Forge and Burner Design Page #2
(This page is a continuation of page #1 of the Forge and Burner Design Page)

©Golden Age Forge

Return to Page #1

The Full Site Map - Lists All Pages on This Site

Contact me by phone: (208) 462-4028
Note: Due to spam problems I no longer post my e-mail address.


Materials and Hardware Sources
Refractories, Burner Flares, & T-Rex Burners

Sources For Refractory Materials

A Word to Potential Advertisers: I do not carry banners, flashing icons, or other eye catching low class graphics to advertise products on my pages. All commercial links posted here are on my page for the convenience of the persons using my page, not for economic gain of the business or me. I receive no payment for any links or comments made here. My comments are honest evaluations of the source to make decision making easier for my page users. If you wish to add your link to my "Sources" section, it must be in keeping with the purpose of this section, and if accepted, it will not have any logos, banners, or icons, other than those already present in this section. If you would like to discuss a link please phone me at (208) 462-4028. Warning, I receive many requests, but accept very few.

I get many e-mails asking where to obtain the Kaowool, ITC-100, propane regulators, and other materials necessary in forge construction. Here are some sources for these materials, and others, that you may use to obtain needed parts and materials. Jay Hayes and Rex Price have refractory materials available at a very good prices, and much lower than available in most places. Jay has a web page which is under construction that you may wish to visit too, and is worth your time to do so, as he has a lovely place. Rex Price also produces and sells the finest and hottest ventouri burners made in the world at this time. You may want to visit Rex's web page and look at his burner selection instead of building your own. These are hotter then anything you can build on your own. Thanks.

Here is another location that you may use, but you will have to phone this company at your own expense. This is "The Potter's Center" here in Boise, where I purchase my supplies when I need them immediately. They have Kaowool board, high alumina kiln shelves in various dimensions shapes and thicknesses, ITC-100, kiln shelf posts, and many other useful materials and items, including pyrometers. They will be glad to mail an order to you, but they do not have a catalog. You will have to work out your order over the phone. Their prices are high for some items and very reasonable for others. Phone (208) 378-1112, or Fax (208) 378-8881. Scott Brown, the owner, also has the following 800 number on his card, but I don't know that it is still in service, try 1-800-498-1126.

I have to thank Matt Bomba for this link. Clayartcenter.com is a one stop pottery supply center, and has a good selection of all the needed components, refractories, and other items that you might need to build your melting furnace, forge, kiln, or heat treating oven. Their prices seem to be in line with the going rates elsewhere. Their online catalog is a little difficult to find, but if you go to the lower left corner of the page there is a box with links to various items. If you click on any of the items listed, you will be taken to the catalog and can easily explore from there. Check it out.

Here is one more useful link. If you are involved with foundry work, this is a must have link. It is also of value for blacksmiths, or anyone who uses high temperature products. They carry refractory products, ITC-100 in its various forms, and a lot more. Budget Casting Supply is well worth a look. You can e-mail Paul if you have a question about a product too.

I have no other sources than those listed above, so please don't e-mail me with a request for additional sources. I suggest you check locally in your area if the above sources are not convenient for you.

A Note About ITC-100 and Kaowool: ITC-100 is more than just a ceramic wool stabilizer. It is an infrared reflector that reradiates up to 98% of the heat that strikes it back into the forge chamber. This will heat your iron faster, and will act like an additional inch of Kaowool, providing lower shell temperature and better fuel economy, as much as 30% better in independent tests.  It will also stabilize the Kaowool so that you don't have dangerous ceramic fibers in the forge exhaust. It will last almost forever if it is not damaged through mechanical injury by you or others using your forge. Even though it is semi expensive, it is well worth the cost as it will repay you in various ways, directly and indirectly, very quickly. It may be used on any kind of forge chamber surface, not just Kaowool. Also, if you have metal surfaces in the forge, there are special undercoatings available to bond the ITC-100 to the smooth metallic surface. I should add that this material is not useful in a coal forge.

Kaowool comes in two temperature ranges. One is about 2300 degrees, and the other 2700 degrees. Although your forge should run up near 2700 degrees, or hotter, you may use the lower temperature Kaowool if its fully coated with ITC-100. The ITC-100 should reflect the heat enough to prevent problems due to the lower temperature range of the material. I would suggest you use the higher range Kaowool however, unless you are unable to obtain it from your supplier. Jay Hayes has both available, so specify which kind you want when ordering. The higher temperature Kaowool is more expensive as would be expected. You should use ITC-100 on the Kaowool whichever kind you use.


Do You Need a Premade Burner Flare?

The following four factors are, in my opinion, what a flared nozzle does to contribute to burner performance, and why you should use one of your own making, or one of the stainless flares offered here.

   1) The flared nozzle acts like the "impedance matching device" found on radio transmission antennas. It creates a "pressure transition" to balance the internal pressure of the burner tube to the lower pressure of the forge chamber atmosphere.

   2) It slows the gas stream to a velocity below the burn velocity...its flame holding and stabilizing function.

