Large Low Mass 22kw Resistive Electric Furnace

Discussion in 'Furnaces and their construction' started by Al2O3, Jul 5, 2024.

  1. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    After four years using my low mass resistive electric furnace,

    http://forums.thehomefoundry.org/index.php?threads/muses-about-a-low-mass-electric-furnace.607/

    I can confidently say I’m quite happy with it. So much so, I’m similarly going to convert my large furnace. It will be a protracted build that I progress as time permits so I thought I’d start this thread to track the project.

    This furnace was originally designed to be fuel (natural gas) fired, but I’ve only used it as resistive electric, with my 8kw electric insert module. I only intended to use that electric insert for heat treating but have instead used it for aluminum melting with my IC 60 Crucible. This furnace body can be fitted to my small furnace lift, but I grew tired of swapping the furnace bodies and built a separate lift for the larger furnace as seen here:

    http://forums.thehomefoundry.org/index.php?threads/furnace-cart-and-lift-build.278/

    Even though this larger furnace is fairly low mass for what it is, it still has nearly 200lbs of refractory that must be initially heated and with only 8kw, plus 33lbs of crucible and up to 60lbs of aluminum, that takes >3 1/2hrs to reach pouring temperature.

    So, I decided to build a new resistive electric module/insert for my large furnace hardware, increase the power almost 3-fold to about 22kw, and reduce the furnace mass by perhaps as much as 2/3. I’m hoping to get 40lb melt times down to an hour or less.

    1 Project Scope.jpg

    I arrived at ~22kw by applying the criteria from the Kanthal handbook for maximum wall loading for a furnace interior that could accommodate my IC60 Crucible and my existing furnace shell and lifting hardware. That ends up being 15” bore by 19” height (the existing refractory insert is 14”D x 19”h, and must fit within a 22” OD) and that is essentially the maximum allowable wall loading at 3.4kw/ft2 @1000C. It was also in the range of convenient power level for using 90amps.

    Unfortunately, I only have residential power of 240vac single phase, but I do have a dedicated 200amp service in my shop. 90 amps at that voltage turns out to be about all the power I could practically throw at this size furnace anyway.

    So besides the refractory insert and heating elements, I’ll also need to build a high power electrical controller, and the mother of all drop cords since my furnaces are portable and get wheeled outside for use. I have been bargain shopping the surplus market for materials and electrical components for months and have now collected everything. At this power level, almost all the cost is in the high-power electrical components. More on that later.
     
    G54AUST likes this.
  2. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Low Mass Refractory Inserts

    One score that I’m pretty excited about is this new (to me) refractory material and acquiring it is what prompted me to proceed with the build. The refractory is really the key to the build. It needs to be the best compromise of density, low thermal conductivity, and strength. This seemed to fit the bill, and at $30/5 gal bucket, I got it for almost nothing from surplus. What a find! I was worried about shelf life so I bought one and tested it in a number of ways. All was good so I bought three more. Shipping would have been 3x the purchase price. It was 5 hours away and I have a friend whose business makes a weekly run between locations so I called in a favor. For that price I should have bought the other 2 they had, but 4 was already 2x what I needed.

    When cured it’s pretty much moldable Insulating Fire Brick. It has the density and insulating properties between 2300/2600 IFB but has a max continuous operating temp of 2500F. Theoretically, along with the upper operating limit (2400-2500F) of Kanthal A1, this gives the furnace bronze capability, though I doubt I’ll put that to the test. Although it is denser than Moldable Ceramic Fiber (MCF) at about 32-36 lb/ft3 dried, it is much stronger.…..much more like 2600 IFB, and being moldable, I can limit the amount/mass of it I need to use and insulate the rest with ceramic fiber blanket. There were several other moldable/pumpable Morgan refractory candidates but they would have cost me at least >10x what I paid.

    Like my smaller low mass furnace I made a plug mold from some cardboard concrete tube with foam profiles glued on for heating elements shelves, and a fixture to apply and dry the refractory.

    2.2 Large Furnace Plug Mold.JPG

    It took me half a day to make the fixture. The material is gunnable, but I don’t have a gun and it’s a small job for that so I just troweled it on with a 3” putty knife. I doubt this would fall within Morgan’s recommended use but I applied the refractory in ½” thick layers and between each successive layer air dried with the aid of a 300w incandescent light bulb (yes they still exist) inside the plug, initially controlled to 120F, and also shrouded the plug to moved air across the outside with a muffin fan.

