The Kitchen Table Gunsmith
Main menu  



Strain Gauges for Chamber Pressure Measurements
by Roy Seifert

Click here to purchase a CD with this and all Kitchen Table Gunsmith Articles.

Disclaimer:  This article is for entertainment only and is not to be used in lieu of a qualified gunsmith.  Please defer all firearms work to a qualified gunsmith.  Any loads mentioned in this article are my loads for my guns and have been carefully worked up using established guidelines and special tools.  The author assumes no responsibility or liability for use of these loads, or use or misuse of this article.  Please note that I am not a professional gunsmith, just a shooting enthusiast and hobbyist, as well as a tinkerer.  This article explains work that I performed to my guns without the assistance of a qualified gunsmith.  Some procedures described in this article require special tools and cannot/should not be performed without them.

Warning:  Disassembling and tinkering with your firearm may void the warranty.  I claim no responsibility for use or misuse of this article.  Again, this article is for entertainment purposes only!

Tools and firearms are the trademark/service mark or registered trademark of their respective manufacturers.

A few years ago, I purchased
the excellent Pressure Trace device available from Recreational Software, Inc. for measuring chamber pressure.  This device uses a strain gauge glued to a rifle barrel to measure the pressure.  The strain gauge is located over the mid-point of the chamber and is connected to the battery-operated device.  The device is connected to the serial or USB port of a laptop running graphing software.  The strain gauge actually measures how much the barrel stretches, which is converted to a pressure value.  I purchased this device to accurately determine when my reloads were reaching maximum pressure.  Relying on visible excessive pressure indicators like cratered or backed-out primers, case head expansion, or separated case heads, is somewhat inaccurate and usually means the loads are already beyond maximum pressure.  I have also used chamber pressure measurements to compare how different powders, or other reloading components, affect pressure and velocity in a given load.

A strain gauge is a resistive grid laid on some type of substrate material.  You install the gauge onto the rifle barrel with special super-glue or epoxy.  Regular store-bought super glue cannot be used with strain gauges as this will cause problems.  When you fire a shot, the steel stretches a small amount, which changes the resistance of the strain gauge.  This change in resistance causes a minute change in voltage, which is amplified by the device and sent to the PC where it is graphed and stored.

The manufacturer of this device sells additional strain gauges with leads and connector for about $30 each by the time you pay shipping.  You have to purchase a minimum quantity of three, which is about $90.  I thought I could beat that price to attach a strain gauge to my other rifles, so I began searching the Internet for materials.  Refer to the end of this article for a list of parts and sources.  

Strain Gauge

The manufacturer of the device states that you need a strain gauge of 350 ohms resistance capable of conducting up to 5 volts.  The gauge they use is CEA-06-125UN-350 available from the Vishay Precision Group.  I looked them up on the Internet and found they have offices in my own state.  I sent a request for a quote, and they responded with a price of $6.40 apiece with a minimum quantity of 10.  So, with tax and shipping, the price came to $7.50 each.  I also asked for a quote for the same gauge with leads already attached, and the price doubled, so I decided I could solder my own leads to the gauge.  (I believe the device manufacturer purchases the gauges with the leads already attached, which directly affects their cost.)

Connector & Wire
The device manufacturer also sells just the leads and connectors so you can attach them yourself to your own strain gauge.  However, each connector assembly costs about $5.00, which again I thought I could beat by purchasing my own wire and connectors.  Once again, thanks to the Internet, I stumbled upon Mouser Electronics.  After carefully measuring the original connectors, I found that Mouser carried the Molex series 51047 connectors I was looking for.  I purchased 10 receptacles and the male pins for them, which came to $0.55 per connector.  I also ordered 100 feet of 28-gauge wire for about $22.00.  Using only about 3Ē per gauge, the price of the wire was $0.06. 

In addition, I had to purchase some special adhesives and heat-shrink tubing, so all together, including tax and shipping, the cost was at, or a few cents under $9.00 each.  That represented a $21.00 savings, which was definitely worth the little bit of elbow-grease, wire, and solder to make my own strain gauge assembly.  (If you think $90.00 was too expensive, remember that this allowed me to attach a strain gauge to 10 of my or my friendsí guns, instead of to only three guns.)

Connector Assembly

First, I assembled the connector.  I cut two pieces of 28-gauge wire about 1 1/2 inches long.  I stripped 1/4 inch off each piece and tinned the exposed end.  I placed the tinned end into one of the male pins for the connector, crimped the end, and then soldered the wire.  I had to be sure the crimped ends were below the level of the pinís locking tab, and that I didnít leave a large solder ball, otherwise the pin would not fit into the connector.  If there were any high spots, I used a Dremel tool with a cutoff wheel to carefully grind them away.  After soldering a pin to each lead, I inserted them into the connector, making sure they clicked firmly into place.

I discovered through trial and error (and wasting a couple of connectors) that the soldered wires became brittle and broke where I soldered them to the connector pins.  To fix this problem, I put a drop of Armstrong A-12 epoxy on the back of the connector making sure I covered the wires and the connector.  Armstrong A-12 is an electrically insulated epoxy; I tried using J-B Weld, but it has iron powder in it so it is not electrically insulated.  After the epoxy dried, I protected the joint with a piece of heat-shrink tubing.  The epoxy and heat-shrink tubing act as strain relief to prevent the wires from breaking off the connector pins through normal handling.  I purchased the heat-shrink tubing from Radio Shack, and the Armstrong A-12 epoxy from Small Parts, Inc (  The epoxy cost over $28.00, but for this application, you only need a couple of small drops, so it should last a very long time.

Strain Gauge Assembly

The Vishay web site has complete instructions for attaching lead wires to the strain gauge.  I followed their instructions using the 28-gauge wire I already attached to the connector assembly.  First I degreased a flat piece of steel I use for impregnating bullets with fire-lapping compound.  Then I degreased a pair of tweezers, picked up the strain gauge being careful not to touch the grid, and laid it on the steel.  I covered it with a piece of Scotch 811 removable tape that covered approx 1/3 of the solder tabs.  I couldnít use normal tape; otherwise I could damage the gauge.  I purchased the tape at my local office supply store.

For the soldering process, I made sure the soldering iron tip was perfectly clean and tinned properly.  I used a variable temperature soldering iron and adjusted it for the lowest setting that would melt the solder.  I carefully stripped about 1/4 inch of insulation from the ends of the 28-gauge wire on the connector assembly and tinned the exposed wire.  I cut the tinned lead leaving only about 0.1Ē of tinned wire, just enough to cover the exposed solder tab on the gauge.  I put a small drop of solder on the tinned end and laid it on one of the strain gauge solder tabs.  I held it in place with another piece of removable tape.  I touched my solder to the underside of the soldering tip so there was a tiny ball of solder hanging down from the tip.  I only needed a small amount here.  I touched this tiny ball to the tinned end of the wire pressing it down on top of the solder tab for one or two seconds.  I performed the same operation to the other lead, being careful not to touch the first lead.  After soldering, I wiggled the lead to make sure the solder joint held.  If not, I repeated the soldering process.  This ďwiggleĒ test ensured I had a good solder joint, which resulted in good electrical contact.

It turns out that this is a very delicate step.  If the soldering iron is too hot or held against the lead too long, it will damage the solder tab on the gauge and the lead wonít stay in place.  If not held long enough the solder wonít stick to the tab.

In order to prevent the leads from coming off the gauge during handling I put a tiny drop of Armstrong A-12 epoxy on the leads.  I made sure that the epoxy did not flow onto the metal; otherwise, I would not be able to remove the gauge.

