Heat Treat Oven for Iron Build

Discussion in 'Furnaces and their construction' started by Melterskelter, Mar 15, 2021.

  1. Melterskelter

    Melterskelter Gold Banner Member

    I am just beginning the build of a heat treat oven for my cast iron castings. Presently I heat treat my 18” and smaller castings in a front loading pottery kiln I bought years ago for not much money. It works great for both stress relief (1150F) and full annealing (1750F) castings. But my larger castings, the largest being 48” long and 9” high or 30” long by 12” high I have to transport 60 miles to a heat treater. That is a big time eater and some expense, so I decided to build a 2000 watt oven consisting of a galvanized-sheet-metal and angle iron box lined with wool and heated with 16ga Kanthal A1. I am using relatively low wattage as that will make the furnace electrical connection more convenient as I have existing circuits in two locations that will easily handle it.

    52x14x14” box that will be top loading. 95% of the time it will be simply stress relieving parts.

    I will be using a 10-key Bartlett controller ( on order and identical to the one on my front-loader) rather than a cheaper 3-key controller as I will often be switching programs and just don’t have the patience for adjusting ramp up, max temp, hold time, and ramp down with up and down arrow pads. The Bartlett appears to require a 24v AC center tap transformer to power the board. I have various incidentals on order as well including a 30 amp switch, Mica high-temp wire, etc.

    I”ll post occasional updates on progress or lack thereof. Not all the details are worked out. The heating element will be in the lid.

    Denis
     
  2. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Should be a fun project. How many lbs of iron per batch?

    Wise move. The 3 keys are a PITA.
    Not sure about that but probably ok. Where will you measure temp? Any air circulation?

    There will likely be very large gradient top to bottom especially with a large charge of iron castings. It may be inconsequential since anneal and stress relief temp schedule are minimum time at temp. Just want to make sure the top of box temp doesn't runaway while trying to get the bottom to temp and everything at set point equilibrium. Temp at/near top would be safer for control, and then monitor temp at bottom. After monitoring one load you may just be able to set a generous time and call it good. Does the Bartlett have an alarm or secondary input that can control and or display both?

    Just curious, why is the lid more convenient for the heating element? Using fiber board?

    Best,
    Kelly
     
  3. Melterskelter

    Melterskelter Gold Banner Member


    The location of the element is not cast in brick or wool yet. And thinking again about it, maybe would be better lower to take advantage of convection to stir the air better than a lid location would.

    Actively stirring the air is somewhat problematic. What would a person use for a propeller/fan blade? At around 1200F strength of SS drops off pretty dramatically. A tungsten t-shaped stirrer composed of tig torch tungstens might help. Powering it easy enough with an exterior-location motor. Perfectly uniform atmospheric temperature is not needed,, but some approximation would be a plus.

    Denis
     
  4. Melterskelter

    Melterskelter Gold Banner Member

    I think one reason I favored the lid location was that keeps the elements out of harms way when loading heavy bulky castings. But, due care could be taken if they were in lower location.

    Denis
     
    Last edited: Mar 15, 2021
  5. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    An ejector. Preheat the (small) incoming stream of compressed air with small stainless steel coil, just like a naturally aspirated burner. -But it may not be necessary.

    I'm just trying to envision the packing density of the oven. Is the internal volume mostly air or mostly iron? If packing density is low it lessens all concerns.

    What do you have in mind for kiln furniture to set the castings on? Bricks I presume? And the floor....IFB? Elements could potentially be in/on the floor with a partially open brick shelve above them.

    Best,
    Kelly
     
  6. rocco

    rocco Silver

    I don't know much about these sort of matters but I'm thinking a bit of strategically placed thermal mass could be helpful. What I'm thinking is a ifb kiln body, heating coils inset in the floor with hard firebrick over the coils. I would think the hard brick should transfer heat evenly to the interior of the kiln. Am I off base with this?
     
  7. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    No not off base, that was sort of my thought as well, that and taking advantage of free convection, but completely surrounding the coil on all sides isn't generally recommended (at least according to the Kanthal design handbook FWIW) because it cause the localized temperature of the coil to be much higher than the oven temperature........but there are those muffle ovens that have to coiled heating elements encased in castable refractory plates.

    Best,
    Kelly
     
  8. Melterskelter

    Melterskelter Gold Banner Member

    I think convection may be adequate for the iron castings. The iron is so conductive that I have observed machining one face will conduct heat to other parts of the casting quite rapidly so that very large differences don't persist. In addition, precision uniformity of heating is not required. I need to simply ensure heating the part to a minimum of either 1150 or 1750 with slow cool down afterwards. It's not like aluminum where the radiant heat can easily melt parts you are trying to heat treat at ranges not far below the melting point of AL.

