Back to
Air Command Rockets


Find Data

Pressure Chambers




Test Procedures

Test Equipment



psi to bar

mm to inches

feet to meters

grams to ounces

Request Experiment

Current Requests

Submit Experimental Data



About Labs

071024-001 Hybrid splice burst tests
Date 24th October 2007
Tested by PK and GK.
Location Sydney, Australia
Test Procedure Burst Test - Splice


  • To test at what pressure the hybrid splice will fail at.

Experiment Setup

The standard hydrostatic burst test procedure was used to carry out these tests. (See above)

Hybrid Splice Background

See the following link for discussion on the Hybrid splice concept:

The total cross-sectional area of all the holes was equivalent to a 32mm Robinson coupling. That is about 15 times bigger than our typical 8mm coupling.

We spliced two 1.25L bottles and let it dry for about 4 days and then performed a hydrostatic burst test on it.


The splice held up to 130 psi. The sleeve was held down by a combination of PL Premium and VISE glues. After we put the PL glue in we noticed there were a couple of minor leaks, so we poured the runny VISE glue in to fill those, and that sealed it well.

Hybrid splice components.
Detail of the location of the holes in the base of each bottle.
An inside view of the same holes.
Bottle bases screwed together using a single bolt through the middle.
sleeve and bottles sanded ready for gluing.
Glued and cured splice 4 days later ready for test.
sleeve cut away showing where the splice failure occurred. Note the cracks between the holes.
Half of the cut away sleeve. You can see where it ripped all the way around.

It is unclear what the initial failure point was but both bottles cracked between the holes in the bases and the sleeve also ripped all the way around. It did not delaminate from the glue, the plastic failed. After a number of discussions with forum members we now suspect the failure was at the sleeve first and when that failed the bottles did. Normally a sleeve like that should hold at least 180-190psi. The bottles still remained together and no shrapnel went flying. It looks like the holes in the sides have weakened the bases too much.

It was a really unusual failure because the sleeve edges are still attached to the bottles all the way around. The straight edge seen on the torn sleeve photos is from the scissors when I cut the sleeve away to photograph the inside.

Conclusions / Analysis

If the sleeve failed first that means that the bolt wasn't doing a good job of holding the bottles together. When we first bolted the bottles together we noticed that there was a certain amount of give. The bottom of the bottles flexed a little when you pulled on the two ends. This was seen before gluing the sleeve on. With the one bolt, this flex in both bases was probably enough to put most of the strain on the sleeve when pressurised. Because the pressure was the same on both sides of the lobes you didn't get the typical crack propagating from the central bolt hole but rather the circumferential cracks between the weakest points.

130psi is not all that great for this particular hybrid design when you consider the VISE glue-only splice held 170psi+ and our Robinson couplings hold also around 170psi+ with bottle burst pressures around 190 psi. A 130psi splice means about 100psi operational pressure.

The reason the sleeve may have failed at a lower pressure is that because the sleeve is only really held down by the ends. The middle could have bulged out under pressure, and placed uneven strain on it. With a normal splice the sleeve is completely held down along its full length by glue and so this bulging is unlikely to happen.

The cross-sectional area of the bottles is 63.6 cm2 and at 130 psi you end up with a force of 581Kg pulling one bottle in one direction and the same in the other direction. No wonder you get a bit of flex in the base of the bottle.

As a result it may be better to try the Robinson coupling Pat suggested in the first place for the hybrid splice. Although the coupling may experience the same flex at the base of the bottle, putting the strain on the sleeve again. It may achieve better results since the bottles are not weakened by the holes in the lobes.

Copyright 2006-2010 Air Command Water Rockets

Total page hits since 1 Aug 2006: