This blog covers the day to day progress of water rocket development by the Air Command Water Rockets team. It is also a facility for people to provide feedback and ask questions.

Saturday, December 26, 2009

Pressure switch testing

I've posted the results of the pressure switch tests here:

The switch seems to be working well so now it's onto finishing the rest of the spliced pairs and getting Acceleron V re-assembled.


Monday, December 21, 2009

Acceleron V pressure switch

This weekend dad and I worked on the Acceleron V pressure switch. On Acceleron's last flight it did not look like the pressure switch(TDD2) worked properly, and the pressure switch(TDD) we were using before on the earlier Accelerons had been somewhat leaky.

The pressure switch is used by the primary flight computer to initiate the staging of the second stage when the booster stops producing thrust.

So we set out to try to make a more reliable one. (I think this is our 5th attempt at a simple and reliable pressure switch). In the photo below is the prototype. The hose clamps are only temporary for the first pressure tests. Also the cut off bottle neck at the end of the switch is only temporary to support the clamp.

We pressure tested the activation pressure and it was around 30 psi. That means that the switch will activate when pressure falls below 30 psi. This is a good value because the flight computer can delay the staging a little bit until the pressure is around zero.

We ran several low pressure cycles to test how well it activates. And the results were always repeatable. We are now gluing it together properly, removing the hose clamps, and when the glue has dried, we will test the switch to full operational pressure to make sure it can still hold up at the higher pressures.

I would also like to set up one of the booster segments on the thrust stand and simulate a launch with the correct amount of water and pressure. The segment will be fitted with the pressure switch so that we can see when the switch activates in relation to the thrust curve. This should allow us to then calibrate the small staging delay for the primary flight computer.

Once the pressure switch is fully tested, I'll do a full write up of how it works on the main website, but it is simple enough for most people to build.

Acceleron V has also been completely stripped down so we can start replacing the bottles with the new spliced pairs.

Deployment Mechanism development

We have also been working on a new deployment mechanism that uses in-line parachute deployment. The main criteria for this is to reduce the weight. Currently the standard nosecone and side-deployment mechanism we are using on a 90mm bottle weighs around 138 grams without the parachute, but it does include all the mechanical hardware, electronics and battery.

The new nosecone and deployment mechanism currently weigh 37 grams (including a servo motor) with the electronics and battery still to be added. The new electronics (FC V1.7) and battery will add about 20 more grams, so all together the entire nosecone should be around 60grams. The design also allows for potentially large parachutes and is also around 30% shorter than our existing designs.

When we've had a chance to test fly it a number of times, I'll again post full details.

Thursday, December 17, 2009

Day 83 Update

I've posted the flight day report from Doonside on our main website here:

It includes a highlights video and photos from the day as well.

I've also updated the flight day report with findings from last night when I was pulling the rockets apart to give them a clean. (see the Update section) It turns out we had some heat distortion issues that we didn't know about on the day. The Polaron VII rocket also flew with a missing o-ring in one of the Tornado couplings!


Sunday, December 13, 2009

Flight testing new splices

Well it's been quite a while since the last update, but rocket development has been continuing although a little slower due to the holiday season coming up. Today we went out to Doonside to test launch the new spliced pairs to see how they behave under flight and landing conditions, as well as sitting in the sun for a while.

The rocket used all 4 spliced pairs previously tested to 140 psi. The rocket was launched a couple of times at 130psi, and the splices held out well without any significant issues. 5 more of these will now be made and fitted to Acceleron V.

I am doing a write up of the whole day along with a few other foam and flour flights we did on the day. I'll post photos and a highlights video in the next few days.


Monday, November 16, 2009

Keeping neighbours happy

This weekend we were going to do burst pressure tests on a range of bottles, but blowing up bottles in one's back yard isn't something to get your neighbours excited about. We are always looking for ways to lower the noise during these tests. Having the bottle full of water greatly reduces the noise, but a little bit of air can enter the bottle as it expands and so the resultant explosion can still be very loud.

