Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Nichrome wire overheating failure in standard air heater

  1. Jan 12, 2016 #1
    Am experimenting with an Omega Engineering AHP 7561 standard air heater rated for up to 240v. Specs include ability to keep a 20 cfm flow exiting @1000F while pulling about 10A @220v.

    Nichrome melts around 2400F. I'd like to produce small bursts (lasting a few seconds) of very hot air, like 0.03 cu.ft./sec. (about a liter/sec.), raising its temperature from ambient to 1500F.
    If I energize the heater with essentially no air flow, it takes about 30 secs to bring the wire from room temp to cherry red (roughly 1400F). At that point several liters of air will exit the heater at about 1000F, rising from ambient to that temperature in a few seconds.

    How long can I safely heat the nichrome under these conditions before melting it and opening the circuit? I do not want to disassemble the system to replace the air heater so playing on the safe side makes sense; still, I need to raise output temps.

    What do you think? Any ideas about a safe schedule for longer times, gradually creeping up on the output temps required? By how much will the service life of the heater be reduced under this kind of regime? How valid an estimate of nichrome coil temp can be made from eyeballing its color?
  2. jcsd
  3. Jan 12, 2016 #2


    User Avatar
    Science Advisor
    Gold Member

    What I have placed in bold seems to be contradictory. My opinion would be that doing something like this may only be good for a couple of times. The heater is not designed for this type of use repeatedly. All the manufacturer is concerned with is safety so if the heater safely destroys itself so the further use is prevented that is probably good enough for the manufacturer. The nichrome wire is not the only thing being stressed in this type of operation.
  4. Jan 13, 2016 #3


    User Avatar
    Gold Member

    I don't know how to determine the effects of the repeated stress of overheating the nichrome wire. You need to talk to the manufacturer.

    A safe way is constant flow and diverting the output. But I don't know what solenoid valve handles 1000F air well.
  5. Jan 13, 2016 #4
    Thanks for the thoughtful responses. Indeed there is a tension between getting the system function required and not breaking the equipment too quickly (it's a test rig), and that prompted the posting. I'm hoping someone has insight into general nichrome wire failure due to overheating but, as suggested, there are other potential failure modes e.g., thermal stress cracking of the ceramic core carrying the nichrome coils.
    BTW, the air control valve is on the cold side of the heater.
  6. Jan 13, 2016 #5


    User Avatar
    Gold Member

    I was saying that a safe way to operate is to divert the hot output air when you are not using it, rather than turning off the input. That way the heater is always getting air flow and running in spec, and the response (hot air availability) is instantaneous. You can do it with a simple baffle arrangement at the output.
  7. Jan 13, 2016 #6
    This seems like a good idea. Will take some rework of the test rig - things are pretty close-coupled. Thanks!
  8. Jan 14, 2016 #7
    It seems like you need hotter air. You might consider using baffles and air from a ceramics kiln?
  9. Jan 14, 2016 #8
    Thanks. Sounds like the right temperatures but a minimum of portability is required.
  10. Jan 14, 2016 #9
    Benchtop kilns exist for the metal (jewelry) working. I don't know if they get hot enough though.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook