Oct 07 2009
ENLIGHT Update
A year ago I introduced ENLIGHT to you as it was first released. ENLIGHT provides an all-in-one software solution to configuring sensors connected to Micron Optics instruments, converting wavelengths to engineering units for hundreds or thousands of sensors, displaying data in charts, graphs or images, setting alarm limits and sending alerts, and saving data. Now I’m back with an update on the features of our next release which is coming soon.
In the past year we have worked to bring the software from an early Beta state to an optimized, faster running tool that’s ready for broad deployment. In response to customer feedback, our major focus has been on data management. For example, systems sampling hundreds of sensors at 1 kHz produce large amounts of data in a short time. New ENLIGHT features will allow files to be managed by size (e.g., once a file reaches 10MB, create a new data file), or by time (e.g., create a new file every day or hour). All files are stored in an automatically created directory structure organized by year, month, and day. Users will be able to choose what data to save, i.e., full spectrum traces, peak locations, FBG wavelengths or the calculated sensor values in engineering units.
Users will also be able to choose when data is saved. Data saving can be either continuous or triggered by an event such as a sensor moving into a warning or alarm condition. Users will select which sensors trigger data saving and then what data shall be saved. Data buffered prior to the event can also be included in the event file.
There will be more flexibility in manipulating data as well. Separate averaging controls will be available for spectra, peaks, FBGs, and sensor values. Another new feature will allow ENLIGHT to calculate the derivative of a sensor value so that the rate of change of a sensor value is known which is useful in fire detection applications.
Error handling will become more robust. Instrument diagnostic tools built into ENLIGHT will allow it to automatically recover to its normal operating state following a power interruption. Additionally, multiple copies of ENLIGHT will be able to run simultaneously on a single processor, thus managing more than one interrogator from a central PC.
For applications where custom user interfaces are required, users will be able to send commands to ENLIGHT via Ethernet to stream processed sensor data, zero sensor values, or retrieve saved data. In such cases, ENLIGHT acts as a signal conditioner simplifying integration of optical data with data streams from other systems.
The bottom line is that optical sensors are continuing to become easier to use. Better software, better instruments and better sensor packages are making this possible. Soon we’ll have more news on new sensor developments at Micron Optics and elsewhere.
A free download of ENLIGHT is available at http://micronoptics.com/sensing_software.php
5 Responses to “ENLIGHT Update”
what happen about this “Error handling will become more robust.”
i m unable to get this exactly…????
and i will wait for new sensor devlpmnt.
thank you.
For example, when you enter an expression that translates wavelength to strain, it’s calculated in terms of FBG names and perhaps even other sensor names. ENLIGHT keeps track of all these variable names so that none are duplicated. If you try to use a variable name twice, the “error is handled” by flagging the input and providing a message that describes the error.
There are several data entry fields that have been improved in this way…
Dear Sir,
I am keen to use some of my research work at NAL, Bangalore and at IST, Lisbon on the algorithms side to modify the software or the firmware provided by Micron Optics for better internal peak detection as I have observed phenomena like
a) peak splitting
b) peak vanishing
with usage of FBG sensors during our experimental tests. How can I incorporate the same and the best way of doing it? Please guide me as I currently possess 1Hz Micron Optics system.
regards,
Amol
It sounds like you’re doing interesting work. If you wish to refine peak detection algorithms for special cases, like split peaks, I think it’s best if you bypass the internal peak detection and process the entire spectrum in your application. Micron Optics’ LabVIEW example (or National Instruments’ sm125 LabVIEW driver) provides an easy means of pulling spectrum to your processor.
Of course it’s even better to avoid split peaks. Micron Optics’ sensor packages are designed to isolate FBGs from the transverse loads that can cause split peaks. Accordingly, the sm125′s internal peak detection tools operates with the assumption that peaks are Gaussian, about 250pm wide, and have a side mode suppression of >12dB. It is flexible enough to handle some exceptions to this ideal, but split peaks are notoriously difficult to characterize and often change dramatically in shape as the load conditions around the FBG change.
Dear Tom,
As you rightly point out, I should work only with the wavelength data. However, my problem is I want to work with as low a size of data as possible so right now, I am looking at taking peak data.
Also, my loading conditions do not change that much and are quasistatic in nature,so the best bet would be to store the last instant wavelength data alongwith the current, wherein incase of peak loss that could be investigated in detail.
I will keep it posted as further work progresses.