   3) It creates a low pressure in the nozzle step that, in effect, invites more gas down the burner tube, allowing a slightly greater output than would occur without it. This factor is not proven or provable with our crude testing equipment, but I can clearly see its contribution in my mind.

  4) The last function is to initiate the burn earlier so as to prevent the injection of cold gasses and free oxygen into the chamber of the forge or furnace....the result is a hotter chamber and less, or no, scaling.

If you would like to buy premade burner flares that have a perfect 1:12 taper, and made of stainless steel no less, there is now a source. Most of the problems that I deal with when helping guys with burner problems relate to the rate of flare of the burner flares that they put on their burners. If the flare is not made correctly the burner will not function correctly. I was astounded when Larry Zoeller sent me one of his "press formed" stainless steel flares to try. I removed one of the old forged flares from the number one burner on my four burner forge and installed Larry's stainless flare. My old mild steel flare came off in fragments, as it had totally corroded through in only two years of use. The best I had been able to do for "idle" pressure with my old forged flare was 2 psi. Below that the flame would start to pulsate or huff. When I put Larry's flare on the burner it easily held the flame steady all the way down to a little below 1 psi! His flare just doubled the "idle" economy of my forge! I have ordered 7 more from Larry, they are that good. I can certainly make my own flares, but to obtain the precision that Larry has achieved I would have to set up a hydraulic press as he has done.

If you would like to obtain some of these superb flares, contact Larry at the address shown below, or at his e-mail address. I am sure you will be very satisfied with the results of your purchase. You may initially think that his price of $6.50, plus shipping, for a burner flare is steep, but remember that he not only makes them from stainless steel, but they are drilled and tapped, and come with two stainless set screws installed also. They are ready to just tighten into place on your burner. BTW, I get no payment of any kind for this "add." I just think that these flares meet a big need out there in the metalworking field and should be made available to the forge and foundry community. Please be aware that these flares can melt and deform if used on the Micromongo Burner when its run at high gas pressures. In the case cited in the linked document, Fred was running them at 25 psi. 

Larry Zoeller
4312 Lahnna Dr.
Louisville, KY 40216
USA

(502) 361-0706

zman59@earthlink.net

Larry's Home Page

Visit Larry's Burner Flare Page. You can see what they look like, and Larry has a price sheet with details there as well.


Premade Burners Are Now Available

The "T-Rex" Family of Burners

There is now a superb new premade burner series available for those who do not want to build their own, or for those who want the ultimate in both quality and BTU output. The T-Rex is a turned, milled, and tuned hybrid designed jet ejector burner which has to be experienced to be believed. For a complete description see the "T-Rex Burner" page.

Another burner that you may want to consider, and part of the developing T-Rex family of burners, is the "Shorty Burner." This is a miniturized T-Rex, and has applications in places where the T-Rex may not fit, or where this reduced sized burner would be more convenient. I have a Shorty Burner Page available if you would like more information.

You may go directly to Rex's own Burner page if you wish. He has a troubleshooting page, and a price sheet there, as well as a short description of the burners he presently is producing, which now include several much larger burners than I have listed here. I will not keep his full listing of burners updated on this page now that he has this information available for you on his page. The descriptions for the T-Rex and Shorty burner I have here are more complete than what you will find on Rex's site, so you may want to read these first, and then link to his site. I have links to his site at the end of each of the above linked pages for your convenience.


BTU Output Calculator

I am very pleased to be able to offer you a very convenient BTU output calculator. It will allow you to calculate as closely as possible how much heat energy you are delivering out the nozzle of your burner, given the gas pressure you are using, and orifice, jet, diameter of the burner. This very fine program was created by Thomas Vincent. It is in Exel format, so you will have to have Microsoft Excel on your computer to be able to use this program. I have it available here in several forms. You may download the zipped file, which includes the "Orifice Document", the "Calculator Instruction Document", and the "BTU Calculator" program. You can then unzip it and use it as needed. Or, you can use it directly by simply clicking on the "BTU Output Calculator" file link. Again, to use this calculator your system will have to have Microsoft Office installed on it in order to be able to open these Word and Excel documents. If you download the files to use on your own system, be sure you read the instruction file first in order to prevent damage to the Calculator file through improper number input. The Calculator program requires entry of your burner "orifice" diameter in decimal form. I have the conversions from drill # to decimal diameter available in my FAQ. You are free to distribute this calculator file so long as you maintain the source information, my site, with the file and leave the authors name on all documants as they are provided to you here. Tom did a lot of work to provide you with this valuable tool, free of charge, so he deserves to be given credit for his work. Thanks.

BTW, I am able to open these files with Office 97, so if you have Office 97, or above, you should be able to use this program without any problem. Due to the size of the Orifice document, it is advised that you download the zipped file instead of using these files directly off my site. 