After the epoxy cured, I took a 1 1/4 inch piece of vinyl electrical tape and folded over one end to make a tab.  I laid the gauge on a piece of non-stick backing from a self-adhesive label, and then pressed the tape onto the gauge.  Notice in the photo how the edge of the tape covers the gauge, but not the solder tabs.  Also, this was a recycled connector as you can tell by the heat-shrink tubing between the connector and the gauge.

I put another piece of electrical tape over the gauge and the leads, then trimmed it with scissors.  I wrote the gauge factor on the opposite side of the backing and I am now ready to install it onto a rifle barrel.

Finger Flick Test
I connected my strain gauge assembly to the device, ran the program, and gently flicked the strain gauge with my finger.  To my surprise, it registered.  I flicked the gauge a couple of more times just to make sure it was working correctly.  In the above graph you can see the results.  By the way, this flick test is recommended by the device manufacturer for the gauges they sell.

Installing Onto a Barrel
The device came with all the instructions for installing the gauge onto a barrel.  Based on those instructions, I needed to decide if I wanted a temporary or permanent installation.  For a temporary installation, I use Loctite 401 available from Small Parts, Inc. (  The gauge can be removed using acetone.  For a more permanent installation, I use the Armstrong A-12 epoxy also available from Small Parts, Inc.  I can also use J-B Weld to attach the gauge as long as I donít allow the epoxy to touch any exposed wiring.  As mentioned before, J-B Weld contains iron powder so it is not electrically insulated. 

At the time I purchased the strain gauges, an engineer from Vishay contacted me with a question regarding my application for the gauges.  I explained how I wanted to measure pressure in the chamber of a rifle and asked how best to attach the gauge.  He stated, ďIn a pressure measurement on a cylinder, the gages that we quoted (single-grid, uniaxial) would be applied perpendicular to the length of the barrel, such that the gage is parallel to the circumference of the barrel.Ē  This agreed with the installation instructions from the device manufacturer.

My test rifle was my Remington 700 BDL in 6mm Remington caliber.  I decided to permanently attach the gauge to the underside of the barrel so it would be out of the way.  Using acetone, I degreased the portion of the barrel onto which I planned to mount the gauge.  I removed the gauge from the non-stick backing being careful to lift by the tab.  I applied a drop of A-12 epoxy to the strain gauge and spread it in a thin layer with a toothpick.  I then pressed it firmly to the degreased portion of the barrel pushing away from the solder tabs.  I routed the wires along the barrel and applied some additional epoxy to keep them in place.  This prevents me from straining the wires during normal handling and perhaps pulling them off the gauge.


Notice in the picture at the left the vinyl tape in front of the recoil lug, under which is the strain gauge.  Also, notice the wires, the connector, and the epoxy holding the wires against the barrel.  The small piece of electrical tape holds the wires against the barrel until the epoxy cures.  I routed a channel inside the stock so it would not come in contact with the wires.  These wires will stay out of the way just below the scope as seen in the next photo.

Live Fire Test
I took my rifle, the unit, my laptop, and some verification reloads to the range.  The reloading manual I used stated the verification loads should measure 62,500 PSI in a tight-chambered test barrel.  I would expect the loads to measure less than that in a standard chamber, but using these loads helped me to judge how close and accurate the strain gauge was.  I wasnít really shooting for accuracy; I just wanted to see how the strain gauge registered.

It is always a good idea to have a verification load.  The adhesive used to attach the strain gauge to the rifle barrel can weaken over time, which can cause a lower reading.  Therefore, when testing new loads, I should always fire a verification load first and compare it with an original verification load to ensure Iím getting the same reading.

The above graph shows the results from the live fire session.  The average pressure reading was 62,362 PSI; close enough to the expected value.  I did have to add an adjustment factor to get the correct readings.  Iím now ready to measure my reloads.

