timgunn1962

I'm about 9 miles from Haslingden, maybe 15 from Bacup. That flame looks like a pretty reasonable starting point to me. There looks to be a decent central cone, suggesting a fair bit of primary air. I've never been able to get a decent photo of a flame because the exposure length blurs out the edges of the flame parts. I'm pretty sure it's just my complete lack of photographic skills, but I find that video tends to show more of what is going on than photos. For HT temperatures down around the 800 degC/ 1500 degF region, I tend to find I need a very rich flame with a good bit of yellow to it. What do you get when you stick that in the forge, deprived of the secondary air? I built a 2BF a few years ago and made the burner hole tapered through the thickness of the wall: narrower at the chamber end. This gave me some secondary air adjustment by moving the torch in and out. It seemed like the obvious thing to do and I didn't really think much about it at the time. It did let me get pretty good temperature control over a few minutes of soak time. It might be worth a try.

Plumbing-type torches can be something of a pain. They are usually quite carefully designed to do a specific job and that job is not heating a forge. Generally, torches are designed to work in the open air. There is usually some primary air, but the bulk of the combustion air is usually secondary air. Primary air is air that is mixed with the gas before it reaches the burner. Secondary air mixes with the gas after the burner. When we put the burner into the forge, it is usually through a hole that is fairly tight to the burner. At best, this dramatically reduces the amount of secondary air available to burn. In most cases it completely eliminates the secondary air supply. When we look at flame temperature, we see that the temperature achieved varies with the air:fuel ratio. It may not be immediately obvious why this is, so I'll try to explain. Combustion is a chemical reaction between the fuel (in this case we'll assume it is Propane) and the Oxygen in the air. Like all chemical reactions, there is a fixed ratio at which the fuel and air like to react. If we mix the air and fuel at this ratio, all of the fuel reacts with all of the Oxygen and we will get the highest flame temperature. We get the highest flame temperature because we have released all of the available heat (the fuel is completely burnt) into the smallest possible amount of gases. If we move away from this ratio, we will see the temperature drop. If we add more air, we only have the energy released by burning the amount of fuel we have burned, but by adding the extra air we have increased the amount of gases that we are heating. With more"stuff" and the same heat input, the temperature will be lower. If we add more gas, we get a similar effect: Oxygen availability is the limiting factor in this case. We only have the amount of heat energy released by burning the Oxygen we have, but the extra fuel gas absorbs some of that heat, reducing the temperature. Once the partially-burnt fuel gas leaves the forge, it mixes with air and starts to burn again, giving the secondary flame usually referred to as the dragons breath. The heat energy released in the dragons breath is outside the forge so does not provide us with useful heat. The technical challenge you are likely to need to tackle is getting the correct amount of air into the flame to achieve the temperature you need. Using a Naturally Aspirated (unblown) burner, the way that the air is mixed with the gas usually involves a gas jet, Bernoulli's principle and something that may or may not bear any resemblance to a Venturi. Almost any Naturally Aspirated forge burner seems to be described as a Venturi burner on the internet, even though very few of them actually incorporate a Venturi. It would almost certainly pay you to research the basic principles on the web. To be honest, I would not expect many people to have done what you seem to be trying to do. I'm guessing the adapter you bought to let you run a small-canister torch off a large cylinder probably cost at least as much as a regulator. I spent a happy weekend a few years ago trying to modify a cheap and fairly nasty torch set to work as a forge burner for a coffee-can type forge. The first thing I did was fit the gas jet from the smallest burner into the biggest one to try to increase the air:fuel ratio. It helped, but was not enough. I then opened out the air holes in the back of the burner. still not enough. Then I slotted the side of the burner, which gave me enough. http://s667.photobucket.com/user/timmgunn1962/library/Toys/MiniForge?sort=3&page=1 The temperatures in the photos are in degreesC. I rather enjoyed playing with it and it performed remarkably well for something so crude. I took it to a hammerin where it did a couple of days of continuous work. My intention was to try to write it up as a "how-to" that a beginner could follow. However, there were far too many variables for that. For less than twice the cost of the torch set (and considerably less in terms of time), I can buy an Amal atmospheric injector, a regulator and the parts to build a near-ideal forge burner, easily able to give stable temperatures everywhere from Heat-Treating to welding. More importantly for me, I can tell someone hundreds of miles away "buy these parts and screw them together", confident that they they will work as intended. I've not returned to the modified torch setup and the original went in the raffle at the hammerin.