BTU Output Calculator

Orifice Coefficient Document

BTU Calculator Instructions Document

Download All Three Documents in Zipped Format


Will Gas Forges Weld?....an E-mail

I have run into two areas of ignorance that I have continually to fight, even on "theforge". One concerns the ability of a gas forge to forge weld. This is certainly not a problem within the members of theforge, but the second is. I was told by a smith on theforge that "everyone knows that gas forges can weld, and have for years", and that certainly is not true. I have posted an e-mail I received that addresses both of these points. I am posting this to show that there is still a lot of ignorance out there that needs to be addressed. We, as brothers in this craft, all need to tend to these issues when the opportunity is presented. Most certainly the gentleman who sent me the e-mail is not one who needs educating. He is anything but ignorant in these matters. I am posting his e-mail for your interest. He was kind enough to give permission for me to post this communication. Thank you Adam.
 


How Hot Can These Burners Get?

I am very pleased to be able to include information about Dave Moore's ultra hot propane fired crucible melting furnace. Dave has been very generous with his time and provided me with detailed information about his melting furnace. This is such a remarkable achievement that I felt it deserved its own page. Please click the image to learn more about this exceptional piece of work. (Note: Dave is normally very safety minded. Please do not ever pour hot metal without wearing proper personal safety equipment. As one person commented, one wandering spider in the mold and you will be wearing the liquid metal, and probably lose your eyes as well. Also, never pour metal while working over concrete. Spilled metal can cause the concrete to explode and blast you with molten metal, or burn down your shop.)


Click  image to visit page.


Burner Chokes

I am adding this short paragraph at the top of this section because I consider it of great importance. In our development of the Mongo series of burners we discovered the full potential of a choke. It will enable you to run the "Reil" burner, or any of the Mongo series, or Side-arm burners, down to levels of gas pressure measured in only a few ounces. The key is to have a choke that can totally close the intake bell, or holes. That way you can balance the air intake to gas injection no matter how small the amount of gas entering the burner. I can now easily run my Micromongo down until the gas pressure gage needle is resting on the peg, and the burner still has a stable flame, no huffing or other problems. I estimate I am running at 1-2 ounces of gas pressure at this point! So to achieve the very useful low end range, that all of these burners can reach, use a well designed and constructed choke. (Remember, for a choke to be of any value, your burner must be capable of running in an oxidizing state, oxygen rich, when running at full pressure.)

I was not interested in burner chokes when I first built my big gas forge because I operated it at higher pressures than I presently find necessary. As I reduced my operating pressures down to the 1-1/2 to 5 psi range I found that more scale formation was occurring on my work. The answer to this problem is a burner choke. It allows you to tune your burner to a neutral or reducing flame at any given operating pressure. 

The following four choke designs are scanned from a small forge building publication that has no author or publication information whatsoever. The only clue to the author's identity is on the front cover which is simply "JF", with a "98" under the letters. I could not contact the author to obtain permission, but was informed by the person who gave me the booklet that it was free for duplication and distribution, so I can only assume that this use of its information would fall within those boundaries. If anyone knows the author I would very much appreciate information as to how to contact him to obtain his permission formally. Thank you.

To use these choke designs it will be necessary to build your burner using the "Bordeaux Modification" so that you will have use of the threads in the throat of the burner bell, or the top surface of the bell. I would also strongly suggest that you use a 1-1/2" by 3/4" reducer bell instead of the 1-1/4" by 3/4" reducer when using a nipple section. This is due to the reduction in open intake area that installation of a nipple section will cause. I would also suggest removing any fins or flange that may be found on the interior lip of the nipple section used. It is important to maintain maximum gas flow through the system so that you can still operate at higher pressures efficiently, therefore choke design #4 may not be a good choice due to the considerable intake restriction it imposes on the system.

Choke Design #1

Choke Design #2

Choke Design #3

Choke Design #4

The Best (?) Burner Choke Alternative

An alternative that is worth considering is to lock your burner jet tube down by screwing a short length, say 1-1/4", of nipple into the mouth of the burner bell far enough to contact and lock the jet tube. You will have to use a bell diameter no smaller than 1-1/2" to do this however to maintain enough  intake opening area. (You may even need to use a 1-3/4" bell if you find that the back pressure of your forge design causes your burner to run too rich.) I recently built a burner this way for a forge I sold and it worked extremely well. This makes the construction of the burner much easier because there is no drilling and tapping, and it also sets up the burner with a short nipple in the bell to which you can add an axial choke, see below, quite easily.

One suggestion would be to attach a small diameter cross bar across the throat of the nipple, and to it weld or braze a short length of all-thread which would stick out of the center of the opening of the burner an inch or two. You could then weld or braze a nut to a disk of steel that would screw down the all-thread and act as a finely tunable choke. The disk would not have to be as big in diameter as the nipple, as it only needs to restrict the air flow, not stop it completely. By drilling a hole through both sides of the nipple just below the lip it would be an easy matter to insert a cross bar that closely fits the holes, and to which you had already brazed on the all-thread to form a "T". You could then align the all-thread to the center line of the nipple and braze each end of the cross bar into place. It would be rock solid and aligned almost perfectly.

