Engine Dehumidifier DIY

don't forget.....the valves. one or two will be open, but not to the breather.

But their openness is on the top side of engine “sealed” off by piston rings... a closed loop dehydrator really isn’t a sealed system, but more like our homes in winter, closed up not sealed up...

I run mine open as I’m putting stuff away and closing door and wrapping cowl etc- to blow the nasty compression gassed out and on the right day you can see the whisps of steam coming out. Then before I leave I insert the draw tube to the dehuey to recycle as much of the air in the system as possible so I’m not always drawing full ambient air having to condition it
 
I run mine open as I’m putting stuff away and closing door and wrapping cowl etc- to blow the nasty compression gassed out and on the right day you can see the whisps of steam coming out. Then before I leave I insert the draw tube to the dehuey to recycle as much of the air in the system as possible so I’m not always drawing full ambient air having to condition it

This is one of the things I was thinking about when designing this in my head - Is there any way to sense when some of these harmful gases are completely out of the system? Or, rather, close enough that further venting would be ineffective?
 
This is one of the things I was thinking about when designing this in my head - Is there any way to sense when some of these harmful gases are completely out of the system? Or, rather, close enough that further venting would be ineffective?

Idk, but I’m glad I’m not the only one nerdy or scared of an OH bill to care to talk this through with :)

I think one would have to be able to have some scientific unit to measure the air quality on the vent side to know for sure.... though I think my air compressor or your fan idea would do a better job of this than my fish pump as it’s so low flow... I wish I could feel a small breeze amount coming out at lease in what maybe we could call the purge portion of dehydrating process.

I’d say if we knew the cubic feet of air or estimate in our cranks if we could pump/blow several fold that volume through it wud clean it out pretty good. I wonder if one could notice a smell difference out the vent tube once combustion gasses are out or if the oily smell wud just be the same...

Excellent thought though on figuring out how to get it all out first... will silica absorb those corrosive gasses I wonder once we close the loop?

Btw if we want to go down a rabbit hole we could also discuss incorporating nitrogen into our flow as well....
 
Btw if we want to go down a rabbit hole we could also discuss incorporating nitrogen into our flow as well....

Why stop there? We could flood the engine with an inert gas like Helium or Argon too. :) But yeah - No oxygen, no oxidation!
 
Wasn’t nitrogen used or is for storage of turbines for its corrosion inhibiting ability ?
 
How much water and acids are in suspension in the oil?

I’m not sure there’s anything I can do to mitigate that even if I knew... so I can only get it when it’s traveling around airborne...

I mean I guess in wild theory if there was an oil filtration system that could remove those you could “change your oil” after every flight... which even if doable it would be a big time and financial investment... we are looking for something to take an addition step in right direction that’s not overly cumbersome... and hooking up the dehuey takes all of 30 seconds addition to packing up...
 
Last edited:
I’m not sure there’s anything I can do to mitigate that even if I knew... so I can only get it when it’s traveling around airborne...

I mean I guess in wild theory if there was an oil filtration system that could remove those you could “change your oil” after every flight... which even if doable it would be a big time and financial investment... we are looking for something to take an addition step in right direction that’s not overly cumbersome... and hooking up the dehuey takes all of 30 seconds addition to packing up...

Changing the oil every 3 months / 25 hours goes a long way.
 
How much water and acids are in suspension in the oil?

Well, with a drier environment inside the engine, that water will go away faster and in a safer way. The acids, there's not much that can be done that is practical, aside from regular oil changes.
 
Well, with a drier environment inside the engine, that water will go away faster and in a safer way. The acids, there's not much that can be done that is practical, aside from regular oil changes.

Without >212F at sea level, the water in the oil is going no where.
 
Evaporate, but what happens when it is underneath a layer of oil??

That depends. How long does it take for the water to settle below the oil after a flight?

Also, does the oil sump heater potentially get warm enough to either create convection currents that essentially keep the water in solution, or get some of it above the boiling point locally, or both?

Finally, does water that's trapped underneath the oil in the sump even matter when it comes to engine longevity?
 
Those are the questions I have too, haven't yet had anyone address those directly but someone over on the red board posted testing done years ago. If I find it I will post it hear.
 
Pour a glass of water out onto a flat, non-porous surface. What happens to it?


Except the oil and acids in your oil are microscopic emulsions that aren’t going to evaporate regardless of how much dry air you blow in the case.
 
Well, I took my laser thermometer to the airport for fun... More on that below.

