PDA

View Full Version : Question: Air Purging



TheUnknownw
04-04-2020, 07:46 PM
Hey there everyone

I am working on a system with very small, custom designed heat exchanger (a cold plate) and i am running water/glycol through it.
Now, inside of it there is a defined path for my cooling agent with channels and different shapes. The device will have the inlet and outlet of the water, on the bottom and all the space where my cooling agent will be, will be positioned above the 2 pipes.

Now my question is: how do i purge all the air out of the system? Is there a reliable way to purge the air out? Or just run the pump and eventually the air will come out?
Because the space designed for the cooling agent space will be on top of the in/out pipes, the air will tend to stay up there, and there is no purging designed into this plate.

Looking forward for any answer.

Best Regards

seanf
04-04-2020, 09:01 PM
If you dont think running the fluid through the heat exchanger will remove the air, could you turn the heat exchanger upside down and fill it with fluid, then cap it full of fluid, turn it back the right way and connect it to the rest of the system?

Or complete the fluid system and orientate the exchanger or the whole system in a way to have the heat exchanger upside down?

Or maybe some way of vacuuming the finished fluid system and then charge it with the fluid?

Josip
04-04-2020, 09:36 PM
Hi TheUnknown,


Hey there everyone

I am working on a system with very small, custom designed heat exchanger (a cold plate) and i am running water/glycol through it.
Now, inside of it there is a defined path for my cooling agent with channels and different shapes. The device will have the inlet and outlet of the water, on the bottom and all the space where my cooling agent will be, will be positioned above the 2 pipes.

Now my question is: how do i purge all the air out of the system? Is there a reliable way to purge the air out? Or just run the pump and eventually the air will come out?
Because the space designed for the cooling agent space will be on top of the in/out pipes, the air will tend to stay up there, and there is no purging designed into this plate.

Looking forward for any answer.

Best Regards

is it possible to upload design drawing of your cold plate ... probably in the way to be semi transparent to see your internal pipes if they are vertical or horizontal regarding to position of inlet and outlet connection ...

it will be easier to give you some advice .... advice of Seanf sounds good ...

Best regards, Josip

Brian_UK
04-04-2020, 10:04 PM
As it is a custom design why was an air release point not included in the design?

There will nearly always be air entrained in the fluid mixture so it will be an ongoing problem. Hopefully the fluid velocity will carry any air through the heat exchanger.

NH3LVR
04-04-2020, 10:59 PM
Brian beat me to this.
If it will not purge air while running it will soon fill again.
Hard to see how it would not purge itself.

TheUnknownw
05-04-2020, 06:41 PM
Well, i did not make the design, and the person that did it, can't change it now...

Also is not made from pipes, it is a narrow space between 2 plates, with guide lines inside.
I know it sounds strange, also by running the pump the air should get out eventually. Tho my questions is, if the air will purge itself out 100%?

I can not move them to fill them up, they are fixed on top of the pipes and the guides inside are particularly dumb if you ask me, but yes, i have to make sure that the air gets out...

Also vacuuming is out of the picture while the filling up system consist in an open container...

Thanks for any other info!

hyperion
05-04-2020, 08:50 PM
Possibly by cycling the pump many times, this could dislodge some of the air. Repeating the process until the system is clear, might work.
How are you going to know for certain that you have cleared all of the air?
If you can have only the inlet pipe connected and very slowly fill the system, you might be lucky that the glycol/water mix pushes that air in front of it and out of the outlet pipe.

TheUnknownw
05-04-2020, 09:42 PM
Well that is the idea, i am looking for a sure way because i dont if the air will be or not inside. I will scan it with a thermal camera and if there are hot spots where there should not be, that means it will have air trapped inside.

Also the glycol/water mix comes from a pipe and fills a shape, it will fill it turbulently.

Thanks for answer. If anyone else has other ideas, i will be happy to hear them!

Best regards

RANGER1
05-04-2020, 10:20 PM
Any pictures?
Helps a lot in this situation

Tycho
06-04-2020, 03:03 PM
So the liquid is going zig zag inside the plate?

With enough velocity, and a reservoir or a air purging valve on the glycol in and outlet it shouldn't be a problem.

However I'm guessing the cold plate is also the highest part of the circuit, so then you are down to velocity to ensure that any trapped air is passed through.

TheUnknownw
07-04-2020, 08:02 PM
Thanks for all the great answers.
Will run the pump for a defined period of time and see what comes out. If air does not come out i will send it back to design haha

HwT
07-04-2020, 08:13 PM
Can you evacuate it with a vacuum pump? We have done this before.