    4.1 Inital Drying Rig.JPG

    I did have some cracking and shrinkage in places on the first coat but I just opened the cracks with a knife and repacked them. I let each ½”t coat of the cylinder and Flanges dry under heat and forced air for a few days, added the ribs, and eventually it looked like this:

    5.1 Molded Furnace Body.jpg

    Then I upped the temp to 140F for a few days. That’s supposed to be the max exposure temp for XPS so I was just trying to get the mold as firm as possible before demolding more higher temp curing and burning out the foam.

    I’ve had a piece 20” x 40” x 4” thick 2600F ceramic fiber board (CFB) for 5 years. It was another eBay bargain I got for almost nothing that couldn’t pass up. I was going to use it for a small high temp electric furnace but it will be perfect for new furnace lid and base. I glued pieces together and fashioned it to fit my existing lift hardware. Unlike the gradual build up of the furnace body, it only took 30 minutes to cut this ceramic fiber board on my band saw and a razor knife on the step in the lid.

    6 MCF Base and Lid.JPG

    I placed the molded center section between the new CFB base and lid and used the same 300watt light bulb to cure it for a few more days at 100C/212F.

    7 Secondary Cure.JPG

    I removed the dense refractory inserts:

    9 Remove Dense Refractory Inserts.JPG

    Applied ceramic wool to the exterior of the molded body, leaving the foam coil shelves in place to be burned out, and swapped it along with CFB base and lid into the furnace lift for final curing/firing. Using the 8kw electric insert, over the course of a day, gradually increased the temperature 1000C/1832F and let it cool overnight.

    The next day I had a look and observed several significant cracks. Although it’s sort of to be expected, I could have hoped for better.

    10.1 Cracked Refractory.JPG 10.2 Cracked Refractory.JPG

    There is a huge temperature gradient from ID (1000C) to OD (100C), and that means considerable difference in expansion, thus stress and cracking. I could take a knife, carve out the crack, pack it with more of the material, and refire it, but I suspect the crack would just recur there or in another place upon subsequent heat cycling. I most likely will do nothing and just use it as is.

    -We shall see.

    10-13-2024 Update

    I added some refractory tunnels for the heating element electrical connections. The plan is to triple up the heating element leads and route them to a ventilated raceway outside the furnace similar to what is done with electric kilns. This is a bit inconvenient because it means the refractory module cannot be installed into or removed from the furnace lift with the heating elements installed, but it does ensure the lowest temperature possible for connecting the heating elements to electrical conductors.

    11 Add Element Tunnels.JPG

    Added a couple inches of ceramic fiber/wool and a protective sheet metal skin.

    11.1 Add Ceramic Fiber Wool.JPG 12.1 Skinned Ref Insert.JPG

    After I paint the sheet metal skin, it's ready to be installed in the furnace lift and have the heating elements installed.

    I made a quicky CNC'd XPS polystyrene mold and cast a low mass/low profile plinth from dense refractory.

    13 Casting Plinth.JPG 14 Demolded Plinth.JPG

    Best,
    Kelly
     
  3. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Resistive Electric Heating Elements

    10-27-2024 Update

    The furnace was designed to be 21.6kw using six 3.6kw resistive heating elements. I used 14ga Kanthal A1 resistance wire which is about .21 ohms/foot. For 240vac that required 75 feet (~16 ohms) for each coil. The furnace bore is 15" but the centerline of the coil shelves is about 16.5" diameter. A one inch diameter coil allowed for about a 3:1 stretch from the solid wound length.

    So I made a 1" diameter coil winder by drilling a hole in a block of wood and using a piece of 7/8"D tube for a winding mandrel powered by a cordless drill.

    IMG_2986.JPG

    Then tripled up the leads for lower resistance and heat at the junction with the conductor.

    IMG_2984.JPG

    Stretched them about 3:1 and installed in the coil grooves.

    IMG_2991.JPG

    Trimmed the leads and connected them to the respective ceramic terminal blocks in the external raceway.

    IMG_2988.JPG IMG_2989.JPG

    I still need to glue in wedge blocks to retain the coils in the grooves. I'm planning 8 spaced evenly on each shelf and just glue the front edge so they can be cut and removed for coil replacement.

    IMG_2987.JPG

    11-10-2024 Update

    Got the elements wired up. There are two banks of 3 elements powered by two separate SSRs. I used MG wire from ceramic junctions with the "tripled" resistive heating element extensions to the power input terminals.

    IMG_2992.JPG

    I have a fan that installs at the bottom of the raceway.

    IMG_2993.JPG

    Here it is all buttoned up.

    IMG_2994.JPG

    ......and with the retaining blocks glued in place. There was 8 per shelf for a total of 48 of them!

    IMG_2996.JPG
     
    Last edited: Nov 10, 2024
    G54AUST, BattyZ and 509Maker like this.
  4. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Build 22kw Electrical Controller

    10-13-2024 Update

    I got started on the electrical controller, did the layout work and sheet metal fab for the control box to house the rather large contactor and breaker. The control box is 18ga steel and 16"x 16"x 10". Next I need to cut the openings for power, controls, and instrumentation.

    15 Control Box.JPG 15.1 Breaker Contactor.JPG

    That's a 12" scale on the box of plugs and receptacles......they are rather large!

    15.2 Plugs Receptacles.JPG

    10-27-2024 Update

    I cut the openings in the control box for mounting the various components and test fit everything.

    IMG_2978.JPG IMG_2979.JPG

    Applied some paint to the control box and furnace module covers

    IMG_2981.JPG

    .....and you knew I had to get a few lost foam castings into the project. Below are the patterns for the control box receptacle, wall mounted receptacle, and Solid State Relay heat sink which mounts on top of the control box.

    IMG_2982.JPG

    11-10-2024 Update

    Had a casting session today and converted the foam patterns for the receptacle boxes and heat sink to metal.

    IMG_3014.JPG

    The heat sink and one receptacle box mount on the controller and the second receptacle box on the wall near my electrical service panel.

    IMG_3018.JPG IMG_3019.JPG

    All the electrical components are mounted, now I just have to wire everything up.

    IMG_3016.JPG
    IMG_3017.JPG

    All wired up.....

    15.64 Wired Up CB.JPG

    15.65Labled Control Box.JPG
    15.66 Power Panel Labled.JPG
     
    Last edited: Nov 30, 2024
    G54AUST, 509Maker and BattyZ like this.
  5. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Install Supporting Electrical Service and Initial Start Up

    11-19-2024 Update

    With the controller completed, I needed to add a 100 amp circuit to my main electrical service panel. I only had one position left and as opposed to adding another sub panel, I just combined two light duty circuits onto a tandem breaker and moved another breaker one position to make room for the new 100 amp breaker/circuit, and mounted the 100 amp receptacle adjacent to the main panel.

    100 amp Service.jpg

    I needed a means to connect the furnace to the power receptacle and to do so, made what I call the Mother of All Drop Cords. It's 35ft of 2 awg-4 conductor SOOW cord with a male and female plug on each end. It weighs about 50lbs!

    MOAD.jpg

    Plugged it into the receptacle........

    IMG_3052.JPG

    ......and the other end into the controller......

    IMG_3051.JPG

    Then fired it up first with one bank of three resistive elements and then the other three. Both drew about 45 amps so I applied full power and the power meter displayed as shown below.

    Power Screen Capture.jpg

    The furnace was empty but wow, the temperature was climbing at >10F/second. I'm sure it will be different with a crucible and 35lb+ charge of aluminum, but initial impressions are very encouraging!

    IMG_3050.JPG

    11-24-2024 Update

    I used the furnace for its first casting session today. It melted 40lbs of aluminum and brought it to pour temperature of 1350F in a little over 50 minutes. The power meter said it used 20kw, so the melt cost $1.90 in electricity. I'm happy with that. I'll have a video in the not-too-distant future.

    I have been using small diameter Inconel sheathed K-Type thermocouples, but I bought one of the heavy gauge kiln style K-types for this furnace. It resides in the center of the lid.

    IMG_3060.JPG

    Because of the high-power density, I suspected the area between the crucible and the furnace wall might be quite a bit hotter than that measured right above the cold 40lb charge of aluminum, so I mounted the previous Inconel sheathed TC right above the annular gap between the furnace wall and crucible so I could compare.

    IMG_3065 Furnace 2 TC Lid.jpg

    I had the PiD set at 1800F on the TC in the center of the lid. The TC near the wall was about 300F higher than the one centered in the lid until about 1000F, then it narrowed to about 200F, which meant the annular region was reading ~2000F when the 1800F was sensed above the crucible. The two were continuing to converge but pour temp was reached before they could. This probably means the heating elements are a couple hundred degrees F hotter or ~2200F. It's probably OK for Kanthal A1 but I may back the set point down to 1700F in the future. I doubt it will make much difference in melt time.

    The PiD I bought for this build is WiFi enabled. At first I thought the remote monitoring was just a novel nicety, but it turns out, by using the Ap, you can key in any parameter and don't have to wait minutes manually stepping on the keypad, so I can make any parameter change no matter how large in just a couple keystrokes. Below are two screen shots. The first five minutes of the melt had rolled off the graph. The little temp twitches are me removing the inspection plug to view the charge, and the large dips are me adding some charge, skimming and eventually pouring.

    IMG_3068.PNG

    Here's my rig just after the pour....

    IMG_3071.JPG

    .......and the fruits of the effort......another intake manifold.

    IMG_3076.JPG

    11-30-2024 Update

    The video documenting the build and initial in-service use is below. There is a five-minute walk-around description of my pouring rig near the end of the video.

     
    Last edited: Nov 30, 2024
    FishbonzWV, G54AUST, 509Maker and 2 others like this.
  6. BattyZ

    BattyZ Silver Banner Member

    This is quite convenient for me because I need to step up my furnace size too! My lift-off retrofit fit kiln works well but it's starting to show its age and I am at the point where a 150# crucible in a tilting furnace would serve me best.

    Essentially this furnace(not willing to spend $10k on it!!):

    https://makeyourowngoldbars.com/col...e-smelting-ore-silver-tilting-pour-myogbtl150

    So far the build looks great. I will be following along, taking notes.
     
  7. Already it looks like a work of art, 90 Amps is some serious current, a lot of switch gear here goes to 63 Amps after that there's a big jump in price. An auction sale induction furnace would be ideal for 22KW of power.
     
  8. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    I must say, I do understand why kiln manufacturers use brick (IFB). Besides being practical to build, all the joints in brick allow some movement without damage. The stainless shell binds it all together. For conversion to lift off, some may require some reinforcement to the supporting structure. Did yours?

    Anyway, for practicality and economy, you just may not be able to do better than surfing CL and FBMP to repurpose an old kiln. They're like exercise equipment......People list them for real money but they almost never sell. You can often pick them up for nothing or near nothing.....and you don't even really care if they work, because I'd rip out the old controller, and probably replace the elements with higher power ones. You'd prefer that the brick be in good shape but even that is easily repaired.

    Getting all the refractory, and build time savings for something (mostly) already built into a furnace and lid for 0-$200? Pretty tough to beat that. If I was starting from scratch, didn't already have my lift hardware, and wasn't so anal (well I actually enjoy it too because I learn!) about building my own equipment, it'd be pretty compelling.

    Not really a fan of tippers unless you are going to use it with a ladle.....then they can be very practical, and even easier to repurpose a kiln for such.......but buying that crucible $$$ will make your eyes water.

    Man, you got that right. I scoured the surplus sites for the better part of year to collect the components. Got some real bargains. But, wait until you see the Mother Of All Drop Cords.......some bargains on the 100amp+ plugs and receptacles but not the 30ft of 4 conductor 2Ga stranded 2/4 SOOW cord.....$13/ft.

    Believe me, I gave it a hard look. There are many issues. Even if someone gave me a real induction furnace, I don't think I could put it into service in my shop environment.

    Best,
    Kelly
     
    Mark's castings likes this.
  9. BattyZ

    BattyZ Silver Banner Member

    Thanks for the advice! Maybe I will have a relook at doing a brick build. I will stop threadjacking now lol
     
  10. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Not a jack. This moldable material is new to me. I'm sure I'll get it to work. The issue will be just how long lived it will be.

    In my smaller furnace, I've had versions in IFB, dense castable, and Moldable Ceramic Fiber (MCF).

    As far as I'm concerned, the MCF I used on my small low mass furnace is proven material for an aluminum duty furnace. The stuff on this build is rigid. The MCF retains some modest flexibility and as a result will not crack. It is lower density, better insulator, but not as strong. It has a little lower operating temperature (2000F-2300F depending upon manufacturer and what you believe) but it would have been >$1000 worth of (new) material versus $60. But, if in the long run I have trouble with this material, MCF or something very similar would be a go-to. The MCF and like materials are commonly used in industrial "Tube Furnaces".

