Amal gas burner upgrade?

[James @ Holkham Forge]

I have an old homemade (not by me) propane forge which has a single gas pipe which has 3 holes drilled in it which sqirts the gas down into the forge. Unfortunately the jets are a bit off target so have a tendency to burn in the top part of the burner rather than down in the forge (particularly at low pressure). I was planning to upgrade it using 3 x 1/2" Amal buners but wasn't sure about how long the pipe should be from the end of the long venturi to the forge? I've tried a 3/4" burner and it seems the shorter it is the better (if better is a more of a roar from the flame). I'm not quite sure how the length of pipe will effect the flame and heat. Also does the offset angle of the burner affect the heat? Thanks for 

existing forgeburners and exhaustsAmal 3/4 inch burner plus screw in pipe

[timgunn1962]

I've not found the length to be critical. I tend to use somewhere between about 8 and about 12 diameters of pipe, but only because it looks about right: I've not done any testing to find an optimum length.

I think the idea is that the flow is laminar for most, if not all, of the copper expansion section and becomes turbulent pretty much where it hits the full pipe diameter. The turbulent flow in the pipe will be where the gas and air do most of the mixing.

I assume by offset angle you mean the 45-degree bend?  If so, I don't think it will cause any problems.

I usually spend a bit of time fettling the Amal injectors before using them: 

I file off the casting flash on the tapered boss the jet screws into, so that it can be choked down to almost zero air gap. 

I unscrew the choke and lubricate the threads liberally with graphite or Molybdenum Disulfide dry lube. Industrially they are usually set-and-forget, but for smithing (and particularly bladesmithing), they are likely to need more frequent adjustment. I've found grease tends to pick up dust and grit and is as bad as no lube. 

I use lots of PTFE tape on the burner pipe thread. Much more than is normal. This is to protect the soft aluminium threads from being damaged by the steel pipe. The mixture adjustment relies on these threads, so they are worth protecting. The gas-carrying threads get an anaerobic pipe seal, usually the Screwfix one. This includes the jet threads.

The 1/2" burners are very small. The heat output varies with area, so a 1/2" has about half the output of a 3/4", which in turn has about half the output of a 1". 

I know from experience that a 1" will  happily handle a 6" x 6" x 13 1/2" chamber in an Insulating Fire Brick forge with about a 3" x 4" opening (built with bladesmithing in mind). The size of the open area can make a big difference.

There's nothing really in your forge pics to give me a sense of scale, so it's hard to know what size burners would be appropriate.  I have found the following jet sizes to work well in forges (Propane and only primary air) 1/2" injector: 30 jet, 3/4" injector: 60 jet, 1" injector: 120 jet.

Amal are/were a carburettor company and size their jets in cc/min of petrol, so tuning jet sizes can be something of a trial-and-error process. 

 

 

[James @ Holkham Forge]

Thanks Tim. Lots of fantastic advice there!

I will certainly clean up the burners and put a bit of effort into keeping it working well.

The Forge is 26" long with an opening about 6" x 5". We make bespoke naturalistic forged commissions and have used it for long heats when a coke forge is too localised, but as I said it has been feeling increasingly inefficient, stopping us making full use of the possibilities of the tool.

I was thinking of replacing the 3 burners with the smaller 1/2" ones to keep the long heat possibilities, although today I welded the 3/4" pipe into one of the exhaust ports and used just a single burner and it heated the front section of the forge area nicely when forging some flower elements in 3mm sheet even at really low pressure which was a bit of a revelation. It was interesting to explore the impact of varying the pressure through the regulator and the choke of the burner. I'm not sure I'm much the wiser yet but it feels much more controllable than the homemade set of three.

How well do you think a bank of 3 in series would work from the same tank?

Thanks again, James

[Richard Furrer]

How do these 3/4" injectors work at 5PSI? I have a max of 5PSI natural gas in the shop.......I would think I could simply increase the orifice.

 

Ric

[James @ Holkham Forge]

Hi Ric,

I was using the 3/4" burner on a nearly empty propane tank at what my old acetylene regulator gauge said was a bit over 0.5bar. I opened the tank tap and only turned the regulator a little and that was enough to create a hotspot 6"x 5" x 9"  for me to forge the mild steel sheet elements I was working on. How accurate the gauge was is open to question but it was very low pressure and it made a lot of heat. I guess 5 bar would be absolutely raging hot!.

One question I did have was about the Jet Size that Tim suggested  " 1/2" injector: 30 jet, 3/4" injector: 60 jet, 1" injector: 120 jet."

The Amal website says the 1/2" is Amal gas jet number 70, the 3/4 number 110 and the 1" number 180. Should I ask Amal for some smaller jets or am I getting confused somewhere?

Thanks.James

[j.w.s.]

Is anyone selling these burners in the states?

-J

[timgunn1962]

I think the Amal factory recommended jet sizing is based on normal industrial combustion processes.

"We" are not normal, at least insofar as combustion processes go.

We tend to run with only primary air (air that mixes before the burner nozzle), so we need more of it than systems that also have secondary air (air that mixes after the burner nozzle). The amount of air depends primarily on the gas speed out of the jet. To get more primary air, we make the gas jet smaller and run at a higher gas pressure to give a higher gas velocity.

There is a relationship between flame temperature and air:fuel ratio, with maximum temperature at an equivalence ratio of about 1, falling away on both sides as the equivalence ratio goes higher or lower. 

The graph is theoretical only and is calculated for ideal conditions. We'll see lower temperatures in the real world, but the graph shows the shape of the curve.