It's still not really clear what you are after, though you've certainly narrowed it down. How are you intending to heat? If you are using a furnace, whether gas or electric, the usual way of getting the steel to a given temperature is to set the furnace temperature and monitor it, then wait until the steel is the same colour as the forge interior. If you are doing this, you can perhaps be a bit more accurate if you put the thermocouple in front of the steel and watch it until it disappears against the steel. I am fairly sure Type R,S or B should get you a temperature reading and I think the Recrystallized Alumina sheath should do it. As you say, generally speaking, and for a given sheath material, the thinner the sheath, the faster the response. If you are taking the steel out of the furnace, heating with an induction coil or doing anything else strange, get a good description of your process down on paper before you talk to the experts. I think there will be several more variables to consider, possibly including emissivity, once you are outside the furnace. I had an IR thermometer that would measure to 1600 degC a while ago. It worked and I was quite impressed by it. Ultimately, it was not much good for my purposes because I was dealing with unknown and variable emissivity as the surface scale changed. I could not just shoot the temperature while the workpiece was in the forge (and emissivity therefore at 1.00) due to interference from the gas flame. I passed it onto a smith who finds it useful: I think he uses it to check he is within a temperature range when making Damascus, so he doesn't need absolute precision. I don't have details. http://www.digital-meters.com/temperature-c3/infrared-thermometers-c17/cem-high-temperature-infrared-thermometer-50-to-1600-c-cem-dt-8859-p250 Edit: Pure Iron melts at 1538 degC and steels melt at lower temperatures. Things are likely to be dribbly before you get to 1600 degC.

I've not used B. I have used both R and S. The application was/is direct fired thermal oxidizers running on landfill gas and the process temperature was 1200 degc (2192 degF) . We've since reduced the process temperature to 1000 degC (1832 degF) and tend to fit Type N now. In essence, it works like a big gas forge. The gas flow is set, then air supply is adjusted to get the temperature required. It gets a bit more complicated than a gas forge because the gas flow is variable, but it's pretty boring. The Type S thermocouples have had 3/8" to 1/2" diameter Recrystallized Alumina sheaths and Stainless Steel support tubes. They are inserted through the wall and are horixontal in use. Apart from needing to be stiff enough to self-support over about 12" of insertion depth, there is no real need for mechanical strength and it's an open-topped chamber so pressure is negligible. I've therefore not used thermowells. I've repurposed one of the type S thermocouples for measuring forge temperatures when I've been playing with burners If you've got any specific questions I might be able to help with, I'll try to answer them. The best advice I can give is to write down what you need (as distinct from what you think you want), make a nice strong cup of tea/coffee, pick up the phone and sit down for a chat with Omega. They are specialists in temperature measurement and control and deal with this stuff regularly. If you have a local thermocouple manufacturer, talk to them as well (not instead of) I've not been keeping up-to-date on IR temperature measurement, but it was looking like it would give pretty good results when looking at the inside of a thermowell or similar, since a hole has an emissivity of 1.0. It may be worth bearing in mind.

The sand tray is a pretty well-established method for dealing with temperature oscillation. It works. I'd suggest an ebay search for "TM902C". It'll bring up lots of suppliers of a type K pyrometer for 5 or 6 bucks delivered. The thermocouple supplied with it is a 3' long glass-fiber-braided thing that is good to about 400 degC/750 DegF and is flexible enough and thin enough to shut in most oven door seals. Downside is the degC-only readout. With a suitable type K thermocouple, the TM902C will read to about 1368 degC, the limit of the typeK table. I've had several on the calibrator at work and accuracy is as good as any of the big-name pyrometers at 10-40 times the price.