    Best,
    Kelly
     
  11. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    I updated posts #1 and #4. The updates are at the bottom of each post starting under "10-13-2024 Update"

    Best,
    Kelly
     
    Tops likes this.
  12. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Had some time to work on the furnace today. I updated posts #3 and #4. The updates are at the bottom of each post starting under "10-27-2024 Update"

    Best,
    Kelly
     
    Tops likes this.
  13. Tops

    Tops Silver

    It's all looking great Kelly, looking forward to more updates. Your sheet metal work is outstanding, have you ever done a tutorial on that?
     
  14. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Don't look too closely :). Don't think I'm qualified to be doing tutorials, but I do have a 36" foot shear, box/pan brake, slip roll, various hand tools which help quite a bit. Would like to build an English wheel, bead roller, and recip machine someday. In my original furnace build on AA forum, I had some detail on the storage cabinet fabrication but that blew up with Photobucket debacle.

    Best,
    Kelly
     
    509Maker and Tops like this.
  15. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    I've been chipping away and made some more progress on the furnace build. I updated posts #3 and #4. The updates are at the bottom of each post starting under "11-10-2024 Update".

    Best,
    Kelly
     
    Tobho Mott, 509Maker and Tops like this.
  16. r4z0r7o3

    r4z0r7o3 Silver Banner Member

    Are you intending for that little PC fan to cool off ALL the connections? If so, I would recommend only having the holes near the top. Or have there be progressively more and more vent holes toward the top. Otherwise 95% of all airflow will just dump right out near the bottom.

    blowerair.png

    (sorry, I dunno where the broken IMG link comes from, there's no other link in the editor, must be a bug)
    [​IMG]
     
    Last edited: Nov 11, 2024
  17. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Actually, since I have free convection on my side, it's doubtful whether I'll install the fan at all, but in any case, don't think 95% will exit at the bottom and don't think that would happen even if the raceway was open with no cover at all. The open area on the perforated plate is >>10x the fan discharge area. Why would you think the fan discharge takes a 90 degree turn immediately after exiting the fan?

    My smaller resistive furnace has no ventilation at all at the resistive/conductive wire junction. That connection is actually made just inside the skin on that furnace and has performed fine for four years now on the low mass version so I figured the combination of the longer tripled leads and exposed connections can only be goodness.

    The fan is a cheapo that can take any voltage input up to 220vac so all I need to do to power it is connect it to the lower coil with a couple of short leads of MG wire. Then it automatically will turn on/off with coil power........but if not needed, it's just something else that can fail and potentially cause more problems than it's worth when/if it does.

    The area you circled isn't a broken image. I used a piece of scrap perforated plate I had laying around for the cover and that location is the seem where I welded a narrow strip of metal on the backside to join the two pieces, as I wanted it to install/remove as a single piece with fewer mounting screws.

    I'm actually thinking the better place for a fan may be on the control box, since it can be exposed to the Sun, is relatively small, and I have a lot of power going through the switch gear. Another candidate is blowing on the heat sink, but since it has so much more area/additional capacity than needed, I suspect free convection will be just fine there too. The one on my smaller furnace controller is similar and I can barely even detect a temperature rise at 8kw full power.........so, although I have provisioned to easily add them, I won't do so unless needed.

    Best,
    Kelly
     
  18. r4z0r7o3

    r4z0r7o3 Silver Banner Member

    Sorry, I show a broken
    Code:
    [IMG]
    , that's what I was talking about.

    Those PC fans mostly push air toward the outside of the ring in a cone-shape, i.e. toward the nearest perforations. Then also, forced air will take the path of least resistance, so maybe not 90% loss but you'll definitely get only a little air over the top connection. But if it's not a concern, then there's no worry and I agree, I'd ditch the fan as something else to go wrong. Esp. if you think the convection will be enough to keep things cool.

    Cooling into/out of your box could make it a dust trap. That can sometimes (eventually) lead to electrical arcing. Since your boxes are metal, they may already have adequate heat-sinking as-is. You could always add a sun umbrella :D
     
  19. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    I updated post 5 and made a short test run. It's a beast!

    Best,
    Kelly
     
  20. ESC

    ESC Silver Banner Member

    That looks great Kelly.
    I have had the coils change length in my furnace and pull out of their groove. Did your blocks hold the coil in when it cooled?
     

Share This Page