When I built the burner I cut the section of nipple off at a point leaving about 1/4" to 3/8" of unthreaded section of pipe above the ends of the threads. This is where the cross bar would go. You would do well to try to align the cross bar with the jet tube below it too for better air flow characteristics. If you drill the two holes in the nipple first thing they can be used to place a rod through the nipple allowing you to screw the nipple in very tightly without having to use a pipe wrench which will damage the surface of the nipple. You may want to drill two sets of holes, one set below the other, so that you will have these "wrench" holes available all the time. I elected to use one set of hole for both purposes.

Please have a look at the image of my axial choke. It is currently mounted on my number one burner which has been modified to the Bordeaux mounting system. I cut two pieces of 1/4" all-thread and brazed them together in the form of a "T". The top of the "T" can be inserted in the two holes in the top of the nipple and locked in place with a nut on one side and wing-nut on the other.  I need to be able to remove the choke when I cover the forge up at night or the axial rod will poke a hole through my tarp. Just use two nuts if you don't have to cover the forge.  I used a large diameter 1/4" washer for the choke disk, and brazed a quarter inch nut in its center to allow it to be screwed down on the vertical stem of the choke's axial rod. It is easily adjusted while the burner is running because it doesn't get hot. It isn't necessary to have a disk as large as the opening in the bell. You only need to reduce the air flow, not stop it. Notice that I am using the holes in the nipple both to tighten the nipple on the jet pipe and to mount the choke. I love it when things serve for more than one function.   :-)  Here is another image taken after I converted my #2 burner over to the axial choke design also.

I now have made some rather important discoveries about chokes that I explained at the top of the Choke section. If you build an axial choke, as shown in my images, I have two suggestions for improvements you may want to follow. First, increase the size of the choke disk so it can fully close the bell. The second suggestion is not of great importance, but is a matter of convenience. Instead of using a nut to braze to the disk, use a wing-nut. That way when you screw the disk down very close to the bell to run your burners at very low levels, it will be easier to move it up again by use of the wings on the nut.



Additional Burner Designs

Note: These burners are designed by others, so if you decide to build one, please contact the designer if you encounter problems. I do not support or guarantee in any way that these burners will function as stated below. I have NO experience with any of them. My comments are based entirely on what the designer has related to me and I have put into my own words. Thank you.


"Monster Burner" and Melting Furnace

(This burner easily melts cast iron!)

Designed by Rupert Wenig, and submitted by Robert Grauman.

This is a corker of a burner that has been perfected by Rupert Wenig. It is an enlarged version of my 'Reil Burner," but Rupert deserves all the credit for this one since I never dreamed that the burner design I posted could ever melt cast iron if scaled up. These guys up in Alberta enlarged the burner and then used it in a melting furnace built to the Gingery design. It easily melted 17 pounds of cast iron! The following images show the burner, the furnace, and the results of their cast iron melt. The burner also was used to melt aluminum, 2 kg in 10 minutes, and brass, 6 kg in 19 minutes, ... very impressive times. The cast iron took a little longer. The top link is the burner design drawing of the "Monster Burner" by Rupert Wenig. I should add that the "Tee" pipe coming off the side of the burner tube in the images is just a support handle to mount it in the furnace. If you are considering this burner, be sure to look at the Minimongo Burner too before making a final decision as to what burner to use.

Foundry Melt Logs

Burner Design Schematic

Image of three burners

Gingery Furnace

Hot Interior of Furnace

Pattern and Finished Cast Iron Casting


******************

Bienstock Burner

Designed by Marshall Bienstock, photo by Bruce Freeman


Click Image for Enlargement

The "Bienstock Burner" design above is a more sophisticated version of the basic propane burner that allows greater control of the forge atmosphere. It is more complicated to build, and also adds to the burner length considerably, which will preclude its use in compact situations. If you have the room, and want to spend the extra effort, this is an excellent burner.


*****************

Jay's Burner

Designed by Jay Hayes, photo by Ralph Kessler


Click Image for Parts Breakdown



Hot News...ITC-100 Needs Recoating Periodically

I partially rebuilt the chamber in my four burner forge today, Sunday, 3 Nov 02, and had a very surprising result. The front half of the chamber was showing all the use and abuse it had received over the last 5 years, so today I decided to replace the front end firebricks, and recoat the front half of the forge chamber, and the new firebricks, with ITC-100. I recoated only the front half due to its constant use. When I finished I allowed the ITC-100 to dry. I then lit the forge and allowed it to run at idle, only a couple ounces of gas pressure, to cure the ITC-100. I was surprised when I looked over after only a few minutes and saw the mouth of the forge chamber glowing a bright orange, and at only idle pressure. When I went over to look, there was a very definite line where the new ITC-100 ended and the old was exposed. I brought the pressure up to working pressure, 3 psi, and allowed it to continue coming up to heat for about 5 minutes. At that point the entire front half of the chamber was bright yellow-white, while the old ITC-100 on the exposed back half, that was in front of the movable back wall, was still dead black, no heat showing in the floor tile at all! There was also a very definite line on the coated Kaowool chamber walls, where the old and new ITC-100 met.

