PIREP - Engine Saver / Dehydrator

[FLYFST]

There is another related post on this topic below, but the last updates were in June of 2019, so I thought some others may benefit from my recent insights from testing the "Engine Saver"

I keep my J hangared in Central FL and since moving here in 2018 I have been concerned about the high humidity and temperature combination.  With the door closed the relative humidity (RH) can easily range between 50% to well over 80%.  In researching the dehydrators I came across the post above and decided to give the "Engine Saver" a try, which I purchased from Aircraft Spruce.

https://www.aircraftspruce.com/catalog/eppages/engsaver.php

After running a simple experiment with only modestly accurate home temperature and humidity sensors, I was surprised to find the Engine Saver dropped the measured humidity at the oil filler port by a high difference, consistent with other reported test data with more accurate and sophisticated sensors.  At the start of the test, the ambient conditions inside the hangar (with the door closed) was 50% RH and 81F, and  the conditions were 51% RH and 79F at the oil filler port.  After 10 hours, the hangar ambient conditions were 50% RH and 77F while the oil filler conditions were 20% RH and 71F.  After another 24 hrs, the hangar ambient conditions were 65% RH and 77F while the oil filler conditions had dropped to 16% RH and 77F.   The numbers would likely vary somewhat with more accurate sensors (the ones I used are supposed to be accurate to +/-5% RH and +/-1% temperature).  But, the drop in humidity was large enough to convince me the Engine Saver works well enough to continue using, particularly when the airplane may go 2-4 weeks between flights.  Since I did this test with a fresh oil and filter change after the annual and sent the oil sample for analysis, the next check will be at the next oil change to see if there is any noticeable change in historical traces for the data reported in the oil analysis.

Hank

[carusoam]

Pirep night at Hank’s house!

Great pirep for the Engine Saver...

With really good tech details too.

Best regards,

-a-

[Danb]

Hank the engine saver you referenced is the one I have it works as advertised 

[GeeBee]

I built my own engine saver. Not that hard to do although it does not have the fancy electronics. I took a Pelican box, which is water and air tight. Tapped some fittings in it and run and aquarium air pump through silica gel. I have recently been doing a lot of research on desiccants and found out that the properties of silica gel are great for bring down moist air but does not work good once down below 40% RH. At that point a molecular sieve with zeolites works better to bring it down even more. So I am developing a two stage unit in which the ambient air first goes through the silica gel then on to the sieve material. You can buy both the silica and the sieve material on Amazon for very cheap prices. 

On the output end, I found that a lot of commercial units do not filter the output. I put a Briggs and Stratton air filter over the outlet inside the box to filter the output air. For the Lycomings I found a rubber flask stopper with a hole in it works great. Put a piece of tubing through the hole and attach it to the output. The Continentals can be a little tougher because the breather tube at the filler neck will leak out the dry air. So I just put a narrow funnel in the fill tube and stick the stopper and tube in the funnel, that way the breather tube is bypassed with output going straight to the crankcase.

I think altogether my home made one is ran about 120 dollars, desiccant included.

 

[tmo]

Ha!  I am not as creative as @GeeBee nor do I have his superior workmanship skills, but I ended up with about 3ft worth of ~4" diameter plumbing pipe filled with silica gel (kitty litter) and three fittings as used for passing wires through to junction boxes.  Bottom one connected to aquarium pump, a motorcycle in-line fuel filter stuck in the top one, business end routed to the dipstick pipe on the engine.  Third one is for wiring for a BME-280 sensor (temp/rh/pressure) that sits inside the piping, and another one measuring outside parameters.

My observations:

• ~3lb of silica gel is good for ~2 weeks - but perhaps a pump with less airflow will make it last longer
• silica gel needs to be baked before first use (2h @250F) - my first attempt was a failure because I didn't do this, the second one gave 0% RH for a week

After ~2 weeks the silica gel does continue to dry high moisture air, but the outgoing air is above ~40% RH, so while likely not corrosive, it is too close for comfort.  Of course this will depend on volume of air pumped, volume of desiccant, and RH of ambient air, so my longer term goal is to have the RPi text me when outgoing air parameters are exceeded (too humid for a longer while).

I also put the silica gel in plastic boxes inside the cabin and baggage compartment to keep the air a little more dry than ambient there.  No measurements yet, need to run a second I2C bus from the RPi to get those.

I'm pretty sure my total cost was less than $100, even including the sensors I fried ;-)

Edited May 11, 2020 by tmo

[GeeBee]

Thank you for the compliment but it sounds like you are doing great work! Heck, I didn't do that fancy sensors stuff!

 

[PhateX1337]

I just got the Engine Saver from Aircraft Spruce as well and like it. Question for you guys - are you able to get the tube & stopper past the ice hole in the breather tube? The instructions say to go past it, but I'm 95% sure when I push the stopper up as far as it will let me it's not putting the end of the Engine Saver tube past the ice hole ... thoughts? Or am I overthinking it?

