[mel]

i'm trying to gain a more fundamental understanding of our heat treat process.

we sometimes do neutral hardening of steel in a batch furnace with an endothermic gas atmosphere. carbon potential in the furnace is controlled using the output from an oxygen probe. Typically, extra natural gas is added at the start of a cycle in order to get the carbon potential up. The natural gas additions kick in when the temp ramps up past 1500F. when carbon potential set point is reached, the natural gas additions stop and for the rest of the cycle, small air additions are occasionally made to the furnace to keep the carbon potential from rising beyond the setpoint.

i recently borrowed a CO/CO2/hydrocarbon gas analyzer. i found that due to the natural gas addition at the start of the cycle, the CH4 concentration reaches ~15% before it begins to drop off.

my questions...

1) how long does it take to arrive at an equilibrium gas mixture when CH4 is added to endothermic gas? does anyone know of a good source for relevant reaction rate equations?

2) i expect a CH4 concentration of 15% is too high. Is there a CH4 concentration that I definitely don't want to exceed during the ramp up to carbon potential setpoint?

Any answers or comments will be appreciated.

mel

[Richard Howell]

First of all, the atmosphere is never in equalibrium. The atmosphere is consistently changing, so additions of either air or methane is always required.

How wet or dry is the endo gas from the generator?
Could you adjust the ratio so the dew point or carbon potential is closer to you neutral atmosphere requirment?
What is the temperature that you are holding the load at for neutral hardening?
Is there a time or temperature after introducing the methane that you the decline beginning?

Methane is the slowest of the hydrocarbons in its rate to thermolly decompose, and is therefor is most easly controlled.
Adding methane at 1500 deg F., the gas reaction will be extremely slow so, in my opinion, the build up of 15% methane would not present a quality issue.

[mel]

Thanks for the response.

Dew point of endo entering furnace ~35F.
Furnace temp = 1600F.
Desired carbon potential ~0.7%.
I think that once a sort of steady state is reached, the endo pretty much gives me the carbon potential i want; it's during the initial transient that the natural gas is added.
I realize that the atmosphere doesn't achieve equilibrium; i was just wondering if anyone had an idea how rapid the approach *toward* equilibrium is.
I also saw a rule of thumb somewhere that said the ratio %CH4(actual)/%CH4(equilibrium) should be less than 10. I think the equilibrium CH4 concentration in my atmosphere should be about .2 - .3%, so the 15% I actually see (although it doesn't stay that high) is in serious violation of the less-than-10 rule.

I'll add a couple more questions...

In my system, carbon potential is determined indirectly from the output of an oxygen probe, which (as I understand it) is basically measuring the partial pressure of oxygen in the furnace. It seems that some of the ways in which CH4 could reduce the oxygen partial pressure (i.e. increase the carbon potential) are: (1) direct reaction of CH4 with oxygen, (2) reaction of CH4 with CO2, which would then pull other atmosphere components into reaction with oxygen, (3) reaction of CH4 with H2O, which also would cause other components to react with oxygen. I suppose there are other things CH4 might do as well. Is there a dominant reaction path that CH4 follows when it is introduced into an endo atmosphere? Is it one of the paths listed above, or another?

I was told once that excessive CH4 concentrations in the furnace would create sooting problems. What is the dominant sooting reaction in a furnace like mine? Is direct dissociation of CH4 the main soot forming reaction? Or is soot mainly formed through a reaction resulting from an undesirable CO/CO2 ratio, and CH4 contributes to sooting by reacting with atmosphere components such that the CO/CO2 ratio is pushed in the direction of greater sooting?

Any answers and comments will be appreciated.

[joakes]

The time needed to reach furnace atmosphere equilibrium has a lot to do with the furnace size, load size and how much air might infiltrate the furnace. A good baseline to determine if this is related to air infiltration would be to check the gas composition of your endothermic generator and then check the furnace with straight endo flowing. That will identify if the furnace is compensating for a continuous air leak or if there is air getting into your endo line.