This was a big shock to me. I didn't know that ITC-100 had such a definite life span. At this writing I do not know what the useful life span of ITC-100 is, but I hope to find out soon. Also, I do not know what signs indicate that the ITC coating has lost its usefulness. The coating in my forge had become very dark, even black on the floor of the forge. I only want to alert the metalworking community that ITC-100 needs to be recoated occasionally. How often still remains to be determined. I have been having some problems with my forge welding, but chalked it up to sloppy technique. Now I believe that I was not getting the temperature that I was previously when forge welding was extremely easy. I will post more info as soon as I learn more. The moral...recoat occasionally when your ITC becomes greatly discolored. You may easily apply a little test dab of ITC-100 in your forge and see if that spot becomes significantly hotter than the rest of the surface. If it does, you probably need to recoat.

Here is the straight skinny, right from the source. This comes from Jock Dempsey, and is self explainatory. I am quoting his e-mail with his permission. His help is much appreciated.

Ron,

I have just spoken with Mr. Feriz Delkic, the owner of ITC, about your questions. He says that at forge temperatures there is absolutely no breakdown of the ITC-100. They have furnaces with oxidizing atmospheres in service for some 20 years without degradation of the coating.

In all probability the problem is soot, scale dust, and other debris that has attached to the surface. Mr. Delkic says with a light coating of debris the ITC-100 is still acting as an isolator and reflecting the infrared, but that the debris is filtering it. In a small forge or furnace this might effect the recovery time somewhat.

Predicting when this problem will occur would be impossible due to the many variables of individual forge usage. I would expect it has as much to do with how many times the forge is fired, and how long it is run each time.

Commercial usage forges may run non-stop for weeks, or at the least for a full 8 hours. In your shop and mine we may run the forge for an hour or two, shut it down, and run it for another hour later in the day. When I am casting with my little melting furnace, I run it for half an hour, shut down when I pour, and then fire back up for the next melt, maybe 5 or 10 times in a day. I would guess that if soot is the problem, then the number of cold restarts is a factor.

Forges are also exposed to scale and scale dust. Fine scale dust blows around in the forge, and the surface temperatures are high enough for it to melt and stick to those surfaces. As you know the ferric oxide can damage some refractories, such as the alumina in Kaowool. It does not damage the ITC-100, but it does cling to the surface. Again, this is not a failure of the ITC-100, but an operating condition particular to forges. Kilns and other types of furnaces each have their own pecularities.

I would say that recoating the forge with a thin wash at least once a year would be good practice. I have found that once a surface is coated that it takes very little to recoat it by brush application. Spraying may take more, as the consistancy must be just right to stick and not run.

Jock


Some Gas Forge Designs
My Four Burner Forge

 The four burner forge I built, using the burners I discussed above, has turned out to be an excellent forge in every respect. It was built using a custom rolled cylinder of 11 gage steel, lined with two 1" layers of "Kaowool." It is coated with "ITC-100", manufactured by "International Technical Ceramics Corporation". The cylinder is 24" long by 12" in diameter. I will not provide any diagrams here as the images and narrative are enough to allow anyone with basic metal working skills to duplicate it, or a modified version.

I used 2" of "Kaowool" liner, but perhaps 3" would be a good additional insurance if you are at higher elevations where forge performance might be marginal in achieving welding temperatures. I highly recommend "ITC-100" over the use of "Satinite" It is a superior IR reflector, reflecting back up to 98% of the heat that strikes it. This contributes to lower forge shell temperatures, higher interior temperatures, as well as fuel savings.

The floor of my forge is rather special in design. I wanted to preserve as much insulation as possible so designed the floor to sit above an inch of kaowool at the extreme edges, and almost two inches in the bottom center. I used one half of a 12"x 24"x 1" high alumina kiln shelf. The 6" wide kiln shelf floor sits on six 1" high kiln shelf posts that are kept in place by 3/4"x 3/8" diameter rods blind welded into holes in the bottom of the shell. This prevents the weight of the floor plate from compressing the kaowool, and provides the necessary space to allow one full wrap of kaowool around the circumference of the forge shell. The second layer runs from one edge of the floor plate around to the other. I also added another strip of kaowool under the plate to fill up the gap in the center. Once all was in place, everything, including the floor, was coated with "ITC-100". When I eventually rework the interior of my forge I will make one modification. To protect the fragile Kaowool walls I will install two full length, by 2" wide, sections of kiln shelf on either side of the floor to act as bumper strips. They will be angled outward and blend in with the surface and curvature of the Kaowool liner. By doing this the forge will be much more resistant to mechanical damage in its interior.