[tmo]

Hm, got me thinking - is going up the breather tube noticeably better than through the dipstick?  The air ends up reaching (and not reaching) the same parts either way, right?  When pushing through the breather, do you depressurize via dipstick or oil inlet (the place you pour oil in) or try to get the engine pressurized inside?

You don't need sensors if you fly often

[PhateX1337]

9 minutes ago, tmo said:

Hm, got me thinking - is going up the breather tube noticeably better than through the dipstick?  The air ends up reaching (and not reaching) the same parts either way, right?  When pushing through the breather, do you depressurize via dipstick or oil inlet (the place you pour oil in) or try to get the engine pressurized inside?

You don't need sensors if you fly often

Yep, I unscrew the dipstick and just let it sit loosely on top so no pressure builds up. I hadn't thought about putting the dry air tube down the dipstick opening - is that what you're doing? The directions that come with the Aircraft Spruce model say to put it up the breather tube, but your point sounds reasonable ... either way the dry air is getting into the engine

[tmo]

Yes, I push the dry air through the dipstick (the tube fits snugly inside) and let it vent out the breather.

[GeeBee]

Only have experience with the Lycoming and big bore Continentals. Lycoming is easy, use a rubber stopper with the feed tube through it and stick it into the dipstick tube. Continental, I put a funnel into the dipstick tube to bypass the breather outlet then stick the rubber stopper with supply tube through it into the funnel. Breather tube feeding would be nice were it not for the ice. hole. While you could masking tape the hole shut, it would be a "MUST DO" preflight item that is not easily discoverable if you fail to perform. Worse it would have dire consequences if it was needed. I've thought about one of those "expanding tube ends" where the tube end expands upon pressure, so you could push the tube up past the ice hole, then turn on the pump and it would seal in the tube, have not actively pursued it.

 

 

 

 

 

[J0nathan225]

I'm in lower Alabama and have been using this for 3 months, my silica is still white/clear, makes be wonder if its working?  I feel air coming out of the hose for what it's worth.

[jetdriven]

Silica beads by nature are white. But you add about 5 to 10% of the indicating beads, those change color when the saturated. I do not like cobalt, because it’s toxic. However, they make orange indicating silica beads which turn green when they are saturated

[tmo]

The question I would ask of indicating beans is what do they indicate.  IOW, what is the RH of the air leaving the system by the time the indicating beads change color.  All being a moot point, since my pipe is not transparent anyway, and if I open it up, I might as well change the beans out for a baked batch.

[DonMuncy]

They also make little cards that change color when dry or not so dry. (I have a supply, but do not remember where I got them.) My desiccant is in a clear glass jar and I can see when it needs "recooking".

[carusoam]

I think you guys may be over thinking some things....

Not sure what they are exactly though...

1) Opportunities for dry airflow... in and out...

2) It won’t matter if you go in one way or out the other...

3) Oil fill tube is a viable entrance or exit...

4) Case vent can do the same...

5) Case vent has a whistle hole... if your dry air escapes out the whistle hole before going in the engine...   you might consider dehumidifying your whole hangar while you are at it...

6) Using a single pass system... dry air goes in... then vents to the outside.... can use up the dehydrator material rather quickly... depending on the RH of the air entering the system...

7) Using a recycle system... may be more efficient... two hoses and a box may be needed...

8) one hose feeds dry air to the engine...

9) the second hose takes air from the engine and feeds it to the air pump in a box...

10) the air pump sends the engine’s mostly dry air back through the dehydrator, then back through the engine...

11) there is very limited amount of moisture in the engine... get it out, keep it out, don’t keep drying tons of air...

12) Or other challenges... you can’t get back to the airport, your dehydrator is fully saturated, and you are cycling outside air moisture through the system...

13) The pigmented stuff in the dehydrator doesn’t tell anything about the air going through the system... it only indicates that the dehydrator has filled to its capacity of moisture over time... sort of a level sensor... not a flow sensor...

I bet recycling is best for the system... the economics, and the ease of use....

Work on the connections at both ends of the engine...

Definitely include a safety for not accidentally leaving Air plugs connected to The engine block...  this would be considered bad form...  

Pilot A cared so much about his engine... then one day... he forgot to remove his dehydrator hose from the beloved engine...  

The accident report is going to read really weird...

PP thoughts only, not a mechanic...

Best regards,

-a-

 

[tmo]

7 hours ago, carusoam said:

I think you guys may be over thinking some things....

But of course we are :-)

I thought about a closed loop system, where the air from the engine is scavenged back to the dehydrator, but figured I don't know enough about how oil and other combustion byproducts will affect the desiccant, so I decided to pass on this idea for now.  But am nurturing it in the back of my head.