    The packing density of the furnace will be very low. It is unlikely that the volume of the furnace would approach 10% iron content. In my present front loader I jsut set the parts in it and set the temps and times. And I have had good results with no hiccups.

    For annealing, the parts will be enclosed in a SS envelope or muffle. So, the enclosure will itself tend to greatly inhibit spot heating. Right now I have a section of 6" pipe to use as a muffle for all but my 36 and 48" straight edges. For them I think I will weld up a SS muffle rectangular in cross-section.

    Right now my plans include IFB "pads" for the castings to rest on. They will be pretty widely spaced in an array otherwise surrounded by wool. The coils will need to be supported either in the lid or floor by standoffs that I plan to make from BluRam. I have done some test pcs of the BluRam that are very robust.

    Denis
     
  9. Melterskelter

    Melterskelter Gold Banner Member

    I've got all the pieces of the furnace collected except for the resistance wire which comes Monday. I have the angle iron shell welded and 20 ga steel panels pop riveted to the skeleton. And I have started filling in the wool insulation. One question I had was how I would retain the wool on the vertical walls and in the lid. What I have decided to try on the sides is to use SS tig 1/16th cut filler wire about 4" from the top of the side laid on the wool and "laced" to the sheet metal sides using short pieces of left-over 20 ga Kanthal wire that loop through pairs of two closely spaced holes in the skin and around the filler rod. Using a twist wire plier the loop is snugged up. A loop about every 8 inches looks like it will be enough to hold the wool in place. Inside HT Oven2.JPG inside HT Oven.JPG
    Here is the twisted wire holding the horizontal filler wire Wire Loop2.JPG
    Here the end is bent over and tucked into the wool so that it is not a poker. WIre Loop.JPG

    THe floor will have 3 strategically placed IFB located one each on the cross pieces and held in place by surrounding wool. The top will have 2 or 3 rows of filler wire/kanthal loops holding it against gravity.

    THe highest temp I anticipate in the furnace will be the occasional 1750 heat with most heats being 1150F.

    Denis
     
    Tobho Mott, Al2O3 and Mark's castings like this.
  10. Melterskelter

    Melterskelter Gold Banner Member

    Not everyone may know about safety wire twist pliers. Here’s kind of a fun video where Adam Savage demos their use and gives his reasons, similar to mine, for owning one.:D They are cool in a geeky kinda way.



    Denis.
     
  11. Melterskelter

    Melterskelter Gold Banner Member

    Made a bit more progress (and thawing myself out as it is 45 degrees in my garage this morning. Heat Treat.JPG Heat Treat2.JPG

    Next come the combination BluRam/10-24 THreaded rod/Kanthal Standoffs for the heater coils and the thermocouple and control units. I also need to buy some more thermal blanket as I ran short. I knew it would be close.

    The way I secured the blanket to the side walls seems really secure. I decided to lace it every four inches. If anyone should try this method, one thing that can save time making the kanthal (or stainless) loops to be twisted in lacing the filler rod would be to take about twenty feet of wire off the spool and put one end securely in a vise and using a vise-grip yank on the length of wire a few times. That straightens out the coil curvature and makes insertion easier. Cutting lengths first and straightening each length individually works but takes a lot longer.

    Denis
     
    Clay likes this.
  12. Melterskelter

    Melterskelter Gold Banner Member

    Today my shipment of 2600F wool came. So, I was able to line the lid with 2" of wool. For this I used a grid of 1/16" tig filler rod suspended from the lid sheathing with twisted stainless wire much as I did with a single wire on the sides. I had drilled enough holes to have the grid wires on 4" centers. But, it was evident that having them on 8" centers should be adequate. If sagging becomes an issue, I can always wire in more filler rods.
    This assembly feels very secure.
    Heat Treat Lid.JPG
    Where the filler wires cross I straddled them with the suspension wire and twisted it, snipped off the excess and bent the tail into the wool.
    Heat Treat Lid2.JPG

    The control box for the Bartlett control board still has lots of wires hanging out, but should come together in a couple of hours tomorrow morning. It will include a electromechanical relay, pilot light, DPST heavy duty switch, hinged transparent cover for the Bartlett, and conduit protected high-temp leads to the heater coil.

    [​IMG]

    Heat Treat Control.JPG

    I still need to make stand-offs for the coils as mounting the coil will be the final step.

    Denis
     
    DavidF likes this.
  13. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Looking good MS. Have a link for that Bartlett Controller? What are you using for the main power switching? SSR?

