DIY vacuum kiln commissioning

[geekamole]

Hi all,
I've been piecing together a vacuum kiln system in my back yard for the last 20 months and have just done some initial testing with it. Some of the older threads on this board have been pertinent to the project so this seemed like a good place to post about it.

It's based on Jeff Hankinson's design from YouTube, on a bigger scale. The pipe is a ~850lb, 9' long, 36"OD, 2-1/4" wall HDPE water main pipe cutoff. I was paranoid about vacuum buckling risk so I made some rings from heavy angle that also help secure the doors. The path from the kiln body to the vacuum pump passes through 50ft of ice cold copper pipe inside the cooler, which is filled with antifreeze mix cooled by an 8000 BTU/h window unit. The condensate drains back into the big pipe and into a scuba drain tank that I can separately empty. I got a good deal on six of the Cozy rubber heating mats. An arduino serves as the cooler thermostat and can control the heating mats individually. The pump is a nice old rotary piston industrial pump and it can pull full vacuum in about 30 minutes. I cut gaskets from 3/4" rubber gym mats, which have been working great. The steel doors aren't perfectly flat since I made them from scraps. I need to re-test it but the empty tank loses less than 1/4"hg per day, if that. Wiring into the kiln is sealed with West System epoxy.

I mainly want to dry turning blanks but I'm interested to try some thick beams and cookies as well, and maybe some more normal slabs from my chainsaw mill.

I've tried two loads so far. Some half rotten cookies didn't dry completely since a heat mat made a seal between two of them, and on the second load I had better results with a few small halves of logs, cherry and brad pear. They were probably too thick to try (around 6-7") and there was a lot of checking. Still, I extracted about eight gallons of water over four days. When I try similar pieces again I'm going to start with putting the heat mats on a duty cycle instead of letting them run full blast, and stretch the process out a little bit. In some of the old threads, Den talked about steam generators so I'm curious how that feature works on the commercial kilns. I will also be adding some more instrumentation: a vac sensor, temperature monitoring, maybe a load cell scale for individual blanks, and maybe probes for a pin-type moisture meter.

The water extraction aspect is working great, though. I only turn the pump on a little each day after I drain the scuba tank. Very little moisture seems to get to the vacuum pump and it has gas ballast anyway.

[Walnut Beast]

Pretty interesting! Thanks for sharing 👍

[Peter Drouin]

 

[Rick Fisher]

Thank you for that,  I just joined here and have been thinking of building a vacuum kiln from a 500LB Propane tank. 

Heating seems to be the biggest hurdle,  the bigger the kiln,  the bigger the hurdle. 

I wondered,  if the pump pulls say 26HG and can be turned off,  I assume the water collects in the bottom of the tank ..  Could it drain into a  smaller tank with 2 shut off valves and be drained daily ?  The loss of Vacuum would be much lower ?   Would that help keep water out of the pump ?

I've looked at Welch pumps,  they're expensive.  Also looked at Liquid Ring pumps but apparently they are high maintenance and also expensive. 

Great post.  

[Ianab]

I think the issue is the the water remains a gas while in a near vacuum. It won't condense unless the vacuum is released? The idea is the vacuum pump pulls that water vapour out of the system, and that's why they need fancy pumps, as they may have corrosive water condensing in them.

One idea is an intermittent vacuum system. That means you can blow in hot air, warm the wood up, then pull a vacuum and remove some moisture. Rinse and repeat until the wood is dry. Not sure of anyone has perfected this idea, but it might be worth exploring. 

If it was cheap and easy, everyone would be doing it. But it's more like the "Rocket Science" end of drying wood.  :P

[beenthere]

Build a wood fire under the tank for heat.  

Appropriate relief valves of course. 

[doc henderson]

just like heating a kiln, pulling the vacuum will take time and energy.  maybe a 2-stage system with a wood chamber and condensation tank separated by a valve so you can dump water without losing all the vacuum.  some heat could be wood fired, but better to have a coil inside and a separate water heater and a circ. pump.  remember in a vacuum water may boil at room temp.

[Ianab]

Build a wood fire under the tank for heat.

Problem is that a vacuum chamber works like a thermos flask, the vacuum is a really good insulator. The problem is getting the warmth to the wood where it's needed. With very little air in the chamber you need direct contact, or alternatively radio frequency energy (Again getting into the rocket science) 

[geekamole]

Re: draining from a smaller tank, that is what the yellow scuba tank is in my pictures. There is a silicone vacuum-rated tube piped to the top and bottom which serves as a sight glass and once it got up to around a gallon, I would disconnect it from the main tank, drain it, then rejoin it. That little bit of air didn't even register on the vacuum gauge.

I considered using a propane tank for a long time and almost bought one, but they are really nasty inside, and it seemed like they would be right on the line of buckling once I cut a door, etc. They do ship the tanks under vacuum so there's that, but I would definitely weld some rings to the outside. The HDPE pipe is DR17 (diameter-to-wall-thickness ratio) so it's pretty resistant to the external pressure on its own.

I still want to experiment with radiant heating via heat or light bulb and green turned bowls, but I've made a couple bowls from the batch I showed and it is already a win that way.

For the last two weeks I've had the whole empty system under vacuum and it's gone from 28.5in-hg to 24in-hg, with no pumping.