I wanted to be able to easily change the opening to the forge to suit the work at hand. Although my "brick wall" may not be as pretty as some other arrangements, it has great versatility in allowing any configuration needed to be instantly arranged. I constructed a "brick shelf" that is part of the forge for the bricks to sit on. All bricks are coated with ITC-100 also. The bricks are not standard fire bricks, but are ultra light weight high temperature kiln bricks. They run from $3.50-$4.00 US per brick, but their high thermal efficiency justifies the expense. Odd as it may seem, the less expensive bricks, $3.50, are actually better than the higher density and higher priced bricks because of their better insulation qualities.

If you are new to forge work, and especially to ceramic fiber insulation, you need to be aware that welding flux, such as borax, will eat through them as fast as water through cotton candy. That is one reason for the heavy kiln shelf floor plate. I also recommend you consider a piece of stainless plate, or sacrifical piece of kiln shelf, placed under your fluxed iron to protect the forge floor, since it will attack that as well.

The burner mounting on my forge was done using threaded pipe sections that screw into pipe couplings that I sawed in half and welded to the shell. The pipe sections were big enough to allow free passage of the burner nozzle through into the forge chamber. The nozzle is held in position by two rings of three "set screws", in this case 1/4" diameter bolts. This allows me considerable adjustability on my burner positioning and aim. I don't believe in the commonly held belief that the burners should be mounted at a tangent to create a vortex in the forge. My experience indicates that it is not necessary, and may in-fact contribute to scaling. I have mine aimed to the side of the floor plate, but not on the wall of the chamber. I can adjust them to dead center if desired, or to a considerable angle to produce a vortex.

The penetration of the burner nozzle goes only as far as the interior surface of the outside layer of kaowool. The interior layer of kaowool has a hole poked through it with a conical punch to continue the path to the interior of the forge. This allows the burner nozzles to be out of the extreme heat of the forge preventing any damage to them. The kaowool seals quite effectively around the burner nozzle preventing any blow-by from coming up between the burner tube and the mounting tube holding the burner.

There are two improvements I have incorporated into this forge that are not normally found on gas forges. I use a movable back wall inside the forge to reduce the forge chamber dimensions to the minimum necessary for a particular application. If the work is small, and I only need one burner, the wall can be slid all the way up to create a very small, but very efficient chamber. The movable back wall is made of a piece of "kaowool board" cut out to allow clearance around it for gas escape. It is also coated with "ITC-100" to further increase its efficiency as an IR reflector. This cuts down on heating unused space and increases fuel efficiency. Note: Please visit my Forge Page for an important update to this movable back wall design. Look under "My Newest Forge Design", at the bottom under "Update".

The second big improvement I incorporated into my forge concerns the plumbing for the four burners. I used individual ball valves for each burner. This allows me to use only the number of burners necessary for the work at hand. The second modification is the use of two valves on the gas inlet line. They are configured in a parallel arrangement with one of them a normal ball valve (on/off), and the other a needle valve to allow precise control of a small gas flow to the burner. With this arrangement I can adjust the input gas pressure to whatever working pressure I need using the regulator, and then with the "on/off? valve to the "off" position, adjust the needle valve for a minimum burn level, an idle flow rate.

With this arrangement I can instantly shut the forge down to a minimum burn rate while I am at the anvil, and when I return to the forge can instantly return to my full working pressure with just a flip of the "on/off" valve. Since I sometimes spend extended periods at the anvil this results in considerable gas savings. To shut off the forge I use the individual burner valve(s). It took me a day to get used to flipping the valve on and off each time I went to the anvil, but quickly became an automatic response for me. I highly recommend this modification. (For more complete information on the idle/full valve arrangement, including images, consult the Four Burner Forge Design description on my Forge and Foundry Page.)

I used all brass valves and plumbing on my forge, but iron will do as well, and at much lower cost. I also have a quick disconnect on the forge so that I can easily disconnect the propane tank. This is a very useful addition.

I will conclude with a few comments on the operation of my forge. Any burners that are being used will remain very cool to the touch while in use. They will heat up once the forge is shut down however, so watch what you touch. The burners that are not being used provide an open chimney to the forge, and the ultra hot gasses will pass right out through them heating them to dangerous temperatures. To prevent this I simply stuff a small wad of paper towel into the throat of each unused burner. When I wish to add a burner, even when the forge is running, I simply use a pair of needle nose pliers to pull the plug out, and then switch on the gas valve. The paper will not burn while in the burner tube.

Some smiths like the thermal mass of a poured in place refractory lining in their forges. There are both advantages and disadvantages in such linings. In my case, I like to have my forge cool down as quickly as possible when I shut down since I use it outside and need to cover it. Poured in place linings take a very long time to cool. The poured in place lining, let's call it "rigid lining", is far more forgiving of the abuse that careless handling of the iron can cause. On the other hand it is far less thermally efficient as an insulator. This may mean the difference in achieving, or not achieving, welding temperatures at higher elevations, and in some case lower ones as well. If you will want to move your forge, the ceramic wool liner will certainly be superior for your needs.

