Muses About A Low Mass Electric Furnace

There’s been a lot of discussion and recent builds focusing on low mass furnace construction. I think principle has a lot of merit for hobby furnaces. My electric was initially an IFB furnace. I really liked it. I then built a dense refractory version. It’s indestructible but takes a lot longer to initially come to temperature than the IFB version. So it got me thinking about materials and construction methods and I've been experimenting.

I took 1" fiber blanket scrap, separated it so it was 1/2" thickness so I had a long strip a few inches tall, brushed a coat of rigidizer on each side, and rolled it up. At the moment, it's very firm but may not be fully cured and I may need to add some heat. I need to check on that tonight.



This construction method yields a rigidized circumferential wall that recurs multiple times as you spiral outward in the diameter so it has a lot of structural integrity even if the rigidizer does not 100% penetrated and fully rigidize the wool. You could imagine just starting the wrap around a polystyrene plug that is removed or burned out creating a vent hole in the lid. I ordered some Satanite since that seems to be to go-to hotface material of choice for ceramic wool. It arrives today. The idea was if the Satanite was applied to the underside of the rolled up lid, it would bond to not only the wool, but potentially to the rigidized wool "walls" than penetrate throughout the thickness of the lid. The topside of the lid may just get a thin coat of rigidizer.

I'd use a similar technique for the furnace body and a hybrid of materials for the base. For the body, I'd mold in grooves in the rigidized wool to house the resistive coils for my electric. Sort of like the plug I used for the dense refractory version. I could paint on a hot face, dry, and then hand lay rigidized wool.



I'm not sure I would even need hot face at my temps and will experiment a little more with that. For my 10" bore lift off furnace that can accommodate an A20, the total refractory mass using 8lb/ft3 ceramic fiber wool would be 14lb. The furnace body alone without the base and lid in the dense castable version was ~70lbs. 14lbs of wool would be a dramatic reduction in mass and time to temperature and dramatic increase in insulation.

I’d also like to experiment with the use of sodium silicate as binder/rigidzer. My thinking is it could be diluted to the desired consistency, and chemically cured without heat with the use of CO2. This would afford the ability to cast very complex shapes with the use of lost foam molds since it would require no heat to cure. Just put it in a garbage bag and full of CO2 and let it sit and diffuse. Sodium Silicate is also considerable less expensive than colloidial silica rigidizer which is about $70/gal, and potentially higher refractory.

The wheels are a turnin'.

Best,
Kelly

 

Yes indeed, I think shapes are possible with blanket and also with bulk fiber. There are compositions with bulk fiber that are blown like a fiberglass chopper gun....not sure it's practical to make one furnace though.

I think it does increase mass and thermal conductivity but only modestly.

My only concern is the temp in the immediate vicinity of the coil. Depending upon the operating temperature of the furnace I think those temperatures could approach the working limits of the wool and binder limits at 2300F. Since I presently only melt aluminum, I only operate the furnace at 1800F.....works fine for me. The coils are only at best a lemon to light yellow at best which suggests <2000F. If that is so, I think rigidized foam will have no problem with that duty except for accidental molten metal contact.

One layer of 2600 wool in that area solves any potential wool issue but potentially not the binder limitations thus my comment about sodium silicate. A Satanite layer or modified insulating Satanite layer between the coil and the wool may add some durability. Continuous duty for Kanthal A1 is 2300F which marginally puts bronze in range.....but that's true for any of my electric furnace versions.

Agree that building inward for the coil shelves is an option but still thinking about coating a plug and building outward because it's easy. It just requires some thought as to how to dry and cure the build up and/or layers.

Thanks for your thoughts.

Best,
Kelly

 

It's a very good point. That sweet spot may be hard to uniformly achieve in a mass of wet wool and experimentation and failure would be expensive at that level.

Heat is a possibility for the lid and base, but a little tougher if lost foam is the method of forming the heating element shelves.

Thanks for the offer. I have some catalyst along with sodium silicate but they've both been on the shelf about a year and half now but are sealed up well. I'll need to check and make sure they're still good and have a manageable working time. I don't remember which version of catalyst I have. If it works well I may be back to you on that propylene carbonate.

The wool really drinks that Colloidal Silica. I haven't estimated how much it would take to do the furnace but I'm guessing 2 gallons. At $70/gal it starts drawing attention but it may not be worth fussing around developing with alternatives...unless they work better. Using the expendable plug method cited above and getting both the Satanite and CS to cure to at least a demoldable green state without heat, and one that can be successfully fired afterward, may be a bit of a trick. I've got some ideas on that though.

Satanite delivery scheduled for yesterday was weather delayed. Maybe today.

Best,
Kelly

 

I figure getting the Satanite to air dry to a green state on the plug is no problem as long as it is applied and well dried in layers of a reasonable thickness. What may be a problem is firing any reasonable thickness of Satanite build up that has only previoulsy been air dried, especially after I stick a bunch of wet, Colloidal Silica laden wool, against it first. That will likely re-suspend and soften the unfired Satanite at the interface. The instruction for Satanite is to fire between coats. It changes state after firing and is no longer water soluble. Not so in air dried green state. My general experience with other materials is thick layers are more prone to shrinkage, curling, flaking, etc during firing because it's hard to apply heat evenly....usually just from one side. The near surface shrinks while the far remains green and that's what causes most of the curing problem. If you cant fire between coats, best bet to combat this is employ slow cure and ramp rate. That I will need to test on a smaller scale.

I think curing in general may be the biggest challenge. After build up and demolding in green state, assuming that can be successfully accomplished, I could simply install the heating elements and fire it. Problem is, the outer diameter of the wool will never see any appreciable heat. I could touch the OD of the furnace shell on my IFB & Wool furnace with my bare hand and it was only mildly warm to the touch when the furnace had been soaked at 2000F. With 3" of wool, I think it will remain very close to ambient. I can get the plug to fit into my larger furnace but not the entire full diameter furnace body....the OD of my small furnace and ID of my large furnace are both 16". I'm not interested in build a furnace to cure my furnace.

The Colloidal Silica can be dried with low heat, time, and perhaps some help to motivate air through the wool. So that's favors the use of CS as rigidizer. With Sodium Silicate, if high heat cannot be applied to the entire body, the SS in the outer regions will never change state and may remain unstable. Not sure how the Ester cured SS will respond in this regard.

Best,
Kelly

 

Thanks for that MS. We're on a similar wavelength.

The bulk density of the Bubble Alumina in that link is about the same as 3000F IFB at about 1gm/cm3 (62lb/ft3). The thermal conductivity of the Bubble Alumina is actually a little greater than 3000F IFB at .57W/MK vs .42W/MK in the 800C-1000C operating range. This comparison is for bulk BA. I expect the cemented BA structure to be both denser and more conductive.

The reason I compare densities, is most of the common and economical refractory materials we use are compromised primarily of Alumina and Silicates. There's not a great difference in specific heat between the two, and it sort of depends on what is creating a lower density mixture for a given material, but if it is air, the thermal conductivity reduces with the density and all the various versions compare very similarly for a given density. The level of refractory can be a different matter.

That density of 62lb/ft3 compares to ceramic wool at 8lb/ft3. So if we are talking low mass furnace, cemented BA is not an exceptional material if considering a monolithic refractory structure made entirely from cemented BA. It would probably be on the order of one of the lightweight insulating castables like HWI Greenlite 45 (about 71lb/ft3), only much higher refractory.

For me, building a monolithic structure from it would probably achieve no better performance than using the lightweight insulating castable. I get no benefit from the high refractory, but you certainly might. My interest in the BA was to use it as a filler to tailor a paintable or maybe dip or flow coat-able hot face. So the build up on my plug would first start with a very thin layer of alumina cement or mortar (or Satanite). The next layers would be the highest percentage BA to cement ratio I could manage and still have a flowable coating that didn't need to be troweled. This would enable the build up of a relatively light, strong, and very high refractory insulating layer. Sort of like building shell on lost wax pattern. That shell could potentially (but not necessarily require) be fired and demolded before application of rigidized wool or built up and cured as a composite structure in the final shape. The BA filler would act like grog/aggregate and significantly reduce shrinking defects and improve drying because you have much less cement and more importantly much less water to eliminate, and this would improve the probability of success when firing a thicker, air-dried green structure.

For Melterskelter, this could also be very useful because it may provide for a hot face with enough structure and insulating barrier to prevent breakdown of the wool (if that's what is happening) and the hot face from delaminating from the wool interface on the lid, while being very high refractory. You could also make the lid any shape you like.....domed for instance.

I did fire off an inquiry to KT regarding price and availability of a 20kg bag of the smaller diameter bubble alumina.

It's hell to be thought in my head

Best,
Kelly

 

Thanks Bonz. Trying to decide how much time I want to spend reinventing the wheel. There are moldable fiber ceramic materials available from about all the typical manufacturers. They have service temperature classes similar to ceramic wools at 2300-2600F. That won't suffice for Meltersklter but probably just fine for my purposes. Their dry/cured densities are typically in the 35-40lb/ft3 range. If I made the entire 3" wall thickness from that material, that would probably have insulating properties and initial time to temp comparable to 2300/2600 IFB. That's ok, but not sure it's motivating enough for me to build another furnace refractory liner. However, something 1/4 that mass would probably be less than 15% the mass of my current dense refractory liner and that is pretty motivating. A compromise might be to use a 1/2" thickness of one of the commercial mastics as a hot face on all interior surfaces and to form the heating element shelves on the furnace body (some of these materials are even rated for metal contact with aluminum), and then use rigidized ceramic wool for the remainder of the wall/structure. I don't think that would require much experimentation. I pretty much just go to it.