    Best,
    Kelly
     
  14. Melterskelter

    Melterskelter Gold Banner Member

    I was leery of using a solid state relay. I bought a mechanical, 12v AC relay from Mouser. There seem to be dozens of 12v DC relays available. But, I had to look hard for the AC version. The Bartlett board wants 12v AC power input and outputs 12v AC to the relay. The relay I got is as linked:
    https://www.mouser.com/ProductDetai...H8XNQOwy55SkkVbtPuFqVLk0SaophkNJFSe/YfCxAIA==

    It is made by Schneider part no. 996-8501CO6V36 and appears to be very robust in construction.

    Bartlett controller kiln controller is sold by several vendors. I used The Ceramic Shop:
    https://www.theceramicshop.com/prod...Lshhz6tIEaqikjMwjn20ivtYRfUjnQP4aAoV8EALw_wcB

    They shipped promptly at reasonable cost.

    I currently have a small to medium-sized front-loading kiln that is controlled by a Bartlett controller and I really appreciate its convenience given its 10-key input and memory for various saved programs. The documentation provided with the board is excellent.


    Denis
     
  15. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    In theory SSR would be more reliable for high switching frequently. In practice the reliability is probably offset by unknown quality of imports. As long as the hysteresis around the set point is reasonable span or adjustable, the cycling about set point should be modest and a mechanical just fine.

    Best,
    Kelly
     
  16. Melterskelter

    Melterskelter Gold Banner Member

    It is the failure mode of the SSR that was the concern. My understanding (based on hearsay) is the SSR's fail in a closed condition. I was concerned about the possibility of uncontrolled heating. I know that the Bartlett on my kiln switches every ten or 15 seconds once operating temperature occurs. So rapid switching is not needed.

    Denis
     
  17. crazybillybob

    crazybillybob Silver Banner Member

    The issue with the mechanical are noisy (electrically) and they like to weld contacts shut (same fail state as SSR).
    Moral of the story.... Keep an eye on them as everything fails, and wants to kill you!
     
  18. Melterskelter

    Melterskelter Gold Banner Member

    Hmmm, the failure mode you suggest for EMR's is at variance with almost all that I read (I have read there are rare instances of contact welding as the failure for EMR's) but actually failure is almost always a coil failure. This is what manufacturers of them say. And this is what a good friend with considerable experience with them reports. In fact, he suggested not using an SSR in this application.

    Here is one reference addressing the issue of selection criteria of SSR vs EMR and failure modes and the applicable excerpt from that site.

    https://www.controldesign.com/articles/2014/in-with-ssr-out-with-emr/


    "Fail Safely

    Electromechanical relays do have a very clear safe-failure mode. When they fail, they almost always fail deenergized, so they are more "adequate" for safety applications than regular solid-state output devices. Those devices could fail dangerously (in short circuit) or safely (open output). So if safety is a concern, either keep the relays or replace them with a good certified safety PLC (way more expensive)."

    On another note: control box size. If anyone in the future intends to enclose one of these Bartlett or comparable-sized control boards along with a relay, switch, and transformer plus the needed heavy mains supply lines and terminal block, I suggest using an 8X8X4" box rather than a 6X6X4 like I started out using. When it all got down to stuffing the components and wire into the 6X6, I felt it was just too crowded in there. So, I had to start over with the larger box---not an especially jubilant moment. But I am glad I did. Putting everything in the larger box is soooo much easier.

    AN 8X8X4 box with knockouts cost 30 bucks. Well worth it.

    Denis
     
  19. Fulmen

    Fulmen Silver

    If you're pushing the surface loading on the coils a thyristor or TRIAC based regulator is beneficial. IIRC they recommend reducing the surface loading by 20% when using on-off.
     
  20. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    True. Might be something to think about on your melting furnace MS. Temps are much more modest here in the heat treat version.

    I have a mechanical contactor in addition to the SSRs in my controller. I use the SSRs to switch/temp control but use the mechanical contactor to shut down the heating element(s) power whenever I access the interior of the operating kiln/furnace. The primary reason wasn't fail safe against SSR failure. Because I have two heating elements in the furnace that are wired in parallel, and the SSR only opens the upstream leg of each 240vac (US) single phase element, that still leaves the downstream side 120v to ground.

    I've seen a number of DIY PiD/SSR kiln/furnace controllers on-the-cheap that only use one SSR with a mechanical switch added to the SSR control circuit for on/off. But in US 240vac, that still leaves the coil partially energized.

    To retain the ability independently energize each heating element I would have needed four SSRs. The mechanical contactor for primary power (the instrument circuit remains energized) seemed to be better option because the (over temp) alarm circuit can be wired in series in the contactor coil control circuit to open it upon alarm.......though you could do the same thing on the SSRs control circuit.

    Best,
    Kelly
     

Share This Page