If you are not familiar with the products I have discussed here, and their sources, try calling your local pottery supply store. They cary kilns and supplies such as "Kaowool" and "ITC-100", as well as "Satinite". If they don't have them on hand they can order them for you. When you are there look around at the various types of refractory products they cary. You will be able to get your kiln shelf floor plate and support posts there as well. You may have to look around however to find a place that will sell small quantities in some cases. If you are unable to locate ITC-100, it is made by "International Technical Ceramics" which you should be able to locate on the internet, and contact to find a local source.

 
See Also the Construction of My Freon Tank Mini-forge


Some Additional Gas Forge Designs

Forge #1

Submitted by: Matthew Wills, design by Matt and others

Image 1 , Image 2


Forge #2

Submitted by: Richard Sevigny Richard's tiny forge was inspired by Ed Halligan's

design as posted by Donnie Fulwood!

Image


Forge #3

Submitted by: Harry Foster A very cute tiny forge.

Image


Forge #4

Submitted by: Larry Zoeller

This is a micro forge that uses a propane torch for heat. Cute forge!

Construction Plans & Images


Forge #5

"Sandia Recuperative Forge Design"

Note on an alternative to the Sandia Forge, the Clamshell Forge design: The Sandia Forge is a very sophisticated forge design requiring an extraordinary amount of labor to build. It is of particular value for wide flat forgings like scrolls. We are building some interesting forges that have a large flat floor, 18" x 24" or larger, and a flat top lid that is attached to a vertical lift system. The lid lift system can be as simple as a camper-jack, or as complex as a camper jack combined with a foot treadle for raising, and a dash-pot for lowering the lid. You can crank up the lid and stack a wall of fire bricks around the edge of the floor in whatever arrangement and height you require, then lower the lid to close it in. You can also lay a refractory rope on top of the bricks to get a tight seal. The burners can be top or bottom mounted, and the top can also have an additional treadle lift for instant opening. This allows total versatility to set up the forge for whatever the task at hand requires. This offers far greater versatility than the Sandia forge, and the work is less than that required to build the Sandia forge. Using high quality burners like the Rex series, an idle/full valve arrangement, and ITC-100 interior coating, the Clam-shell will almost certainly match, and probably exceed the Sandia in both operating temperature and fuel efficiency. I say "probably," because there are so many possible variations of this forge, and different insulating options, that you would have to compare specific forges for a definite answer. I have a Zip file of information about the Clam-shell that you may download if interested. Ralph Sproul provided me with some of the images in the Zip file, as well as a copy of his very well done Clam-shell forge plans. I highly recommend you contact Ralph if you are interested in this forge design. Nahum Hersom provided me with some of the information too. Also, talk with me by phone because there are some very useful additions and modifications I can recommend to you. The Clamshell is as close to being a "do-all" forge design as exists.

Submitted by: Chuck Kishaba This very fine forge is a recuperative propane forge built from the "Sandia National Labratory" plans which are available from ABANA. The images show various views of the finished forge, along with two images of it heating some 1" diameter bars. This is such a fine piece of work that I am going to quote Chuck's entire e-mail explaining the following images. His excellent description follows. Much thanks Chuck.

Chuck' Description;

Well, here they are. Let me give you some info on each JPEG photo.
1) forge5.jpg is a 3/4 front view of the forge
2) forge6.jpg is a 3/4 rear view of the forge
3) forge3.jpg is a front shot with the door open
4) forge11.jpg is a close up shot of the heated bars
5) forge8.jpg is a shot of a cold forge with the cold samples prior to light up
6) forge9.jpg is a shot of the bars being heated
7) forge10.jpg is a shot of the bars after 8 minutes from a cold start

My story:
I needed an economical means to flatten the ends of 1" diameter solid round
bars. I looked into induction heaters (incredibly expensive) and commercial
propane forges. I'm not a blacksmith and needed something that I could light
up, do my work, and shut down. I liked the efficient design of the Sandia
forge and decided to buy the plans from ABANA. The forge was not easy to make.
It helps to have sheet metal skills and a lot of patience to fit the ceramic
insulation board. It took me about 3 months working in my spare time.

Description:
Air enters 2 flattened stainless steel preheater tubes from the rear of the
forge. This air then mixes with the propane, and the air/fuel mixture goes into
a ventouri that necks down to the burner tubes. The exhaust gases exit the rear
vertical chimney which pass around the 2 flattened stainless steel preheater
tubes. This feature preheats the intake air to 1000F. The preheated air
increases the velocity of the mixture and this results in a hotter forge,
and a 25% savings in fuel.  The white insulation covering the burner tubes is
1" kaowool wrapped in stainless steel wire screen. The main insulation is 1"
thick 2600 ceramic insulation board. I obtained it from Thermal Ceramics.
I paid $26.00 per square foot for the board.

Changes:
The biggest change I made from the ABANA plans was the door. The original
plans call for a door without the oval opening, only a notch in one corner. My
door does lose some efficiency, but as you can see from the photos, works quite
well for my purpose. The stand is also different from the ABANA plans but has
no effect on performance.