Lack of 3-phase power is not usually a problem unless you need to run big stuff. A 3 HP single-phase motor can (usually) be run off a 13A socket. 3 HP is about as big as single-phase compressors get. You can go a bit bigger if you hard-wire to a dedicated circuit with a suitably-sized breaker back at the distribution board and can find a single-phase motor bigger than 3 HP (2.2 kW). You can use a phase converter to convert from single-phase 230V to 3-phase 400V if necessary, though the power will be limited by the capacity of the incoming supply. A VFD will convert single-phase 230V to 3-phase 230V and the Chinese ones off ebay are pretty cheap (usually under a hundred quid delivered from within the UK) and seem to work ok. If you can, it's a good idea to get a circuit or 2 put in with industrial 230V sockets and type "C" MCBs which gets you over the 13A limit. If you are running a welder, a 32A socket and type "D" MCB is wise, as the inrush current can be pretty high. http://www.screwfix.com/c/electrical-lighting/industrial-plugs-connectors/cat830086#category=cat830090

£1000 over 5 years is £200/yr, £4/week, £1/day in round figures. At a rate of £20/hr, that's 3 minutes/day to check the fuel, oil and water, top-up if needed, start the thing, stop it at the end of the day, sort out the fuel deliveries, maintain it and do all the other assorted faffing about that goes with a generator. You need to be pretty good to get it all done on that sort of timescale. Even if you cost the time at minimum wage, it's only about 9 minutes a day. Add in the noise, the need for bunded fuel storage, maintenance, etc and it gets even more convincing. It has not been mentioned, but mains 3-phase is also much more versatile in terms of what it will run. Inverter drives (VFDs) tend not to like generators very much, if at all. Apparently this is because of the "dirty" waveform and the harmonics. I've certainly had a number of VFDs that simply will not run reliably off generators. Starting fixed speed motors on generators can be troublesome as well. Direct-On-Line starting a motor draws 6-7 times the rated motor current. Star-Delta start is gentler, but still pulls 2 1/2-3 times the rated motor current. You therefore tend to need a big generator for motor loads, just to handle the starting surge, whereas the mains can generally cope with starting surges without a problem. It may be a consideration if you are likely to want to add to the shop in the future. Hydraulics in particular can be heavy on starting.

Unless it needs to be movable, I'd go for the stump and probably bed it in silicone. I made a stand from end-grain timbers and routed in a pocket for the anvil to sit in, then cut a piece of Anti-Vibration matting to fit the pocket (Fab-Cel 25, left over from a job at work). http://uk.rs-online.com/web/p/anti-vibration-pads-chips-tape/3660150/ The difference in noise level with the mat and without it is amazing. There is no discernable ring at all with the mat, but it rings like a bell without it. If I was doing it again and didn't have the mat, I'd be going for silicone instead. I'm pretty sure it's important to have the entire anvil base in contact with the material for acoustic coupling: the sound waves get straignt out into something that doesn't transmit sound very well, rather than being internally reflected off the bottom of the anvil and bouncing around in side something that does transmit sound well.

The burner sputtering at low flow is fairly usual. When the mixture speed in the burner tube is low, the flamefront can move down the mixture faster than the mixture moves up the burner tube. The flamefront runs down the tube until it runs out of mixture, when it goes out. Fresh mixture flows until it reaches the hot chamber, where it ignites and the whole thing happens again. If you do nothing, it will often get worse as the burner tube heats up and the flame speed increases. Flames burn faster at higher temperatures. You will sometimes find the sputtering only starts once the forge is at working temperature. Standard fix is to turn up the pressure, thereby increasing the mixture speed, until the sputtering stops.

It really depends on the burner, the forge, and what you are doing with it, but as a general rule, it shouldn't. Usually the burner is recessed into the Kaowool and in some cases, the Kaowool and its coating are formed to make a burner flare.

If my quick calculation is right, your 144 M3/hr is around 85 CFM of N2 or 180 kg/hr (400 lb N2/hr). I assume you are using Oxygen-Free Nitrogen. If you can live with lower purity, an on-site membrane-type Nitrogen generator may be a (much) cheaper option. It needs a clean, dry compressed-air supply and uses membranes to selectively remove Oxygen. Pressure-Swing-Adsorption systems are also available and offer higher purity than membranes, but I have no experience of them. 100 CFM is not a big system in industrial terms. I'm guessing you are not in the USA by your use of units. I'd suggest speaking to your nearest Atlas-Copco distributor to get an idea of what is available, then shop around once you have an idea of what you need.

