Measuring Strain on a Holdback Bar with Piezoelectric Sensors!

Posted by Steve Hanly on May 11, 2017

hold back bar

Ever wonder how a fighter jet builds up enough thrust to clear the short runway on an aircraft carrier?

You may know that they use a steam catapult system to add thrust to the aircraft's. But a release bar is needed to hold back the jet until enough force builds up to ensure a safe launch.

Starting in 2008 as part of an SBIR project, Mide developed a health monitoring system using a piezoelectric sensor bonded to the bar.  We learned a lot as part of this effort that I'll share in this post.


What's a Holdback Bar?

You may remember it littered throughout the opening of Top Gun.  Here is a video of a couple F-18 takeoffs where you'll see the bar in action.

This bar that temporarily holds the aircraft back is not-so-creatively called a repeatable release holdback bar (RRHB).  It is a purely mechanical system (check out the patent... and check the date on that patent!) that releases a pin once the strain in the bar reaches the required level.

The Problem

The reliability and effectiveness of the RRHB is critical to successful and safe aircraft launching. There is currently no technology on the holdback bar to provide either release force confirmation or proper bar reset information. Due to the harsh environment and application of the RRHB’s they have a tendency to fail or degrade with use. This requires the Navy to rework and replace the holdback bars at pre-determined intervals or “shots”. This pre-determined shot number is for good reason low to avoid catastrophic failure and potential loss of life and equipment. If a RRHB is not operating properly before this pre-determined interval the safety of the aircraft and pilots are at risk. If the RRHB is still working properly after this interval the Navy is replacing perfectly good RRHB’s
and wasting resources.

SBIR to the Rescue

In 2008 the Navy solicited small businesses to develop a health monitoring system through the SBIR program.  This solicitation needed a solution that would benefit from Mide's core competencies and we saw the commercial opportunity if the Navy were to like the solution. It checked all the right boxes for Mide - therefore we proposed and won a contract to develop a health monitoring system.

Piezo Sensor Health Monitoring System

Mide proposed and developed a piezoelectric sensor based health monitoring system. There were two needs:

  1. Measure the strain/load in the bar - bonded piezo sensor
  2. Check that the pin reset properly - camera to determine location

Then of course the system needed all the intelligent electronics to make sense of the sensor output and communicate that to navy sailors.

Piezoelectric Sensor over Strain Gauges

First question you (and some of our internal engineers) asked is... why the heck do you need to use a piezo - wouldn't a strain gauge be better? 

The answer is no for two main points:

  • Strain gauges are not stable across a wide temperature range
    • Piezoelectrics are more stable... but we later learned there are other problems
  • Piezo sensors do not need to be powered!
    • A piezoelectric sensor can turn on the system using energy it harvests from strain.  A strain gauge based system would have to always be on or require a user to turn it on (and remember to turn it off).

Bonding the Piezo Sensor

Mide's packaging process allows us to make piezo packs that can be later bent - creating many piezo fibers.  We used such a design to bond a piezo wafer (turned into many piezo fibers) to the bar as shown in Figure 1.

piezo-sensor-bonded-to-bar.jpg
Figure 1: Poweract piezo is bonded to the holdback bar.

Piezoelectric sensors (depending on the material choice) has a very stable performance across temperature... but the epoxy we used did not (and we tried quite a few).  This required a ton of temperature compensation but we were able to make it work. Unfortunately it made strain gauges a more viable option because the temperature compensation there may have been more straightforward.

The Results

We did a ton of testing on a representative steel bar using a hydraulic test rig custom built for this program.  We had an example load profile from the Navy with a nice clean load build up to the release load and then sudden-but-manageable release/drop in load. The piezoelectric sensitivity was dialed in and the electronics were working beautifully with a charge amplifier to accurately determine the load in the bar.  And it was repeatable, even across temperature.  What could go wrong?

Then we started testing on the real holdback bar.

Turns out that when the holdback bar releases the load, there is a vicious stress wave that propagates through the bar.  Figure 2 compares the output from the load cell to what the piezo measures directly bonded to the bar.  Prior to release, the piezo sensor and charge amplifier slowly and accurately integrate the piezo output to match the load.  But when the bar releases a high amplitude and frequency stress wave propogates through the bar.

piezo-sensor-output-aircraft-carrier-takeoff.jpg
Figure 2: The load measured from a load cell (blue) is compared to the load calculated from strain in the bar using the bonded piezo sensor (orange).   We can't divulge the exact load due to ITAR restrictions.

This stress wave threw a wrench into the electronics algorithm; and occassionally the piezo would generate so much charge that it would saturate the system.  It was also difficult mechanically to keep the enclosure we had designed from breaking.  Mide was able to compensate for this stress wave but it highlighted how important it is to test on the actual system as early as possible.  In hindsight we probably tested our "representative" holdback bar for too long.

The System

Both systems ended up working well and as intended. The release load system is capable of wirelessly transmitting the shot data and both systems provide the user with immediate visual feedback as to the accuracy of the pin reset and release load. The video below shows testing of the two systems. Overall both Mide and the Navy were happy with the results of the Phase II SBIR program effort!  

What's Next?

This is an example of both the good and bad of an SBIR program.  The Navy could clearly benefit from a health monitoring system on these holdback bars. But... the health monitoring system initial cost and resulting cost savings over the life of the holdback bar wasn't enough to motivate the Navy to continue investing in the technology.  For all we know, they may have known from the start that this was more of a science project for them than a viable long term solution to a real pain point.  The manual maintenance of these bars ended up being less painful for them that we had thought or hoped.

The good for Mide though was that we had the opportunity to learn a lot more about using piezo sensors to measure highly dynamic events.  We also learned a few lessons in electronics and firmware development that helped lead to the ongoing success of our Slam Stick product line. 

So all in all it was a success.  We developed a working system that met the requirements of the Navy and we learned a lot along the way.  Unfortunately it turned out to be an SBIR that went no where directly; but indirectly it continues to pay dividends as we apply lessons learned to current programs.  And you should too!


Stay in Touch

If you'd like to hear more lessons on SBIR and other small business government grants that our company has learned, subscribe to our engineering blog.  We've planned a blog series around sharing best practices, and additional resources.  And if you're interested in partnering with us on a development opportunity in the future, please don't hesitate to reach out and contact us!  Hopefully we get to work together on an exciting new technology soon, one with a large commercialization opportunity that we're both excited about!

Link to Blog Subscription Page

Steve Hanly
Posted by:
Steve Hanly on May 11, 2017
Find me on:

Topics: Piezoelectrics, General Engineering


Comments Section