I don't think the air leaving the dehydrator can ever be more moist than the ambient air (silica traps the water and lets it out only at temperature above the boiling point), so it won't be worse.  But we're looking for "better".  The reality of the matter is the dry air is only getting to parts of the engine, so this is more of a "feel good" thing than an actual solution to a corrosion problem.

From my POV an option would be for the monitoring system to stop the pump when a threshold RH of outgoing air is met; I do have a relay that could work somewhere...  But I do like the excuse to have to go to the hangar, too.  Maybe if I could get a text reminding me to come and save the day...  As a CB, I need to find one of those USB GSM modems from the previous decade that no one uses anymore, an actual GSM module probably costs around $25...

Don's clear jar solution is brilliant in it's simplicity, with the caveat that if you're already there to check the color, why not change it out.  In case you can't go to the hangar for whatever reason.

[neilpilot]

4 hours ago, tmo said:

I thought about a closed loop system, where the air from the engine is scavenged back to the dehydrator, but figured I don't know enough about how oil and other combustion byproducts will affect the desiccant,

Back in 2012, when I stored my 64E for 4 hot & humid Memphis summer months, I build a simple (i.e. cheap) closed loop dryer using a 1 gallon clear plastic container and an aquarium air pump.  Air in the oil dip and out the breather tube.  I placed a bit of fiberglass mat between the air return and the desiccant.  After 4 months there was no visible contaminant on the pre-filter.  The desiccant was only about half spent.  If I had used an open system as opposed to closed loop, I'm confident the desiccant would have been spent prior to the end of the 4 month period.

 

PS - don't stress over the breather bleed hole.  The flow out will be relatively insignificant, due to the relative low differential pressure at the point.

[carusoam]

A raspberry pi solution....  ?

 

There would be some risk in taking impurities from the engine and accidentally drawing them through the dehydrator.....

There has to be at least a tsp full of moisture in there... after the engine has been running...

Some exhaust by-products...

possibly some oil vapors...

Most of these will fall out of solution by the time the air gets back to the pump in a box...

 

If you have an rh sensor in the system, connected to the WiFi.... what a great closed loop control and monitoring system that Could make...

Best regards,

-a-

[Rick7576]

here are some pics. of the Eng.Saver I just made. cost me less than $50. total. And so far it seems to be doing the job. I set the timer to run for about 30 min. twice a day. the timer is one of those switches you can set when you go out of town. The Silica gel will turn green when its saturated. place it in oven at 250 over night and it will be ready to use again.

[tmo]

I would be really interested to know the relative humidity (RH) of the air leaving the device when half the beads turn green.  I suspect at that time pumping the air is doing no good, or worse.

I base my suspicion on my measurements, where about 3 lbs of silica gel lasted only two weeks, and it weren't especially wet weeks.

I have a Raspberry Pi with two Bosch BME-280 sensors, one measuring air at the top of my "dehumidifier tower" and one measuring ambient air.  Initial RH of 0% lasted for a week, but after two weeks it was approaching 30% regardless of ambient air.  After three weeks it was around 50%, a 30% decrease from the 80% ambient air.  I suspect that at 50% it doesn't make sense to keep pumping new air in, better to leave what's in the engine there.  I have gotten similar results over two iterations (the 2nd one is at 25% RH after less than two weeks).

I am focusing on doing my first oil change, but once that is done and I'm back to flying, I was going to test a) a smaller pump, because how much air volume can there be inside an engine, and possibly b) having the RPI turn the pump on and off based on RH ratio.  Your idea with the timer is brilliant, and I'll do that together with the smaller pump.

[tmo]

Additional data point - my silica gel is as dry as it will be after 2h at 250F, based on weight.  1h gets most of the moisture out (75%); the decrease in weight after the 2nd hour is about 10%; after that it is a measurement error at best.

It could very well be that my silica gel (which originally was called kitty litter, but is 100% silica gel) is not as good as yours, although I suspect they are one and the same in terms of performance.

[Rick7576]

Well I guess I'll have to get my hygrometer out of the humidor and put it in the pickle jar for a week or 2.

that way I'll be able to gauge the color of the silica and the relative humidity on the hygrometer. I'll let you know what we get.

this device is a closed system so I am assuming that once the air in the crankcase has circulated through the silica a few times 

it will be pretty dry. i know there will be a small amount of outside air still entering the engine, but it cant be that much. we'll see

[tmo]

With a closed loop you are way ahead of me and yes, your desiccant should last much longer.  Dang those moving targets

[neilpilot]

23 minutes ago, tmo said:

With a closed loop you are way ahead of me and yes, your desiccant should last much longer.  Dang those moving targets

As I posted above my closed loop dryer, with maybe 3x the amount of desiccant Rick uses, easily lasted 4 months in the humid Memphis summer.