I think the OP missed some of the details: "This Ironton single phase 120/240V, 2 HP compressor motor with capacitor start design is ideal for use on compressors, pumps, pressure washers and other clean and dry applications that require a high starting torque." Copied from: http://www.northerntool.com/shop/tools/product_200580992_200580992 It looks like it's an Open Drip Proof design: only offers protection from vertically-falling drips. I'd pass.

While Frosty's got his burner head on, a couple of other things spring to mind. What is the forge construction and is it properly dry? What happens if you increase the gas pressure on the .023 tip? Pics will definitely help.

I'd go with the muffle idea and try to avoid direct heating by the burner flame altogether. I tend to use Venturi burners and the idea of trying to balance the mixture (and thereby the flame temperature) on multiple gas mixers seems like way too much effort: I'd go one big one every time. With blown burners, you have a better chance of balancing them, though getting them all to run evenly off one mixer tends to mean more pipework filled with a potentially explosive gas/air mixture. I'd still go one big one. Type K thermocouples look quite good on paper (the tables go up to 1370 degC, 2500 degF), but tend to suffer from "drift" when used above about 1000 degC. 1000 degC is 1832 degF, but note the "about". I'd expect your application at 1800 degF to cause drift in type K, albeit slower than at higher temperatures. Thermocouple types R and S are very stable and make a great cost-no-object choice up to 1300 degC, 2372 degF, but are Platinum-based and priced accordingly. R and S are actually good to around 1760 degC, 3200 degF and there are no cheap options at these higher temperatures. If your temperature readout will accept type N thermocouples, these are an excellent choice up to their limit of 1300 degC, 2372 degF, having been developed as a sort of improved type K without the drift problems.

Do a google search for D-bits. They are dirt cheap, easy to make in the shop from drill rod and they work extremely well. There's a pretty good description of how to make them half-way down the page at: http://modelengineeringwebsite.com/Holes_part_3.html Silver steel is the British version of drill rod and is water hardening. Modify your HT to suit the available material.

Just a thought. I've pinned firebrick forges together with welding rod to give them a bit more structural integrity. I tend to use 2.4mm (3/32") stainless gas/tig rods as the pins. I grind away half of the end 10mm (3/8") or so to make a rudimentary D-bit. I get a bit of scrap wood a couple of inches thick and drill a hole through it on the drill press to get a guide hole at 90 degrees to the face, hold this on the brick with enough of the welding rod in it to go all the way through the guide block and the workpiece, chuck the rod in a battery drill and just drill through. When the rod comes out the other side, I snip the excess off and leave the pin in. It works well on Insulating Fire Bricks because they are soft and the waste just seems to get packed into the pores so there's no need to peck. As they are single-use, there's no need for a hardened tip. The longest I've drilled and pinned through IFB is 19 1/2", but the limit is probably the length of rod available. Incidentally, D-bits work well on other materials too, though they are usually best made from drill rod and hardened. They'll drill deep straight holes (though they need a *lot* of pecking due to the absence of flutes) and cut flat-bottomed holes. Used with care (slowly and with lots of pecking to minimize heat buildup), they will drill through holes in wood blocks for stick tangs at a fraction of the cost of an extra-length drill. Shop-made D-bits were a mainstay of British model engineering until low-cost drills and milling cutters started arriving from the Far East. They are still very useful for one-offs and some of the more specialized jobs. If you've not come across them, they are worth Googling.

I'd suggest the biggest problems are probably, in no particular order, a hot climate, possibly a problem with the unloader, and single phase. If there is any way you can realistically get 3-phase to it, it will greatly increase reliability. 3-phase motors can generate a decent amount of starting torque where single-phase have to have life made easy for them simply in order to get started. Religiously change the pastille in your non-return valve every year (maybe more if you are somewhere hot; I'm in England and change them annually) and check that the unloader on the bottom of the pressure switch spits out air when it switches off, but at no other time. The idea behind these 2 components is that the NRV stops tank pressure getting back to the pump head and the unloader dumps what was there when the pump last ran. This lets the pump get a revolution or so in before the pressure builds between the pump head and the NRV and lets it get somewhere near to full speed before starting to do real work. If the pastille goes brittle, it stops sealing, pressure bleeds back to the pump and it has to start against a load.