I also shined a bright light down my oil filler tube. It gets way narrower prior to entering the lower end of the engine than I thought it did. It also has another exit tube above the engine - What would that be? It's got the opening for the cap at the top, the opening into the engine at the bottom, the breather comes off one side, and I could see that farther down there was another opening... No idea what that goes to, but I was looking at it through the oil door and the Mooney is very tightly cowled and has the intake on top of the engine so it was very hard to see in there. I'll have to take a better look next time I have the top cowl off.

In any case, I'll likely need a rigid tube to go from the cap down into the engine when I build my dehydrator. And, that tube is going to need to be maybe 1/4" in diameter, so I'm going to need more than an aquarium pump to get a reasonable airflow into the engine. I'm also thinking that I'll go ahead and use the breather as the "exit" rather than building it into the cap, provided that whatever that other connection is won't be harmed at all by a bit of extra air pressure.

Finally, because of the size of the entrance from the filler tube into the engine, there's no way I'm going to be able to actually take measurements of temp and humidity inside the sump itself, unless there's a way to have a REALLY tiny probe that can be fitted to the end of the tube that will go down into the engine. Anything measuring more than 1/4" thick is definitely not going to work, and 1/8" would be better. I wish I could find a spec for the size of the filler tube so I knew how much space I really had to work with, but... I'll probably just need to have the measuring happening in the filler tube, and thus I'll need to move enough air through the engine that the air in the filler tube is at least reasonably representative of what's inside the engine.

Now, on to the laser thermometer! The cylinders were a little over 100ºF when I got to the hangar, thanks to the engine heater, cowl plugs and blanket. After shutdown, I measured the front cylinder on the left side (#5?) at 234ºF, with an OAT in the low 50s. After I moved the plane into the hangar, I measured the oil dipstick at 130º and when I pointed it down into the sump via the filler tube, it was 165ºF in there.

The prop was 51ºF near the tips, and 56ºF near the root of the blades. It would be interesting to measure both of those over time starting at shutdown to see how much of a heat sink the prop is. I'd bet it's a pretty good one. I'm going to measure this again *before* flying to see how much heat from the engine heater is coming out of the prop. I'll measure tips, blades just outside of the blanket I wrap around the roots/spinner, and inside the blanket immediately after taking it off.

Anything else that would be interesting to measure?
 
Well, I took my laser thermometer to the airport for fun... More on that below.

I also shined a bright light down my oil filler tube. It gets way narrower prior to entering the lower end of the engine than I thought it did. It also has another exit tube above the engine - What would that be? It's got the opening for the cap at the top, the opening into the engine at the bottom, the breather comes off one side, and I could see that farther down there was another opening... No idea what that goes to, but I was looking at it through the oil door and the Mooney is very tightly cowled and has the intake on top of the engine so it was very hard to see in there. I'll have to take a better look next time I have the top cowl off.

In any case, I'll likely need a rigid tube to go from the cap down into the engine when I build my dehydrator. And, that tube is going to need to be maybe 1/4" in diameter, so I'm going to need more than an aquarium pump to get a reasonable airflow into the engine. I'm also thinking that I'll go ahead and use the breather as the "exit" rather than building it into the cap, provided that whatever that other connection is won't be harmed at all by a bit of extra air pressure.

Finally, because of the size of the entrance from the filler tube into the engine, there's no way I'm going to be able to actually take measurements of temp and humidity inside the sump itself, unless there's a way to have a REALLY tiny probe that can be fitted to the end of the tube that will go down into the engine. Anything measuring more than 1/4" thick is definitely not going to work, and 1/8" would be better. I wish I could find a spec for the size of the filler tube so I knew how much space I really had to work with, but... I'll probably just need to have the measuring happening in the filler tube, and thus I'll need to move enough air through the engine that the air in the filler tube is at least reasonably representative of what's inside the engine.

Now, on to the laser thermometer! The cylinders were a little over 100ºF when I got to the hangar, thanks to the engine heater, cowl plugs and blanket. After shutdown, I measured the front cylinder on the left side (#5?) at 234ºF, with an OAT in the low 50s. After I moved the plane into the hangar, I measured the oil dipstick at 130º and when I pointed it down into the sump via the filler tube, it was 165ºF in there.

The prop was 51ºF near the tips, and 56ºF near the root of the blades. It would be interesting to measure both of those over time starting at shutdown to see how much of a heat sink the prop is. I'd bet it's a pretty good one. I'm going to measure this again *before* flying to see how much heat from the engine heater is coming out of the prop. I'll measure tips, blades just outside of the blanket I wrap around the roots/spinner, and inside the blanket immediately after taking it off.