Sound suppression attempt #1:
We filled the bottle with water, and then placed the whole bottle in a small plastic bathtub, filled it full of water, covered the bottle underwater with a towel and held it down with two bricks. ....
Result: Big boom, tub split in half, lawn received a healthy watering.

Sound suppression attempt #2:
So we drove over to dad's house, filled the next bottle with water and suspended it about 1/2m below the water in the concrete in-ground swimming pool.
Result: Quieter boom but what was amazing was the fact you could feel the concrete shake under your feet with the boom! This was a bottle full of water with about 100mL of air in a large pool and it burst at ~160psi. The slow-mo video standing on a tripod on the side of the pool shows the blur as the shockwave hits shortly later followed by tiny ripples on the surface doing interference patterns for about 1-2 seconds. Obviously the shockwave travelled through the water and echoed around the pool. .... WAY COOOL! ... but not the objective. Could the shock wave damage the pool? Unlikely, but we didn't want to take too many more chances.Tiny ripples on the water are the sound waves bouncing around in the pool

Sound suppression attempt #3:
Dad brought out an old SCUBA cylinder with the bottom cut off. This aluminium tank has a wall thickness of about 1cm, and so we place the bottle in it, stuffed old rags in the hole, placed it up against a rock wall, put a bench, lead weights and seat against it to stop the tank from flying in the other direction, wrapped the tank in two layers of foam and burst the bottle ....
Result: A barely audible thud.
Subsequent tests showed that you really only needed the rags in the end of the tank to suppress pretty much all the noise. There was no need for the foam wrap, or weights to hold the tank down.

Putting the bottle in the SCUBA cylinder

Stuffing in rags to suppress the sound.

Tank is under two layers of foam and bench with weights is meant to stop the tank from flying away.

In the end 9 bottles were blown up, and the neighbours didn't complain one bit.

Wednesday, November 11, 2009

So far so good ....

During the week we made up four new spliced-pairs of 2L bottles. Each has a capacity of around 3.15L. The splicing and reinforcing technique seems to be working well so far. The Sikaflex is doing its job sealing the splices.

Today they were all pressure tested to 140psi without any leaks, and the bottles and tape showed no stress marks. We are very happy about that. Having a good yield rate means we can produce more of these with less bottles.

Now they will all be assembled into a Polaron type rocket. The initial target launch pressure will be 130psi with a 15mm nozzle. While we are building this rocket we will continue to make up more spliced pairs for the rest of Acceleron. Some of these will also be converted into boosters.

I think I have an approach too now for the pressure switch, so I'll be doing some prototype work on that soon.

Friday, November 06, 2009

Investigating water behaviour in flight

At Doonside this weekend we flew the Axion rocket again 3 times with a camera on the boom to try and get a better shot of what is happening in the lowest bottle in flight. We used food colouring to enhance the contrast and no foaming agent this time. We also painted the fins a neutral gray to prevent the auto-iris on the camera making everything else too dark.

The full launch report is available here: http://home. au/~aircommand/ day82.htm

Acceleron V

Since a few people have now asked about when Acceleron V is going to fly again, I might as well do an update. We are in the process of getting Acceleron V back in the air. There were a couple of issues that we really wanted to solve before trying again. The number one priority is getting the spliced pairs reliable enough to be able to launch with them and not worrying about being very close to the burst limit. The last explosion showed that the spliced pairs just weren't there yet. Even though they were tested to 130psi, they were obviously close to that pressure. Although we could fly the rocket at lower pressure, the rocket's flight envelope was really not designed for anything lower than 110psi. Ideally the rocket needs 130-140psi.

The need for stronger spliced pairs is the reason we have been doing the splice tests recently. If further tests with the new splice technique continue to go well, we are hoping to replace the existing spliced pairs with the new ones on the rocket. We have 4 new spliced pairs curing at the moment and we will hydro test them to 140psi. We will then make a Polaron style rocket and test fly them at 130psi. If the spliced pairs perform well in real life situations then we will make up more to replace those on Acceleron V.