Dave, can you post pics of your setup? I have struggled to find any with the search function. The only things on the inlet side that I can think of that might help, would be to increase the mixture pressure in the plenum (which would increase the gas speed through the holes in the ribbon burner block), or to adjust the air:fuel ratio (maximum flamefront speed is usually somewhere around the stoichiometric mixture, so moving away from the stoichiometric tends to reduce the flame speed; of course it also affects the flame temperature, which reduces as the mixture gets further away from stoichiometric). Air:fuel ratio adjustment "just" to prevent the explosions would obviously limit the operating air:fuel ratio range you could use. Much of the beauty of a blown system is the control it provides over the air:fuel ratio, so it would seem a shame to have to limit it. Once the flame breaks through into the plenum, the only difference that any design changes upstream can make, will be to increase or decrease the volume of mixture that is available to explode. Having a bit less volume would be no bad thing, but it would seem preferable to address the cause of the problem first. I am pretty certain the cause lies somewhere in the forge and burner installation. Obvious things to check would be that the forge flows freely enough to allow the burner to work unrestricted, and that the feed pressure is as per the burner manufacturers spec. The pressure can usually be measured easily enough with a U-tube manometer (a length of clear tubing bent into a "U" shape and half-filled with water. Apply pressure (or suction) to one end and the water column will become lopsided. Measure the height difference between the water levels in each leg and you have the pressure in inches of water column. Colored water helps, and make sure the length of the U-tube is at least one-and-a-half times the maximum anticipated pressure, so it can't blow the water out. Getting a handle on how well things are flowing in the forge itself is mainly about looking for anything that might restrict flow. There is no easy way that I know of to get good quantitative data. I gather you are using the second-smallest of the Pine Ridge burners; presumably that's the GH190? It looks like it needs 7"-56" WC of pressure. Higher pressures will give higher speed through the burner ports and less chance of the explosions. What pressure are you actually getting? How big is the forge? The GH 190 is recommended for "up to 1.5 cu ft" (about 2500 cu in) That seems pretty big to me. My guess is that the burner would want a good 10"-plus between itself and the opposite wall to allow it to flow freely (there is a photo on the Pine Ridge website of a "miniburner top-mounted in a small forge" that seems to show the ceiling about 9" above the floor. Other photos on the site seem to show the burner recessed about 1" into the wall). If you've built an unusually-shaped forge for specific work, this might be an issue. Like I said, pics would be good.

Dave, what I think is probably happening is this: On startup, everything is nice and cool. The gas/air mixture in your plenum is also nice and cool and the holes between the plenum and the forge chamber are similarly cool. When you light the forge, things are still cool, but the hot face of the ribbon burner block starts to heat up. The back face is being cooled by the incoming fuel/air mixture. As the hot face gets hotter, the heat moves through the burner block and it gets hotter. To work as intended, the speed at which the gas/air mixture moves through the holes in the burner block, needs to be faster than the speed the flame-front can move through the mixture. As the temperature rises, the flame speed increases. Also, as the pressure rises, the flame speed increases. When you get a moving flame-front, it radiates heat ahead of itself, raising the temperature of the mixture ahead of it, and, because there is huge expansion during combustion, there is also a pressure wave ahead of the flamefront. If it is relatively cool, the burner block "should" help to slow down the flame front by absorbing much of the radiated heat. This is most effective with small holes: it is pretty much the same principle as a flame arrester. At some point (45 minutes to an hour in your case), the temperature of the block reaches the temperature at which the flame speed is higher than the mixture speed, the flame runs through one or more of the holes in the burner and reaches the plenum, where there is a big open space and the walls are far enough away not to absorb much of the radiated heat. The flame-front accelerates across the plenum until it runs out of fuel, but may reach the speed of sound as it travels. A supersonic flamefront is "detonation". A Subsonic flamefront is "deflagration". Does it seem likely to you?

I'm assuming the reference to running a compressor for 2 hours at a time is for some sort of Babbington-style burner? If so, I think you may have misunderstood the (essential) compressed air requirement for this type of burner: there is a (very) small flow of high-velocity air to provide the atomization and this needs the high pressure supply, but pretty much all the actual combustion air comes from a low-pressure fan or is provided by natural draft, depending on the detail design. The typical air jet for the atomization part seems to be around .010" with atomization airflow under 1 CFM. I am sure that there are designs out there that supply all the combustion air from the compressor, but it's not the only way to do it.. The Babbington-style air atomization is often favoured for waste oil burners because it is relatively tolerant of dirty oil. Atomization systems that rely on feeding pressurized oil through small jets are much more prone to clogging and, as far as I can tell, there tend to be more difficulties when dealing with waste oils of different viscosities/temperatures. If you can get the oil clean enough and the viscosity low enough, a modern Common-Rail Diesel would seem offer some potential as a parts source for a highly efficient atomisation system.