Anything else that would be interesting to measure?


Man idk what to tell you on the tubing... I’m such a visual learner I’d have to see it to get all you meant. I agree a aquarium pump is crazy low on “oomph”... I thought I shared on here but looked back and didn’t on my idea to use a small air compressor from Harbor Freight Aviation. With a valve you could adjust for whatever airflow you wanted... the act of compressing bleeds off some humidity and an air/water/oil separator could be put on outbound line.... I’ll look to see if I can find my rough sketches of the idea... if you do a closed loop you could mount a hygrometer in-line both input air and exiting air, exiting air should give you rough idea of rh inside crank, though if located in area temp is far different that would throw it off, but would give a decent indication...

If you have a temp gun could you get the front bearing area of engine when you arrive at the airplane one time with wrapping prop/spinner up and one time leaving prop and spinner naked... to see the differential. I’m with you that I bet that prop is a first class heat sink...
 
Last edited:
Man idk what to tell you on the tubing... I’m such a visual learner I’d have to see it to get all you meant.

Well, this is what I have to work with:

IMG_3969.jpg

That hole at the bottom that goes into the engine is 1/2" diameter, tops. The hose clamp and black rubber hose on the right just above the dipstick is the breather. You can see that there's another opening near the bottom of the filler neck on the left, that goes to parts unknown.

What I'm thinking now is having a spare dipstick/filler cap with rigid tubing instead of the dipstick, that will go straight down into the middle of that tiny hole at the bottom and push air directly into the bottom of the engine... Or, I suppose, pull it out!

Which brings up the question. Which is better, pressure or vacuum? If the crankcase as at a positive pressure relative to ambient, some of those awful acid-laden gases would get into the cylinders after leaking past the rings. On the other hand, once the engine clears out and has mostly-fresh air inside, that air inside will be drier than the outside air, so would help to get moisture out of the engine.

If the crankcase is at a negative pressure relative to ambient, then outside air might get into the cylinders, and a bit would leak into the crankcase through the cylinders. I'm not sure which is worse: Dry, acid-laden gases, or moist "clean" air.

I agree a aquarium pump is crazy low on “oomph”...

Maybe... Or is it? They put air through a tube that is pretty small as well, so it might not be so bad... A heavy-duty aquarium pump might do just fine. Something like this: https://www.amazon.com/VIVOSUN-Comm...ocphy=9018723&hvtargid=pla-322824042950&psc=1

If you have a temp gun could you get the front bearing area of engine when you arrive at the airplane one time with wrapping prop/spinner up and one time leaving prop and spinner naked... to see the differential. I’m with you that I bet that prop is a first class heat sink...

I'll have to see if I can shoot down in there without pulling the cowl apart. ;)
 
Man your set up complicates the hook ups!

So are you now thinking open loop? Push or pull air in, vs recycling it?

Could you have long straight in tube to slip just inside that lower portion that thins down... and some form of tube you could mold to the right angle to slip into that vent tube???

That’s funny you brought up that big pump, I was just googling bigger air pumps and seen that one. Looks like for a few more bucks you can get a quieter version...

As I “engineer” this in my head I’m weighing out constant noise of running a pump like that vs an on/off cycle of a compressor pump... aquarium would be quieter but constant, compressor would be louder but periodic... wondering how much resistance those big aquarium pumps can handle, pushing or pulling through desiccant without getting hot and how long the diaphragms last...

On the temp thing that might be tough with a tight cowl. Maybe same test one wrapped one naked just of prop root and temps...
 
I've always kind of wanted it to automatically switch from open to closed loop as well. Need to find out some sort of actuator for that.

BTW, there's a couple of really good threads about this over at BeechTalk, too:

https://www.beechtalk.com/forums/viewtopic.php?f=37&t=165231
https://www.beechtalk.com/forums/viewtopic.php?f=37&t=76166

I’ll check that out, yea a valve on timer could do it, one way valve is turned it’s exhausted to atmosphere and other way into the closed loop system... though I guess it would require a timed valve to allow ambient air into the system while exiting air is in “purge” mode...

Maybe I better start learning to use an arduinos or whatever those little diy computers are called! We’re getting quite the rig conceptualized!
 
So are you now thinking open loop? Push or pull air in, vs recycling it?

I've always thought about it being a combined (and automated) open/closed system. It would be open until the humidity inside the engine got within a specified amount of ambient, at which point it would switch itself to closed. No point putting all that junk into the dessicant right after shutdown, and having it open for that time period would help get rid of the corrosive gases as well.