One good way to test this is check the dew point in the furnace while running straight endo. Make sure you shut off the gas and air valves at the furnace. If the dew point of the generator is 45 then the dew point in the furnace should be within 5 points. For example if you are at 1600 with a endo gas of 45 the carbon should be in a range of .4 to .5 carbon. See the following chart http://www.supersystems.com/CARBDEW.PDF. If the carbon is lower, most likely you have a door leak or tube leak. CH4 will be added to overcome the air. If the CH4 being added is not able to crack then you will see an increase in CH4. As the furnace gets lower in temperature the CH4 will read higher due to the gas not cracking.

Did you happen to use the gas analyzer on the output of your generator? It would be interesting to know what the CH4 value is after the gas is cracked. In many cases you can achieve the desired dew point at the generator with a bad catalyst which would produce higher methane values coming out of the generator.

Using a three gas will allow you to determine if the gas composition on recovery. Once you reach set point you should see the CH4 drop and the CO climb to 19.0 or as high as 20.0. If you have the opportunity to run a three gas on a furnace that does not have any issues it will help build a baseline of what the CO and CH4 should for a given temperature and carbon set point.

Thank you,
Jim Oakes – Super Systems Inc.

http://www.supersystems.com/

[aoneill]

I'm sorry that I cannot answer the questions you asked but I did notice you stated 'small air additions are occasionally made to the furnace to keep the carbon potential from rising beyond the setpoint'.
Normally you'll have a supply of 'burn out air' to be utilized when you start up the furnace to burn out carbon that might build up inside so you don't have a fire, and then for the probe you'll have a supply of 'reference air' which is on all the time and 'soot burn out air' that is automatically turned on/off via program and a solenoid valve to keep the probe clean.
With regards to your questions, contact your furnace manufacturer and have them explain it all to you, they should have a vested interest in your successful use of their equipment.
Good luck!

[BethRyan]

Maybe this will help:

Troubleshooting Endothermic Generators -

http://www.secowarwick.com/metalminu...ps/endogas.htm

[DrOrpheus]

I gotta ask. What do you run at 1600 and .7% carbon? Also what kind of furnace,style,size,cfh flow of endo going in? Have you plugged in and check on your endo generator? What temperature do you run that at? And when you say you sometimes run neutral hardening out of the furnace, what is happening with it and the generator when you don't?

[josephmarkgreene]

A good source of information is the ASM Handbook. I believe that it is the Heat Treating volume that has a whole section on furnace atmospheres. Another good source would be the Heat Treat Doctor, Dan Herring, who has worked with carburizing in endo atmospheres for longer than most of us.

Back when I worked with atmosphere carburizing, the rule of thumb was that the natural gas enrichment could peak as high as 20% during the initial transient. At 15%, it sounds as if your equipment is operating within normal ranges. If it were not, the side-effects would show up by way of sooting of the furnace or carburizing of the parts that you wish to neutrally harden. Sooting is also a function of temperature. Carbon monoxide is actually more stable at higher temperature (thus the high operating temperature of your generator), so a given gas concentration in the furnace is more prone to sooting at lower temperatures, such as for carbonitriding. A given gas concentration also has a higher carbon potential at lower temperatures.

Steel isn't very receptive to carbon below ~1450°F, so starting the enrichment at 1500°F makes sense from that standpoint as well as from the stability of the atmosphere with respect to soot. Also, because of the parts have thermal mass, their temperature follows that of the furnace thermocouple.

