it could keep it elevated for an extended period of time.

progbuddy
11-20-2008, 05:55 PM
You could use that secondary PSU mod in the tutorials section, and then get a timer and wire it between the two PSUs so the first shuts down but the second keeps going for a while. Maybe a few capacitors to keep the green 5v signal going?

Kill_Switch
11-22-2008, 09:31 PM
You could ahve it so a capacitor is charged and once powered down it will disipate all the charge into a cd fan for a length of time. ? Maybe, just a thought. I like the idea

*edit sorry progbuddy I somehow didn't see your last sentance. I have to start actually reading things out :)

rendermandan
12-14-2008, 02:13 AM
damn spammers!

Here are my thoughts on the subject. A water cooled system simply takes heat from the cpu and transfers it into the coolant. then the radiator and fans transfers that heat to air. Now a CPU is generating heat due to resistance. If you shut off the heat source, then there is no more heat generated and therefore whatever temperature the water block is at, at the time of shutdown, it will not increase. In fact, since the water in the block is not as hot as the waterblock itself, it will continue to transfer heat from the water block into the water until it evens out.

DaveW
12-14-2008, 02:16 PM
It's a nice idea, but it's overkill. I thought of a similar idea for the TIP-3000, basically using fans to cool the system if it's hot. However, my PSU automatically does this so I left it.

All the same, looking back I realise that even that is overkill. With no more heat being produced, it's better to let everything settle down. I think a second PSU is likely to introduce more heat back into the system than the pump will remove when it's turned off.

By that, I mean the heat cost of keeping PSU2 running will be equal to or greater than the heat removed from the system by the pump.

-Dave