Mistakes I Made
This project took me a few years to complete.  I started this project in December, 2004 and completed it in April, 2008.  I made many mistakes along the way, but this is how we learn.  Iíve listed my painful learning mistakes below:

  • The solder tabs on the gauge are very delicate.  I ruined two gauges by having my soldering iron too hot.  I was also so nervous I didnít leave the soldering iron on the leads long enough to get the solder to stick to the tabs.  Eventually I perfected the process.
  • I didnít epoxy and heat-shrink the end of the connector where the wires were soldered to the pins.  During my initial test, the wires broke off the connector.  Of course, this happened at the range so I couldnít test anymore!
  • I got very erratic traces during my initial test.  It turned out the battery was weak in the device, and of course, I didnít have a spare battery with me!  Again, this happened at the range.  The device has a standby mode that uses a very small amount of battery power.  After a short time, the battery goes dead.  I installed a switch in the unit to completely disconnect the battery when Iím not using the unit.  This saves battery power.  I also bring a fresh 9V battery with me just in case.
  • I used J-B Weld epoxy the first time I sealed a connector and the soldered leads on a gauge and got no readings when I did the ďfinger flickĒ test.  J-B Weld has iron powder in it, which means it is not electrically insulated.  It caused a short, so I got no readings.  Scratch another strain gauge assembly!
  • I attached the first strain gauge to my rifle using Loctite 401.  Over time, this affected the gauge so I didnít get any readings.  I destroyed the gauge removing it from the barrel.  I attached another gauge and decided to make it permanent so I used epoxy instead.
  • I didnít ďflickĒ test my work, so when I went to the range to test my first gauge and installation, I got no readings.  Thereís not much you can do at the range, so I packed up the rifle; a wasted range session.  (Not really, I always bring other guns to shoot.  I was, however, disappointed that I didnít get any pressure readings from my rifle.)
  • During the initial ďflickĒ tests, I had the sensitivity of the unit set too low so it never registered.  I had to set the sensitivity high enough to register.  This may also have been part of the problem why it didnít register at the range, but now if itís sensitive enough to register a flick, I can adjust the sensitivity to register a live shot.
  • I installed the first two gauges in parallel with the chamber, rather than perpendicular.  This caused erratic or no readings.  Of course, I destroyed the gauges when I removed them from the barrel.  I installed the gage in parallel with the barrel because the gauges initially sent with the device were installed this way, and I thought I could install the new gauges the same way.

So now that I can measure chamber pressure, how do I use this information?  I have a pet .45 LC hog load of 20 grains of 2400 under a 300 grain hard cast wide flat nose gas check bullet.  This load chronographs at 1527 feet per second (FPS) from my 24-inch Marlin 1894 Cowboy.  Using a strain gauge, the pressure for this load measured 32,454 PSI.  Needless to say, this is above SAAMI maximum for the .45 LC.  Please remember, this is my load for my gun, and must be approached with caution.

At a local gun show I purchased 8 pounds of Accurate Arms (AA) #9 powder.  This powder can be used for the same loads as Alliant 2400.  In fact, they are right next to each other on the burn rate chart.  But, just how interchangeable are these two powders.  So, I put some loads with 20 grains of AA #9 through my Marlin 1894 Cowboy to compare the measurements.

The data for 20 grains of AA #9 showed that the pressure went up an average of almost 2,000 PSI, but the velocity only went up an average of 14 FPS.  I would probably drop this load about 2-3 tenths of a grain.  In fact, maybe it would be better to measure by volume rather than weight to get equal performance.  However, without the strain gauge I would not be able to make this comparison.

Parts Sources (prices as of 4/1/2008 without taxes or shipping)


Part No.




Vishay Precision Group


Original strain gauge used by manufacturer



Mouser Electronics


Plug Housing



Mouser Electronics


Pins (2)



Mouser Electronics


28ga stranded wire, 100í



Small Parts, Inc


Loctite 401



Small Parts, Inc


Armstrong A-12 2-part epoxy



Radio Shack


Heat Shrink tubing



Office Max


Scotch 811 removable tape



Strain Gauge Assembly Price




   © Copyright 2008-2012 Roy Seifert.