While enriched endo truly is not in equilibrium, it is generally thought of as relatively close, much closer than direct-feed atmospheres such as nitrogen-methanol, which is especially slow to crack at carbonitriding temperatures. I don't know how to quantify the reaction rate, but the rate-controlling reaction in terms of carburizing is generally published as the reduction of carbon monoxide by hydrogen. This is not what is measured by your oxygen probe, but because the CO concentration is relatively invariant at ~20% (assuming that the generator is fueled by natural gas), the oxygen partial pressure turns out to be a reliable indicator of CO/CO2 and H2/H2O ratios that figure more directly into the relative equilibrium. If your endo is produced from propane (LP gas), then the CO content of the endo is higher, I believe just over 23%. While this makes all of the calibrations change, the application of the oxygen probe works much the same. In certain direct-feed atmospheres, the CO content inherently varies over a wider range. This introduces an additional variable into the equilibrium calculation, and an oxygen probe by itself turns out to be inadequate without a CO sensor.

I am not sure what the dominant reaction is of methane, but assume that it is quickly oxidized to CO by H2O, CO2, and O2. Thus, as CO as reduced by hydrogen and the carbon incorporated into the steel surfaces, the CO content of the atmosphere is maintained relatively constant by the oxidation of methane by the three most strongly oxidizing species. Maybe someone can confirm this?

Tweaking your generator to lower dew point will require less natural gas addition at the furnace. However, 35°F is right in the sweet spot for keeping soot down and generator maintenance moderate. I no longer recall where this came from, but 20-40°F dew point at the generator was recommended as optimal from the standpoint of generator maintenance. Personally, I am partial to 30-35°F.

Hope this helps.

[josephmarkgreene]

A good source of information is the ASM Handbook. I believe that it is the Heat Treating volume that has a whole section on furnace atmospheres. Another good source would be the Heat Treat Doctor, Dan Herring, who has worked with carburizing in endo atmospheres for longer than most of us.

Back when I worked with atmosphere carburizing, the rule of thumb was that the natural gas enrichment could peak as high as 20% during the initial transient. At 15%, it sounds as if your equipment is operating within normal ranges. If it were not, the side-effects would show up by way of sooting of the furnace or carburizing of the parts that you wish to neutrally harden. Sooting is also a function of temperature. Carbon monoxide is actually more stable at higher temperature (thus the high operating temperature of your generator), so a given gas concentration is more prone to sooting at lower temperatures, such as for carbonitriding. A given gas concentration also has a higher carbon potential at lower temperatures.

Steel isn't very receptive to carbon below ~1450°F, so starting the enrichment at 1500°F makes sense from that standpoint as well as from the stability of the atmosphere with respect to soot. When you consider that the temperature of the work follows that of the furnace, there is no reason to be overly hasty about starting the enrichment.

While enriched endo truly is not in equilibrium, it is generally thought of as relatively close, much closer than direct-feed atmospheres such as nitrogen-methanol, which is especially slow to crack at carbonitriding temperatures. I don't know how to quantify the reaction rate, but the rate-controlling reaction in terms of carburizing is generally published as the reduction of carbon monoxide by hydrogen. This is not what is measured by your oxygen probe, but because the CO concentration is relatively invariant at ~20% (assuming the endo is produced from natural gas), the oxygen partial pressure turns out to be a reliable indicator of CO/CO2 and H2/H2O. If your endo is produced from propane (LP gas), then the CO content of the endo is higher, just over 23% if memory serves. While this makes all of the calibrations change, the application of the oxygen probe works much the same. In certain direct-feed atmospheres, the CO content inherently varies over a wide range. This introduces an additional variable into the equilibrium calculation, and an oxygen probe by itself turns out to be inadequate without a CO sensor.

I am unsure what the dominant reaction is of methane, but assume that it is quickly oxidized to CO by H2O, CO2, and O2. Thus, as CO as reduced by hydrogen and the carbon incorporated into the steel, the atmospheric CO content is maintained relatively constant by the oxidation of methane by the three most strongly oxidizing species. Maybe someone can confirm this?

Tweaking your generator to lower dew point will require less natural gas addition at the furnace. However, 35°F is right in the sweet spot for keeping soot down and generator maintenance moderate. I vaguely recall a recommendation of 20-40°F dew point at the generator as being optimal for generator maintenance. Personally, I am partial to 30-35°F.