Most industrial processes run out to the right-hand side of the graph. We tend to want to run on the left-hand side where we get a reducing atmosphere to limit scaling. 

I have tried to size the jets to get about the peak temperature with the choke fully open and make the full range of temperature adjustment available by closing down the choke and running down the curve on the left-hand side. The ones I have suggested are the jet sizes I have found worked best in my testing.

 

 

 

[timgunn1962]

There is someone in the States listing 3/4" NIB Amal Atmospheric Injectors on ebay. Search for "Amal Atmospheric Injector".

I think the jet sizes will be much too big for Propane, as they are Natural Gas injectors.

http://amalcarb.co.uk/downloadfiles/amal/amal_gas_injectors.pdf

I don't know whether Amal made a version with NP threads for the US, but I doubt it. As standard, the threads are BSP, which might be a minor pain. 

When I was messing about with jet sizes for the 1" burner, I used a 1/8 BSP to M5 reducing bush and M5-threaded miniMIG tips. My 1" burner worked well with a 0.6mm tip (.023"), so a 3/4" would want a smaller jet. As .023" seems to be the smallest standard MIG tip, that option is out.

Burlen Fuel Systems supply carburettors for old, mainly British, cars. They certainly ship to the States.

 

[James @ Holkham Forge]

Thanks Tim: more amazing info and insights!

I spoke to Amal/burlen today and ordered a 60 jet that will fit in the 3/4" burner I already have so fingers crossed I can achieve the efficiency and heat that seems possible from this great little burner.

 

[James @ Holkham Forge]

I thought I'd post an update and I just wanted to say thanks Tim. I've got two 3/4" Amal burners working now with 60 jet nozzles and today the heat from my old homemade forge was insane. I was forming scrolls around a jig and the heat was pretty much 2ft long on batches of up to 6 pieces of 16mm square mild steel, and the gauges on the regulator said it was only 1.5 bar on a bit less than a quarter of a tank. Thanks so much for your help and advice. It is now really a productive bit of kit! I can screw the burners on and off which means a really flexible workspace.

[Charlotte]

Congratulations on your success!  That is a  very nice picture an a good taper on the scroll!   Good Work!

[Mikey98118]

 

MIG contact tip sizes are listed for the MIG wire size they are meant to  feed. A .023" MIG contact tip actually has an orifice diameter of .031"

BTW, theoretical is exactly right; that chart is showing the mathematically derived adiabatic temperature of an air-propane flame,and is wildly over optimistic. Air-acetylene flames are only rated at 3600 F!!!

[timgunn1962]

5 hours ago, Mikey98118 said:

 

MIG contact tip sizes are listed for the MIG wire size they are meant to  feed. A .023" MIG contact tip actually has an orifice diameter of .031"

BTW, theoretical is exactly right; that chart is showing the mathematically derived adiabatic temperature of an air-propane flame,and is wildly over optimistic. Air-acetylene flames are only rated at 3600 F!!!

The clearances on MIG tips seem to vary slightly. Most of the ones I used were no-name M5-threaded minimig tips and the hole sizes were around .006" to .008" above the nominal wire diameter.

The chart is indeed a plot of theoretical adiabatic flame temperatures. I used the online calculator at:  http://elearning.cerfacs.fr/combustion/tools/adiabaticflametemperature/index.php

Clearly it has its limitations, but it was the best I could come up with and my main concern was to get an indication of the shape of the curve, rather than any actual value for achievable temperature.

If you can point me at a more useful graph, or some data on which to base one, I'd be most grateful

As I pointed out when posting the chart, we can expect to see lower temperatures in the real world. How much lower is not easy to quantify. There would seem to be a number of variables and it's probably safe to say that the maximum attainable temperature will vary quite widely between forges.

I've measured actual forge temperatures of over 1550 degC ( 2822 degF), but have not really tried to push for any higher. It's hotter than I'm ever likely to need and I don't want to risk an expensive Platinum-based thermocouple chasing unnecessary temperatures. In all honesty though, I don't think I'd get much above 1600 degC, even on a good day, without going significantly bigger.

I was primarily interested in covering the full range of temperatures useful for bladesmithing, from Austenitizing at around 760 degC (1400 degF) to welding at 1320+ degC (2400+ degF) with a single (relatively) cheap-and-cheerful setup that a novice could build with limited equipment. 

Near the top of the temperature range:

And towards the bottom end of the range:

 

I was trying for 800 degC (1472 degF) and 797 degC is 1466 degF. Incidentally, the minute and sixteen seconds shown is because the pyrometer has autoshutoff after some arbitrary period of time and I'd switched it back on a minute and sixteen seconds earlier. 

The Mikey burner had seemed difficult to build for a beginner without a fairly well-equipped shop, while the design of the Frosty T-burner seemed to present difficulties in arranging a choke that would give the sort of fine and progressive control that makes it easy to hold an Austenitizing temperature (here in the UK, we have easy access to O1 Ground Flat Stock, but most other blade steels are difficult to find and often involve high international shipping costs, including 1084, so being able to soak for the time O1 needs can often repay the extra cost of a burner that will do it). I'd been using Amal injectors at work for over 20 years, but had no idea what they cost. When I found out the price, it became a no-brainer to use them. IIRC, the last 1" injector I had cost around 55 GBP delivered (and including 20% Value-Added Tax), equivalent to about 85 USD. Not cheap by hobbyist standards, but it seemed very good value.