Could you have long straight in tube to slip just inside that lower portion that thins down... and some form of tube you could mold to the right angle to slip into that vent tube???

I'm thinking that I'll have a long straight tube that goes down through the lower portion, and just hook the other end to the other end of the breather. Regardless of which way the air is flowing, the pressure differential would move the air through the breather. I just need to be sure that whatever that other tube is on the left doesn't go to something that might be harmed by the additional pressure (be it positive or negative).

That’s funny you brought up that big pump, I was just googling bigger air pumps and seen that one. Looks like for a few more bucks you can get a quieter version...

As I “engineer” this in my head I’m weighing out constant noise of running a pump like that vs an on/off cycle of a compressor pump... aquarium would be quieter but constant, compressor would be louder but periodic... wondering how much resistance those big aquarium pumps can handle, pushing or pulling through desiccant without getting hot and how long the diaphragms last...

Well, it seems to me like the "stones" they put the air into these days must have a fair amount of back pressure to them.
 
I’ll check that out, yea a valve on timer could do it, one way valve is turned it’s exhausted to atmosphere and other way into the closed loop system... though I guess it would require a timed valve to allow ambient air into the system while exiting air is in “purge” mode...

Maybe I better start learning to use an arduinos or whatever those little diy computers are called! We’re getting quite the rig conceptualized!

Yep, I was always planning on doing something Arduino-based that would measure both ambient and engine moisture levels and open and close the loop as necessary. :)

I think this rig is maybe made by one of the guys in that Beechtalk thread, and appears to do a lot of what I'm interested in doing: https://www.enginedryingsystem.com
 
What I know about aircraft engines could be carved into the head of a pin with a hammer, but FYI, we use desiccant and dryers for our transmission lines. These are big, air-dielectric coax runs that have to be pressurized with very dry air. In the old days, we used nitrogen, but nowadays, we use pumps that force air through a silica or carbonate-based desiccant.

So ... the only tip I can offer is to remember that the desiccant goes "stale." Here in Alabama, that can happen with depressing speed, too (less than a week in really humid weather). You can recharge the desiccant by baking it in the oven. If you use the silica that has the blue indicator beads, you simply bake it until the beads turn dark blue again. When they stop turning blue, even after baking for a while, the desiccant has gone bad and needs to be replaced.

If the kitty litter that you're using doesn't have reliable indicator beads in it, go to your hardware store, in the pneumatics section, and buy one of those little inline desiccators. Break it open and dump the blue beads into your litter. Cheap, effective and you feel even more like McGuyver.

(Or, you could buy industrial-quality desiccant with the blue beads already in it, but where's the fun in that?)
 
As mentioned the desiccant isn't going to last very long and going to become a royal pain. I would scrap that and the harbor freight compressor, it might not last that long either.
 
As mentioned the desiccant isn't going to last very long and going to become a royal pain. I would scrap that and the harbor freight compressor, it might not last that long either.

I shopped for commercially available dehydrators a few years ago, and one had an interesting solution to this problem: It held the desiccant inside a box that also contained a heating element, and when the desiccant lost some portion of its effectiveness, it automatically cut off the line to the airplane, vented the box and baked the desiccant.

I definitely don't relish the thought of dumping the desiccant into a box to take it home and bake it myself.
 
OK, a few temperature measurements today. Here is the airplane as it sits in the hangar:

IMG_3973.jpg

I'll call the blades 1, 2, and 3 starting from the left (so 2 is the vertical blade). As you can see, blades 1 and 3 are partially covered on the front face with the blanket, while blade 2 is mostly uncovered.

The temperature measurements were a little weird to me, especially on blade 3. I measured the temp of each blade at the tip, roughly mid-span just outboard of the blanket, and at the root. OAT (well, not outside, but in the hangar) was in the low 50s. The following were the measurements:

Blade: Tip, Mid-span, Root (ºF)
1: 54, 65, 80
2: 56, ??, 77 (didn't measure mid-span because it wasn't covered with a blanket... But I probably should have.)
3: 58, 60, 78

What I found surprising was that the #1 blade had an 11º difference between tip and mid-span, while the #3 blade was only a 2º difference. It would seem that maybe heat conduction is better on the #3 blade, because its tip was warmer and its middle was cooler?

In any case, clearly there is plenty of heat loss happening through the prop.

For other measurements - I aimed it as far in just behind the spinner as I could, and got a reading of 94ºF. The #5 cylinder was 125ºF, the #6 cylinder 135ºF, and aimed down the filler tube into the sump I got a reading of 148ºF!