Then it is not the design from Porter's book. From your photos, it looks to me as if you've taken Mike Porter's carefully developed and documented construction methods for a Venturi burner, dispensed with any part of it that actually resembles a Venturi, and produced something akin to an "upwind" burner. Albeit a beautifully-made upwind burner. I may be misreading the photos, but it looks like there is no reduction to produce a Venturi throat and therefore no subsequent expansion; the essential components of a Venturi. http://en.wikipedia.org/wiki/Venturi_effect The upwind design is well proven, so you should not have any great trouble getting it to work, but it is unlikely to give quite the control over air:fuel mixture that a good Venturi design can.

If you are "only" looking at a £300 phase converter, it's probably not worth doing 3-phase distribution wiring. For the guys across the pond, I should point out that UK and European mains distribution are rather different to North American ones. Our domestic supplies are 230V single-phase to Neutral and we get one phase of a 400V phase-phase, 230V phase-Neutral distribution system. This means that the first stage in a UK phase converter needs to be a step-up transformer from 230V to 400V. After that, things are the same both sides of the pond (except the Voltage of course). The transformer is a big chunk of cost and £300 represents maybe a 3HP static. If you've lucked into a 5HP, or larger, Rotary at £300, the wiring makes more sense. For reasonably modern motors up to about 3HP (say 1970 onwards, maybe even the late '60s), you usually have 6 terminals in the motor and can run in Star (Wye) for 400V and Delta for 230V. These can be run in Delta from a 230V Variable Frequency Drive: single-phase in, 3-phase out and you get variable speed as a bonus. The price for a new Chinese 2.2 kW VFD, buy-it-now on Ebay, has just dipped to £69.50 delivered (shipped from within the UK, so no import duties). I've played with 3 of them so far (I bought 2 myself and set one up for someone else) and they all worked perfectly. If you are going to be messing around with your shop wiring, it's never a bad idea to have a nice meaty circuit with a 32A blue CEENORM socket on it and a 32A type "C" MCB feeding it, as it gives you somewhere to plug in welders, phase converters, etc, over 3 kW.

That would make sense Mike. Except there seems almost never to be a jet size mentioned when I see pressures quoted in relation to welding temperatures. Like Frosty says, the wide variety of burner designs out there seems to make the likelihood that apples are actually being compared with apples pretty small. That's before we even start on the differences between forges. Personally, I'd rather build a Venturi burner to give welding temperature at around 30 PSI and take as much turndown as I can get for when I don't need to run that hot. To use half the gas, I am fairly sure I'd need to drop from 30 PSI to 7.5 PSI, as pressure vs flow through an orifice follows a square law. Going up from 30 PSI to 60 PSI would give around a 41% increase in gas input on that basis, but I think the flow chokes somewhere around 30 PSI so the increase would actually be less. Unless I've completely misunderstood the physics, turning down from 25 PSI to 12 PSI actually reduces gas consumption by only around 30%.

It's a bit bigger than the home-scale systems you are referring to Frosty, but this is the gas flare on a site I occasionally work on. The site has 5 anaerobic digesters that process around 350000 tonnes of organic waste a year. Generation capacity using the Methane produced is around 7 MW into the grid, using 4 MWM spark-ignition-engined generators. The flare is only used to burn off any excess gas and is big enough to cope with the full gas production if we have a mains failure and cannot export the power. Thermal capacity is 18 MW (around 61.4 million BTU/hr). It is lined with 4" or so of Kaowool and it PID-controls the temperature to 1000 degC (1832 degF) by adjusting the air:fuel ratio using the 6 air dampers around the base of the unit. Basically, it's a slightly scaled-up gas forge.

If it's a Flamefast-manufactured part and they will not sell a diaphragm separately, you are pretty much stuck with having to replace the valve. If it's a bought-in standard part, you have a chance of bypassing Flamefast. Got any photos of the offending valve? And where in the world are you?