The aspect of what the water is doing inside the kiln is pretty interesting. Cody'sLab has a video about freeze-drying a pumpkin in a regular freezer that is almost describing the same process. I definitely drain a lot of liquid water even while the vacuum gauge shows 28.5in. It's tempting to just consider the phase diagram for water, but the normal one is for a system at equilibrium. Since different parts of the kiln are at different temperatures, the vapor pressure is different, and there is heat flow. The water may boil in the 140F wood but will still condense in the 35F cooler, and once it is liquid there is an energy barrier that has to be overcome for it to be a gas again. I could see small bubbles forming in the silicone tube, so the system would be slightly more efficient if I insulated the scuba tank and tube, but it seems to be running pretty efficiently already.

I wonder if heated steam could be used in a sort-of-discontinuous mode--still keep the air out of the system, but heat the contents periodically with steam, which later condenses. Probably the heating rates involved would be too low.

I have a feeling that with active cooling like I have (or an outside setup and cold weather), you could probably use any old vacuum pump as long as your system is sealed up well. Just pump the system down first, then start the heating and disconnect the pump.

It's been waiting on some bandsaw upgrades but I'll probably run another test batch soon.

[Rick Fisher]

I had pictured aluminum plates with perhaps a block heater attached ?   That could be crazy but a block heater that only reached 150F might work.  

If you stacked wood on the aluminum plate,   turned on the Block heater it would heat the sheet of aluminum up pretty fast.  Another option might be hot water pipe attached to an aluminum plate and a circ pump. 

It would likely mean another aluminum plate every 2 rows of wood,  if you dried all 8/4 .. you would need 5 sheets of aluminum in 2 feet ?     A block heater may be too powerful for this idea but Aluminum transmits heat very well.  I think its second only to Copper.

Silicone heating pads may work on the aluminum as well,  there must be a way. 

If the aluminum plates are heated but not allowed to touch the pipe itself, the pipe would be very cold and the vapor would condense ? on the walls of the pipe,  away from the wood.

[Crusarius]

I wonder how efficient it would be if wrapped with heat blankets?

[doc henderson]

I am sure the wires in the blankets should be ok, not sure about the thermostats ect.  but prob. ok.  anything that is sealed off at 1 atm may burst in a vacuum.  during non vacuum periods you could go back to lights to heat the chamber, but I predict a halogen bulb would fall into the "burst in a vacuum" category.  you need a method to cycle out the water vapor.  ideally letting in dehumidified air. I believe at full vacuum the water vapor will set there at equilibrium until it can be displaced by something.  water condenses better under pressure, and boils in a vacuum.  think ac units with condenser and expansion/evaporator.  you could use commercial inert gas at 2200 psi in a tank and dry to add to the chamber to cycle out the water vapor.  just my thoughts. :P

[KenMac]

I'm not sure this is entirely applicable here in this application, but in normal evacuation (think hvac) moisture is expelled with air as it is removed from the appliance. Theoretically, an area in a vacuum of 28.66" mercury should be dry. As moisture is released by the wood it should be expelled by the vacuum pump. These kilns may not operate in as deep a vacuum as hvac but I think the theory is still the same.

[geekamole]

In my tests I've had the vacuum pump off for an entire day, and come back to a gallon of water in the scuba tank and a strong vacuum. As is I think I could do an entire drying cycle with only the initial pump down. If I had a water pump that could pull a 33 foot water column I could drain the water without introducing any air .

I did research how bulbs would work under vacuum and people say they do ok. I hadn't looked at halogens but they run at 7-8 atmospheres so one additional atmosphere is probably fine. Regular incandescents are under partial vacuum anyway. There are ceramic heat bulbs too, but they might get too hot without some convective cooling.

It would be nice if it were feasible to put the hot side of the A/C in a separate reservoir and use that to heat the wood but the amount of cooling I'm doing is way way less than the amount of heating (the kiln pipe walls do a lot of condensing too) so it would end up freezing the condensate before any serious heating happened. They do make heat pump hot water heaters, but $$$...

Den's kilns use aluminum plates with hot water circulating inside. Re: plates touching the kiln walls, it's definitely better to isolate the wood and heat from the walls. Next test I'm going to add some foam insulation feet to my platforms.

[doc henderson]

the bulbs will be really easy to test.  I am surprised that the water vapor itself does not drop your vacuum some.  when looking at gasses (blood gas) we have to calibrate for barometric pressure, altitude and water vapor.  Of course you are not at atmospheric.  as the water boils (goes from liquid to gas phase) it should displace some of the vacuum.

[geekamole]

The vacuum does drop a little bit in the first ~1hr, maybe around 0.5in-hg as some vapor builds. After that it's all about the overall rate of evaporation/boiling vs the rate of condensing.

[Stephen1]

1st cherry and pear dry with checks and movement, so do not judge by that wood. 
Why reinvent the wheel? 
My idry runs a liquid ring vacuum pump for 4 years of with no issue. 
If you have no air you need to heat the wood to boil the water out. Hense the aluminum plates between the wood. 
If you have 8" of vacuum and add a fan to the inside of the chamber, you can blow air around your wood. you can put your heat blanket around the outside to heat the chamber. 153F is the boiling Point of water. Probably easy to attain.
I shut down and drain my kiln every 12-24-48 hrs depending on type of wood.