Flame Detector for Hydrogen Fires

  • Thread starter Waddell, Reginald E
  • Start date

Thread Starter

Waddell, Reginald E

I have an application that requires flame detection for hydrogen fires. Does anyone know of the type of technology that would work well in this application (UV/IR) and what vendors manufacture this type of equipment.

Reginald Waddell, PE
Process Control Engineer
BP Amoco Chemicals
843.881.5153 (voice)
843.881.5175 (fax)
[email protected]
Responding to R. Waddell's query:

You didn't include application. If related to pipe-flange leak (and subsequent hydrogen flame detection) then contact BP-Amoco Instrument Dept at Whiting, IN Refinery. Or Bill Mostia on the List.


Phil Corso, PE
Trip-A-Larm Corp
Deerfield Beach, FL

McConnell, David P

The facility in which I work tests large liquid fueled rocket engines with LOX and LH2 as the most common fuels. Over the years we have found that the most reliable hydrogen fire detect system is one made by Kidde. It consists of a coaxially constructed "wire". Heat causes a drop in the resistance between inner and outer conductor (in fact it usually destroys it!).

This "wire" is routed, on support brackets, around flanges, valve domes and other leak prone locations. There have been many fires over the years and to my knowledge the system has never failed. It does require ongoing maintenance to keep water out of the connections and prevent false alarms. If you are into esoterics, it is also ugly and unkempt looking, but it works!

For us the optical sensors don't work since we have some very large highly luminous exhaust plumes and the occasional transient free air combustion as engines start and shut down.

If you want, email me direct and I will put you in touch with the guys who engineer and maintain it.

Regards, Dave McConnell

Preston Todd Johnson

There are two companies I have reviewed for IR imaging. Their cameras have come down quite a bit in price in the last few years. These companies are FLIR, and Indigo Systems. Both work very well with LabVIEW software which can analyze the data and serve the results via OPC or Modbus to your SCADA network or DCS.

Here are some websites to review:

These are the ones I am most familiar with.

There are others out there from Honeywell, Fisher/Emerson, etc.


Todd Johnson
Interspace Electronics
Responding to Dave McConnell's 14-Jul reply:

If anyone wants to know how BP-Amoco resolved the problem of false alarms associated with the KIDDE "wire-based" H2 flame detection system, contact me directly.

Phil Corso, PE
Trip-A-Larm Corp
Deerfield Beach, FL
Dear Reginald, The flame detector that I have in mind is an optical instrument. Hydrogen flames (1 to 3 ft high) can be detected with UV detector in the 180-260 nm range. The downside is that UV detectors false alarm to arc welding and get quickly contaminated. IR is less suitable since they usually detect radiation in the 4.4 micron range which is the CO2 emission in a Hydrocarbon fire. Burning Hydrogen does not emit CO2 so cannot be detected bt most UV/IR, IR/IR or IR/IR/IR detectors. Fire Sentry in Brea CA manufactures the SearchFlame 22 detector for Zellweger Analytics. The unit is based on Visual, Near IR, Wide band IR and 4.4 microns. A microprocessor analyses the sensor in put and is also able to detect non-hydrocarbon fires like Hydrogen. Please contact me if you need more information. Jan Nijkamp Zellweger Analytics UK [email protected]

Brett Boudreaux

I encourage you to contact Buff Crosley with Detector Electronics in Minneapolis. The new Det-Tronics X3302 multispectrum Hydrogen flame detector has been tested and approved for government and other installations as Best in Class.