Engine dehumidifier

[BillL]

Originally Posted by Bavafa

......My first gallon lasted me only three months and I found recharging them, they are not nearly as effective as when they are new.
This time around I am going to take more care in recharging them, too much heat will kill them (turning black)
There have been multiple times that I have observe the pump continuously running, unable to bring the humidity to the 10% level and the silica is still blueish color.

Hope this adds value to everyone's attempt in building/using this.

Mehrdad, I would challenge the accuracy your humidity measurement for the engine as it is shut down but believe it is what your set-up measures.

On to regeneration. I get my beads from Sorbent Systems in CA and after lots of research and issues, here is my summary:

"It is recommended the white beads be heated in an oven at 280-300º F for 90 min. The indicator beads (blue/pink) should be baked at for 90 min @ 230-250º F. The color changes long before they are truly regenerated. These temperatures allow variation in actual temperature without damaging the silica gel."

For several reasons, I switched to white beads and find with the 290F regeneration temps the batch lasts longer. I find the oven saturates with moisture and have to open the door several times to clear it. Although they say (documented) use only 2 beads thick, I have good success with 1/4" layer. There are curves for silica gel performance and as temperature increases the relative humidity achieved is reduced, although not enough for us to become concerned. It is really all about dew point temperature.

Originally Posted by Steve Melton

I think this is after 6 flights. the return line humidity sensor would reach 80% after an hour and then drop to 10% over the next hour. 10% is it's lower reading capability. it's a cheap sensor. the return line is plugged into the crankcase breather exit. there is a small vent hole above that so it is probably pulling in some amount of ambient air. a timer will extend the life hopefully by a factor of 2X. I'm OK with that. the dehydrator gives me piece of mind when I don't fly.

This makes more sense, but the life is much shorter than my experience with 2 quarts with my original recirculating system. If you are going to use the timer (60 min) it will last longer between regenerations in open loop style. You have a huge amount of beads to regenerate. Looking at the numbers you will agree. 10% humidity at room temps is consistent with my testing as well, although the time constant due to design for mine was horrible, and modified it with a tube to blow air directly on the sensor.

I found the almost everything leaked as I tested each container, the pump housings, and the barb fittings. The cheapest jar from WalMart sealed perfectly while the more expensive jars and kitchen containers did not. With a single pass system the leak issues are less challenging. YMWV

I would not be without a dryer!!!! So NO challenge to that, regardless of the details.

[Bavafa]

Originally Posted by BillL

Mehrdad, I would challenge the accuracy your humidity measurement for the engine as it is shut down but believe it is what your set-up measures.

Bill,
Can you please elaborate on this, I am not sure if I am getting the meaning all well. FWIW, I semi trust the sensor accuracy as when I check the standard weather report, it jibs down with what the sensor is reporting.

Any info on where you get your beads or type of beads that you have had better luck with it. My first try was in the Microwave which made a huge mess as the beads headed up the plastic container and poured out. We still find the beads here and there in our house.
For the last couple of times, I have tried them in the oven, at 185F for 90 minutes. My way of indication has been change of color, perhaps I should not use the color as a way of indication since I found them very ineffective when I used them. I can see the humidity goes down from 50% to about 30% and then struggles to go below that. With the new beads, it takes 10-15 minutes to go from 50% to 10%

[TimO]

Mehrdad,

I can tell you that you'll want to get at least near 250F if you're going to regenerate them, and my experience while doing the task was the when I had them too thick in a layer, I could stir them all and they would be pinkish on the bottom and blue on top. After I stirred them enough that they were all blue, I took them out and let them cool so I could put them back in the jugs. I was surprised to see that they then all turned slightly pinkish again. That told me that I could be fooled by the heat changing their color without the beads actually being fully regenerated.

So that's why I watched a little videos on regeneration and came up with doing it for at least 3 hours. When I did that, in shallower pans (I used multiple 13x9 pans on 2 oven shelves) and stirred them occasionally, they stayed dark blue when I got all done and were indistinguishable from a brand new bottle of beads that I had. Stirring them also gives the benefit of opening the oven to vent moisture.

So I think you'll have much better luck once you try it again with higher temperatures and some different techniques. It wouldn't hurt to run them more than 3 hours if want to get the absolute most benefit out of the process, too.

As you found, I don't think the microwave is the right way to go for us on this project. And like you, I found that these beads bounce a lot when you pour them and I was chasing the ones that flew out for a while. We found one yesterday under something on the counter.

[RV7 To Go]

I used to use our oven set to 250 for a few hours to dry the beads which were in a 10"x14"x2" pan. Used to stir them a few times while drying. I now use an old crock pot that we had laying around. Dump all the beads in and run it on low (only has low or high settings) leaving it overnight. I usually try and remember to stir them once after a few hours to let the moisture out and mix the beads. I also leave the lid cracked open an 1/8" to vent the moisture. Next day turn the pot off, close the lid fully and let it cool for hour or two. Beads come out a very dry dark blue every time. One of the guys on the field drys his in a microwave. Many small thin batches. I never liked the idea much...too much work.

[BillL]

Originally Posted by Bavafa

Bill,
Can you please elaborate on this, I am not sure if I am getting the meaning all well. FWIW, I semi trust the sensor accuracy as when I check the standard weather report, it jibs down with what the sensor is reporting.

Any info on where you get your beads or type of beads that you have had better luck with it. My first try was in the Microwave which made a huge mess as the beads headed up the plastic container and poured out. We still find the beads here and there in our house.
For the last couple of times, I have tried them in the oven, at 185F for 90 minutes. My way of indication has been change of color, perhaps I should not use the color as a way of indication since I found them very ineffective when I used them. I can see the humidity goes down from 50% to about 30% and then struggles to go below that. With the new beads, it takes 10-15 minutes to go from 50% to 10%

sorbetsystems.com - - look for bulk silica gel in 2-4 mm beads. I posted the temperatures above. Hard to go wrong. I measured temperature excursions during cycles in the oven (thermocouple) and set at 240F it will range between 230F to 250F, which will not damage the indicator beads. A large pan and thin layer with regular opening of the oven door to release the moisture works for me.

I am interested in your system and the humidity controller.

Let's assume blow-by is half air and half exhaust. Exhaust is 1/2 water, so the water content of blow-by is very high, may be 15 times higher than humidity on a hot day. So your % does not sound like an accurate representation of what is really in the crankcase initially, but after circulation for a while, it should be more representative.

[Bavafa]

Originally Posted by BillL

sorbetsystems.com - - look for bulk silica gel in 2-4 mm beads. I posted the temperatures above. Hard to go wrong. I measured temperature excursions during cycles in the oven (thermocouple) and set at 240F it will range between 230F to 250F, which will not damage the indicator beads. A large pan and thin layer with regular opening of the oven door to release the moisture works for me.

I am interested in your system and the humidity controller.

Let's assume blow-by is half air and half exhaust. Exhaust is 1/2 water, so the water content of blow-by is very high, may be 15 times higher than humidity on a hot day. So your % does not sound like an accurate representation of what is really in the crankcase initially, but after circulation for a while, it should be more representative.

Bill,
I am using a Inkbird temp and humidity controller, I am not using the temp control portion only the humidity control side.
https://www.amazon.com/gp/product/B0...?ie=UTF8&psc=1


50-60% humidity is what the normal air measures so it is not surprising that after shut down, I get about the same humidity from the engine.

The instruction on the silica that I bought says to heated in the oven at 220 but I was so worried to ruin them (over heating will turn them to black and if black, they are no good any more) that I set the oven at 180. Perhaps I should try higher heat or longer.

[Finley Atherton]

Here is my closed loop dehumidifier made from 90 mm storm water pipe and fittings.



There is an empty chamber at each end and a middle chamber which holds 1 kg of beads. The beads are held in place at each end by metal screens (water tank outlet mosquito screen fitting) which should help promote even air flow through the beads.



The homemade window allows me to see when the beads change colour - clear acrylic 90 mm tube may have been an easier alternative?? Most of the fittings are glued except at the left end where it is pushed together and sealed by the black tape so the beads can be removed to dry. The 90 mm Tee piece at the other end stops it rolling around. The humidity meter probes and air in and out tubes go into the empty end chambers. The blue tube is air out after being sucked through the beads.

Rubber chair leg ends (idea from a local RV10 owner) from Bunnings (Australia) connect to the oil filler tube and breather outlet. The black rubber is reasonably soft and flexible and fits snugly - 38 mm for the oil filler and 16 mm for the breather. Note I have the antisplat oil separator but the outlet is at the cowl air outlet so is easily accessible.





I don’t have mains power in the hangar so use a 12 volt deep cycle battery. The two humidity meters with probes, aquarium pump, timer, solar panel and charge controller are all inexpensive items from eBay. The little motor pumps plenty of air but only seems to work sucking rather than blowing through the beads. The timer turns on daily at 4:30 am when the engine is cold (higher internal relative humidity) and runs for 30 minutes. The first 30 minute run brings the RH down to about 26%.

I take out the dipstick and connect this to the breather outlet for a few minutes to purge the hot moist air before connecting the dehumidifier.



Fin 9A

[thinkn9a]

But
Nice compact arrangement !

[Dean Pichon]

Originally Posted by control

If you have a 3D printer, you can make this plug that a friend designed for the oil filler:
https://www.prusaprinters.org/prints...CXkm-tOtpId5Kg

My brother-in-law just printed this cap for me using the model at the link above, but it printed a bit "small" Could there be more than one size of filler cap? He also printed this adapter part at the same time and it fits perfectly, so I would not expect there to be a scaling issue. Thoughts anyone?

https://photos.app.goo.gl/vrHAZLRwHX1cP1Zj8

https://photos.app.goo.gl/Wmyr9eiPuEuFZVPM7

Thanks,

[rv8ch]

Originally Posted by Dean Pichon

My brother-in-law just printed this cap for me using the model at the link above, but it printed a bit "small" Could there be more than one size of filler cap? He also printed this adapter part at the same time and it fits perfectly, so I would not expect there to be a scaling issue. Thoughts anyone?

https://photos.app.goo.gl/vrHAZLRwHX1cP1Zj8

https://photos.app.goo.gl/Wmyr9eiPuEuFZVPM7

Thanks,

Salut Dean - I printed it using "shapeways" web service for my Mattituck TMX IO-360 and it fit perfectly. Any chance it's a problem with his 3d printer or the scaling? If you want to fly over LSGY you can have a look at mine in person.