Best,
Kelly

 

Well my furnace is a lift off so it has a base, a cylindrical body, and a lid. The body houses the electrical coils. The refractory is actually a module with its on sheet metal skin that sit's on a ledge in the bottom of the lifting barrel like shown below. So when the body is lifted, the heavier hot face is not supported by anything except shear with the remaining refractory wall. If it was just wool I think it may droop. My dense castable liner has flanges on each end so that is what supports the weight refractory module. If the refractory was rigidzed wool in my jelly roll type construction, I think it could easily support it's own weight and the weight of the hot face. The hot face needs to me moldable or machinable (that how I did the IFB version) for the heating element shelves.



I got an answer on the Bubble Alumina cost. $550 + shipping for a 50lb bag........let me think about that.....ok, NO! Here's a couple examples of that moldable ceramic fiber I mentioned earlier. At $80/gallon, that's a possibility for the hot face as opposed to me dinkin around trying to formulate something. I'm not sold on the stuff yet, but guess what? It can't be allowed to freeze. It's going to be -15F ambient here tomorrow. -They can't/won't ship me. RATS!

http://www.theceramicshop.com/product/1197/Inswool-Moldable,-1-Gallon/
http://www.nutec.mx/technical-sheets/nuevo-data-sheets/MaxMoldable-DataSheet.pdf

The wheels are still turnin'.

Best,
Kelly

 

Pretty common method in planar walls, ceilings, and modules. Ever done that in a circular arrangement like that? I looked at that and did a small scale mock up. The ID is 2/3 the OD (10" vs 16"). It either super compresses the ID pleats or leaves the OD loose and it felt pretty floppy on the OD.

Not sure I'd have a good way to do that.

I had wire retainers for the coils in my IFB furnace every 90 degrees. The coils still wanted to grow out of the grooves in some places between the retainers. I found overstretching the coils so they were loaded against the OD and letting them float in the grooves works better. Have had no issues with latest iteration.

Best,
Kelly

 

I've only used the ceramic wool as is, captured between rigid surfaces. I'm having a hard time imagining cutting a channel with a knife in unrigidized wool. Cut, yes. But cut it with any precision without tearing and pulling the wool out of whack, and have it do the job as a heating element shelf....not so much. -Even rigidized wool.

I think it would just spool up on any rotary tool and I wouldn't want to be around or spew a whole bunch of that stuff into the air and onto the lateral surfaces in my shop. Cutting some off the role with a pair of scissors....ok. Same thing goes for IFB. I'll cut and shape it with power equipment but only with a decent respirator and out doors.

I'd be interested in your impression of the Inswool caulk. The stuff I listed is just two examples of 7 or 8 similar products I found. Most were in that 30-35lb/ft3 range cured. That's still 4x the density of wool. Some were called moldable ceramic fiber (with a pumpable version of same), some called it ceramic fiber mastic....but they were all the same thing. If I could paint or trowel it on relatively thin, say 1/4 to 1/2" that would be between 3-6lbs of hot face material and mass for the furnace body.

The thing that made me hesitate was the statement in the material description "its ability to compress" makes it a good material for gaskets and expansion joints. On one hand a semi-rigid or compliant hot face might be good to resist cracks and cyclic dimensional changes, as long as it had enough structure to support the coils. One of the big pluses I was surprised to see was almost all were rated for aluminum contact and most common use were troughs and transport systems. Now I had better not be splashing aluminum on the furnace wall because that will fail a heating element in short order, but that sure is a nice quality for the furnace floor. I just need to get my hands on some of the stuff. I'd rather not have to invent a hot face material.

Here's what I was thinking. Take that plug I posted above and lay it on its side and spin it with this 6rpm gear motor. Raise a bath of paintable hot face under the plug so it rotates through the bath to (very thin) coat the plug. Let it slowly spin to keep a uniform coating thickness until it sets. Dry with warm forced air. Do a second thin very thin coat of hot face. When it quits drip but still wet, replace the bath with a dry catch pan, shower the wet plug with insulator (like bubble alumina except not $600 worth!). Collect the excess from the catch pan and shower again. Repeat until you have coverage.....Dry and repeat until you get the build you want.

I think the net density if you used a thinned mortar like Satinite or alumina cement along with a light weight insulator could be ~70-80lb/ft3. So that hot face would be double the weight/mass of the moldable fiber. Probably a lot stronger and rigid (and brittle) too. But now I have 6-12lbs of mass in the furnace body hot face. Hand lay rigidizied wool between the coil ridges and spriral wrap the rest of the diameter ala the KC jelly roll method with rigidized wool, compress the OD and dry it out.

.....or just use the moldable ceramic fiber stuff for the hot face, finish the remaining diameter build up the same, and call it day.

Follow?

Best,
Kelly