Performance:
Photo (forge8.jpg) is a shot of a cold forge with five 1" solid round 1018 bars
ready to be heated.
Photo (forge9.jpg) is a shot of the forge 30 seconds after startup with the five
bars inserted into the door.
Photo (forge10.jpg) shows the five bars after 8 minutes. This is all from a cold
start with the bars inserted immediately after startup!

I've ran the forge for about 1-1/2 hours from a 30 lb propane tank with no
freeze-up. I usually run about 5-7 psi of pressure.

Conclusion:
I am extremely pleased with the performance of this forge. It works better
than I imagined and was worth the expense of the ceramic board and difficulty
of construction. I would like to thank Robb Gunter, Karl Schuler, and Ronald
Ward of Sandia National Labs for their excellent design.

ABANA is on the web at: abana.org. To order the plans, go the the site
and click on "ordering" On the ordering page you will find a listing for
"recuperative forge plans".

Good Luck!

Chuck Kishaba
Los Angeles CA

tmcgroup@aol.com

Forge #6

Building a Large Gas Forge

By: Steve Smith

This is a copy of a newsletter article that Steve Smith wrote, and has very generously allowed me to post to my site. It describes the construction of his large gas forge in detail. If you are considering building a forge you may want to read this document to get another description and viewpoint of building a large gas forge. Click the title to visit this page. Thank you.


Links To Forge & Burner Designs, or Images

Don Fogg, Bladesmith-Gas Forge and Burner

David W. Wilson's Forge Design Page (Two designs)

"Tales of a New Blacksmith" by: Marc Godbout...must reading for the new smith.

 


Coal Forge Designs

There is now a counterpart to my gas forge site that is dedicated to coal forges, Forge and Fire. Matthew Rutz has done a superb job of posting a large body of coal forge information to his site. Be sure to explore all of his site, it is extensive and rich in content.

I have already covered the construction of my big coal forge in some detail on my "Forge and Foundry Page", so I will refer you to that location. Forges for coal can be very simple, no more than an old barbecue fixed up with an air supply coming in from the bottom with a clay lining in the pan, to the very complex, and difficult to build, forge you will see on my page. I will post pictures of coal forges from other smiths as I come across them and feel they are particularly useful or clever in some way.

BRAKE DRUM FORGE: You can quite easily build a very serviceable forge from a big cast iron truck brake drum. The hole is already provided in the bottom to bolt in your air grate, a piece of 1/4" steel plate with holes drilled in it. You only need to attach a plumbing "Tee" on the bottom to provide an inlet for your air, and also to be your ash dump. Put some legs on it, and a blower of some kind, and your in business. These forges do not even need to be lined with anything, and most of them have a fairly deep bowl giving you a good deep fire. Their only down side is that you will have a difficult time getting long stock into them for heats in the center area of the stock. A friend of mine, who is a very good smith, uses one of these forges and is able to make remarkable things with it. He cut out two slots, one on either side, and installed doors to allow long items to be heated. He also has a chimney installed to prevent breathing too much of the smoke.....a great idea.  I hope to have a picture of his forge here soon. Robert Bordeaux has provided a unique design in which he uses the brake drum as the bottom fire-pot, and expands upward into a blower housing, creating a much larger pan. The forge is lined with a 1/2" thick layer of furnace cement. Robert deserves lots of thanks also for his contribution of the "Bordeaux modification" to the burner design I have posted above. It is a superb modification.

NOTE: Coal smoke will do bad things to your lungs and body over time, so a chimney is very important on any coal forge, indoors or outdoors.

FIRE-POT DESIGN: Bill Franchini has provided the following home built clinker breaker fire-pot design. It is straight forward, except that the drawing shows the use of a square section tuyere, instead of a 4" diameter pipe. He built it this way simply because he didn't have 4" pipe available. It should be easier to construct using the pipe.If I were building this fire-pot I would make the fire grate out of two separate pieces. One part would be a grill fabricated out of the materials shown in the drawing, with "fingers" for the grill elements, and the portion that is actually attached to the rotating handle would be another set of "fingers" that would pass between and through the grate fingers as it was rotated, thus reducing the loss of coal Bill mentions in his instructions document.

Fire-pot Drawing

Instructions Text
 


Additional Coal Forge Designs and Misc

Brake Drum Forge at Anvilfire.com

Paul Stevens' Brake Drum Forge


Blacksmithing Terminology

Thanks to the efforts of Paul Stevens there is now a glossary of blacksmithing terms available for new folks to consult if needed. Check it out.  :-) 


 The Personal Side  

Me in my Utilikilt - superior wear for blacksmithing -or- how to get toasty knees. :-)

Me, the Z, and the Kilt

My Younger Daughter and Me at the Hagerman National Fish Hatchery
 


The Full Site Map - Lists All Pages on This Site

Return to Page #1

23 Nov 07

©Golden Age Forge