I didn't think to measure it right away after removing the blanket, but a couple minutes later I measured the exterior of the cowl: 56º near the bottom on the exposed portion, 84º at the top under where the blanket was. I also measured the right-hand exhaust pipe, which was 50º at its tip (basically OAT), and 53º where it comes out of the cowling. I was expecting this to be somewhat higher, but I guess the metal is much thinner and it's farther from the heat sources, so conduction isn't helping it nearly as much as it does at the prop.

It would be interesting to see what all these measurements were on the coldest day of winter. Even the prop was well above the dewpoint today, but I'm wondering if that would still be the case on an exceptionally cold day.
 
What I found surprising was that the #1 blade had an 11º difference between tip and mid-span, while the #3 blade was only a 2º difference. It would seem that maybe heat conduction is better on the #3 blade, because its tip was warmer and its middle was cooler?
What is observational error on your temperatures? What variation does your thermometer give for repeated measurements of the same item at the same temperature? Is it possible the differences are just due to statistical variation?

Not that it matters very much- just note that depending on the variation on the measurement technique, the data may not be good enough to draw many conclusions.
 
What is observational error on your temperatures? What variation does your thermometer give for repeated measurements of the same item at the same temperature? Is it possible the differences are just due to statistical variation?

Not that it matters very much- just note that depending on the variation on the measurement technique, the data may not be good enough to draw many conclusions.

I thought of that too, and so I took several more measurements of those two blades. The results remained consistent, even though of course I was hitting a slightly different part of the blade each time.
 
As mentioned the desiccant isn't going to last very long and going to become a royal pain. I would scrap that and the harbor freight compressor, it might not last that long either.

As long as it’s designed to change easily it’s no biggie- you can buy silica gel as a premium kitty litter at Walmart- crazy cheap
 
I definitely don't relish the thought of dumping the desiccant into a box to take it home and bake it myself.

No need for that. A $50 microwave at the site takes care of it. Cook for 5 minutes, stir, repeat until the blue indicator beads are the right color. We've been doing it for years. Works like a charm.

One thing, though ... I would imagine that silica dust would cause problems in the engine. Better put a good filter between the dehydrator and the crankcase. Otherwise, it will fill with silica silt, and I imagine that might not be good.
 
One thing, though ... I would imagine that silica dust would cause problems in the engine. Better put a good filter between the dehydrator and the crankcase. Otherwise, it will fill with silica silt, and I imagine that might not be good.

And that's something I've been thinking about. I'm not a fan of the desiccant, nor am I a fan of the cycling dehumidifier... The compressor thing interests me. Are there any other options for dehumidification?
 
And that's something I've been thinking about. I'm not a fan of the desiccant, nor am I a fan of the cycling dehumidifier... The compressor thing interests me. Are there any other options for dehumidification?

There are all sorts of air dryers. The desiccant approach is cheap and fine for our application, so that's what we generally use. Honeywell makes some units that cool, and then reheat, the air. (Think air conditioner, then heater. Makes sense.)

Membrane dryers that work on a type of reverse osmosis are also available, but they typically run at very high pressure (at least 90lbs on the ones I used), and need a lot of pre-filtering.

I strongly recommend an OIL-LESS compressor before either. I don't care how many oil separators you put after the compressor, your desiccant or membrane is going to get gooey and sticky. (Speaking from experience.)

Here are some membrane dryers. Grainger isn't the cheapest, but you can look at them. (Then again, everything in aviation is ridiculously expensive, so prices that make me pass out probably wouldn't even phase you steely-eyed missile men.) :)

https://www.grainger.com/category/pneumatics/compressed-air-treatment/membrane-air-dryers

I wrote a detailed article for about the Andrew membrane type that we had on our 6" coax at 101.1 in Cullman, AL, many years ago, but I can't find the link now. Basically, the treatment chain, from input to output (I'm leaving out the regulators) ...

- Input dust filter
- OIL-FREE compressor, run as high a pressure as you can (we were at 90-100lbs, as mentioned)
- Air cooler, the bulk moisture condenses
- Moisture separator to remove the bulk moisture
- Membrane dryer to polish it off, lowers the final air dew point to -20 to -40 degrees
 
The air conditioning type won’t really work if we are t in a heated hanger would they? I’m interested to see the membrane type when I can get on computer later...

Oiless compressor makes all the sense in the world. I would think a simple particulate filter would grab any silica dust... the lack of temperature demand of desiccant is a huge advantage to desiccant too...
 
Last edited:
Back
Top