Copper Air Dryer Design - Check my Math

Oh did not think you guys were running a refrigerated dryer. So did you ever run your system with just the refrigerated dryer? That what I use but nothing else. I do have motor guard filters on my powder coat equipment and plasma cutter but that’s it. I have never changed the filter on the motor guard as it looks like new.

1 Like

I have a bead dryer, I haven’t installed the after cooler yet.

I’ve only had to swap the beads out once this year. A big part of that is I usually turn the compressor on first then get everything if ready to let the air cool in the tank. I also have a second tank in-line before the bead dryer.

I live in a dry climate tho.


I hooked my dryer up straight for giggles. Just used quick couplings to see how it changes things. Yeah I it froze up so much moisture.

I can’t tell you how much my first one catches I leave the bottom drain loose. It looks like a water hose running…

Just a standard fill up my compressor runs for a few I use 140 gallon capacity worth of being a dry air tank. I can hold my hand on pipes not even warm. My example earlier was a extreme condition guess should have made that note.

1 Like

Anyone ever try is fan guard mounted radiators?

This one from Amazon looks like it’s good for 20 CFM. The M15 from the pdf would be equivalent to the Amazon link.

More money than the Derale 15300 but would make a slick installation. I also like the configuration of the ports better and it would likely have better pressure drop characteristics.


I am curious if and how much hotter it would make the compressor. I did a remote cooler so Ii would not add heat to the compressor.


I think the compressor would definitely run hotter with the radiator between the pump and the check valve of the tank.

I also believe that running hot air from the radiator then over the heads of the compressor is a robbing Peter to pay Paul situation as far as heat transfer goes.

And with the radiator upstream from the check valve the whole radiator has to unload every time the compressor cycles so you lose all that work that’s been done to the air in the radiator.

I’m sure there’s a dramatic amount of water that comes out of that radiator but much of that water would have condensed anyways in the tank in the first few seconds so you’re using this radiator to do some work that would have already been done naturally.

There’s a lot of reasons why having your after cooler after the tank is better practice.

That being said it does work but it’s just inefficient on quite a few levels.

The examples being

that you lose the volume of compressed air in the radiator every time.

The added back pressure between the pump and the tank will cause the pump to run hotter.

Much of that initial moisture would have condensed in the tank anyhow making 60% of the work it does redundant.

Cooling the radiator by blowing heat across it and then over the pumps is a robbing Peter to pay Paul’s situation.

I’m not saying it doesn’t work because it obviously does but it comes at the daily cost of efficiency and overall capacity.

I have my active after cooler about 30 ft away from the compressor.

It definitely works but in my opinion it’s not ordered correctly for the components to do the jobs that they are made to properly and efficiently.

The other factor which I don’t know will make much difference or not is that the air flow between the pump and the tank will almost be like a heartbeat pulse where the air flow after the tank will be a real steady state flow.


I have wondered about back pressure before doing this. I had actually replaced my compressor a few years ago. Did a mod on copper line changed it all to 3/4.

The benefits I see is less water in the tank.

I have one after my tanks also that never catch any moisture unless I am sandblasting for a while…

Who know which way is best?? Anything we do to help in one place hurts in another.

1 Like

Tin, I agree with a lot of your statements…
I had been looking at a rad cooler between the head and the tank…but mine would have been on a separate fan system…therefore avoiding the Peter to Paul heat transfer…
so here is my design…

I am removing the compressor unit off my 60 gallon tank…and going to mount it at a serviceable height on the wall.
coming off the compressor I will be running the air through a simple 3/4" copper wall cooler 3 runs of 5’ with auto drains at the bottom of the runs.
then into a ceiling mounted 60 gallon tank with an auto drain on that…
then when it leaves the first tank it will go through a refrigerated cooler…and into a second 60 gallon tank…
this second tank will supply a 1/2" loop around the shop for shop use…

then before the plasma I will have the beaded desiccant and then a motorguard and into the plasma…


Assuming 10 ft of 3/8 pipe for the cooler(rough estimate and for simplicity)

20cfm @ 175 psi you would loose 1.2 psi

15 cfm @ 150 psi = 0.8 psi

It’s been a while since fluid dynamics so it’s possible and likely I over simplified this.

I think some of these losses would be replaced by storing cooler air in the tank?

I think this is a good point and a very good reason to provide a second fan for the cooler and not rely on the pump pulley fan.

Another good point I hadn’t thought of compressed air is already pretty inefficient energy wise…

Looking into improving my air I have come across a couple interesting articles discussing the cost of compressed air in manufacturing.

Here is one that discusses the costs of losses in air systems. Obviously this is geared more towards large scale manufacturing but still interesting.


I see a lot of people putting their beads and filter right at the plasma.
Why not let you whole shop benefit from extra dry filtered air?

1 Like

Because you really dont need that level of dry air for things like pumping up a tire, blowing dust of a machine or even for the high volume air tools like drills, die grinders and sanders ect.

Painting, sand blasting and plasma are the only operations I do that I would use the bead dryer for.

You certainly can run all your shop air through a bead dryer but you aren’t getting any benefit from it for most compressed air operations.


I agree I have a dry air tank that has no restrictions. High volume.

Like in his case using 2 60 gallon tanks why not have all filtration before the last tank and all air is filtered.


Oh I see what you are saying… that makes sense :+1:

1 Like

Sounds like a great setup! Looking forward to seeing some pics of the new shop all setup!

1 Like

I’m going to get into some more of your reply when I get back to computer but you can’t calculate a radiator like this.

Where it may have 10 ft of line your calculating 10 ft of linear length.

What we care about is equivalent length.

With 16 180° bends in it would likely be closer to the equivalent length of 150 ft of linear pipe.

Also there is a little bit of a difference with the check valve being the line in the Sand between how one part of the system operates and the other.

I think spending some time looking at this logically from a bird’s eye view is going to help everybody out. It is a fascinating and complex situation, there’s a lot of variables.


I know I over simplified it. I should have been more clear that I left the elbows out and the calc was meant to show the loss in a 10 ft section of 3/8 pipe… not necessarily reflecting the losses of the radiator as a whole. I did the calc on a napkin fighting my 2 year old daughter for the pencil. :stuck_out_tongue_winking_eye:


She was just correcting your mistakes dad​:rofl::rofl::rofl::rofl::rofl:


She won in the end. Grabbed the pencil and ran away laughing!


I wish I had done some testing before and after the mods…

Temp and cycle time. That would told us some useful information.

1 Like

So using this estimate(150 ft) the loss would be 18 psi and 12 psi respectively …which seems like allot. would be interesting to have a gage before and after the cooler and see what the loss is.

1 Like