The other issue is the pressure switch. We want to have a reliable pressure switch to detect burnout. The pressure switch we used on the first flight just wasn't reliable enough, and the old pressure switch (TDD) leaks a little. When these issues are resolved, then we will try to put Acceleron V up again.


Thursday, October 29, 2009

Splice test #3 - Sikaflex - Success

We did a successful burst test on the Sikaflex asymmetric splice with a PL attached sleeve. (splice #5) The bottle ends were also reinforced with a jacket made from another PET bottle and held down with glass strapping tape. We learned a couple of things from this test:

The Sikaflex works great as a sealant and the width of the sleeve was more than sufficient to hold the splice together.

The final burst pressure was 190 psi! This is a better result than was expected. The actual failure was due to the strapping tape breaking, weakening the bottle reinforcing and then ultimately bursting the bottle. The sleeve still held though. The normal burst pressure of these bottles is about 165 psi.

This is a photo of the splice under test. The needle on top of the bottle was added to see how much bottle distorts close to the sleeve.

I'll make up three of these splices next and see how reproduceable the results are. I won't be testing them to destruction but pushing them to 140 psi, for an operational pressure of 130psi in actual rockets. If the results can be replicated, it would give us a nice safety margin.

I do have good slow motion video of the bottle's distortion and ultimate burst. I'll post the video with the next main website update, as I am busy the next couple of nights getting ready for this weekend's launch event.

Sunday, October 25, 2009

Splice test #2 - Epoxy

After letting the epoxy splice sit for a week we pressure tested the spliced-pair to destruction today. At around 120 psi a small leak developed. We kept increasing the pressure until the splice failed at ~150psi.

This time it was the PET bottle that failed rather than the splice. It fairly cleanly tore itself off from around the edge of the sleeve. The epoxy glue though separated cleanly from the bottle that flew off. This means that the epoxy did not do as good a job of holding onto the PET, even though the 1cm x 1cm tests showed promising results.

The sleeve was glued on with PL premium.

We are now waiting for the sikaflex and PL to cure on the next splice test.

The reinforcing shells worked well and the bottles did not show any signs of stress in the neck area. We have taken them off again, and will use them on the next splice test.


Friday, October 23, 2009

Bits and pieces

We've updated the main website with some of the things we've been working on over the last month or so:

The update includes discussion of the splice variants we are currently testing as well as details of my involvement in the rocket challenge on Channel 7's 'Beauty and the Geek - Australia'.


Monday, October 19, 2009

Splice test 1

I tested the splice yesterday, but it sprung a leak at 130 psi, which was disappointing, but not unexpected for the first attempt. It was only 4 days since I did the splice so it may not have been quite long enough for a full cure of the splice. Except for the leak the splice held well though, and there were no stress marks on the bottles. All up the reinforced double-walled spliced pair was 3.15L and weighed 139 grams.

The reinforcing shells used on both ends of the rocket can be easily removed and reused for the next test since they are not glued, saving a couple of bottles there. They protected the inner bottles well during the test.

Having recently performed the 1cm x 1cm shear tests on the Araldite super strength epoxy, I've made a new splice using this glue last night, and it looks like it has a much better seal. It does not bubble up like the PL. I used very gentle heat from the heat gun to make it flow better into the corners once the splice was assembled. Tonight I'll add the extra outer sleeve and glue it on with PL.

If the epoxy fails, next I'll try the sikaflex on the inner splice for a better seal, and then again use PL for the outer sleeve.


Wednesday, October 14, 2009

New Splices

We have been using the same symmetrical splice technique on our 2L bottles for a couple of years now, but I have never quite been happy with it. We have had quite a few failures along the way, and the no-leak yield hasn't been as high as we would like. This meant a lot of wasted bottles and wasted time. The other issue has been the relatively low pressures the spliced pairs could hold. 130psi has been the limit we could practically push them to.

We used to use 2.25L bottles as the sleeve since they fit well over the 2L bottles and no shrinking was required. But recently we have found that some of the 2.25L bottles don't fit as well as they used to so it has been difficult to use them as well. To try to stop the leaks we have been using a different glue to do the sealing in the joint and PL to hold the splice together but this made the process more complex and we still ended up with leaks although less often.

So we are currently trialing a different technique that will hopefully yield better results. The technique is not new but a combination of several different techniques used together. We are using just PL glue for this splice and only the one type of bottle. I'll post the full technique on the main site, but here is a quick run down:

We first heat shrink about 2-3 cm of one of the bottles using hot water so that it fits inside another bottle of the same type. Then curl the edge on a pan. This has the effect of giving a tight seal where the heat shrink section ends and meets the other bottle since they are the same diameter at this point. We sand and glue these sections together. We then make up a sleeve about 4 cm wide made out of the same type of bottle. Because the diameter is the same we simply cut the sleeve to turn it into a strip. We then use another section of sleeve about 4 x 4cm and glue half of it to the sleeve strip to cover the gap. We let the glue dry for a couple of days. Then we glue the sleeve strip over the splice to provide further strength to the splice.

The drawback is that inserting one bottle into the other means that the volume is smaller, but not by too much.

We have had to previously reinforce the necks of the bottles with rings made from other bottles, but it only strengthened the necks not the rest of the bottle. This time we are going to use Richard Wayman's bottle-on-bottle technique to not only strengthen the neck but a significant portion of the rest of the bottle.

Hopefully this technique will yield better sealing results and will be able to take higher pressures.

The glue is curing at the moment so tests will be carried out at the next opportunity.


Wednesday, October 07, 2009

It's been a while since the last update because there has only been little rocket related development in the last two weeks.

Crestwood Fair and other stuff

Last Sunday I spent the whole day helping to man NSW rocketry association's stall at the Crestwood fair. David and Neville did a great job of organizing the stall and getting all the display materials together. We had all sorts of rockets on display, and I even brought along a couple of water rockets including the Acceleron V to sit in the corner. It was a fun day talking with people about rockets. Quite a few interested people stopped by for a chat. We gave out close to 50 fliers on the day, so we'll see how many people turn up to the next launch at Doonside.

Over the last two weeks there were also several birthday parties, accountants to see, kid's school concerts, and kids cricket games which meant a lot of time away from the rocket workshop. Now that the dust is settling down, we'll be back to more regular development. Dad is also back from Europe so we'll have more time to spent on developing the next set of rockets.

Work has also been continuing on the website extension. There is just a lot of typing to do, and it's taking a lot longer than I originally expected. There are around 70 pages on there so far.

New workshop
We had a long weekend this week, but instead of rockets I decided to concentrate my efforts on making significant progress on the new workshop under the house. The current workshop is just getting too small. Dad and I spent all three days putting up the internal walls, which I'm happy about, because those are done now and the ceiling is next. We had the electricians come in earlier in the week to put in all the power points and lighting, and we'll need them to come back to finish everything off after we have painted. There is probably 3 weeks or so more work on it, because we need to put in new doors, do the little trim bits to finish it off, and then fit it out with benches and shelving.

Coming up
Last week we've also made a booster for Paul's pyro rocket, so we are hoping to fly it this upcoming weekend if the weather improves. It will be our very first two stage pyro rocket launch.

We are bringing a couple of water rockets with us as well of course. We will fly the rocket with the camera mounted on the boom again to see if we can get a clearer view of the water behaviour in the rocket during flight. Acceleron V is currently slated to fly at the end of the month's NSWRA launch.

Wednesday, September 16, 2009

A different perspective

We had a good flight day on Sunday at Doonside. Although the wind was fairly strong we managed to get 3 flights in.

The first flight was that of a modified rocket giving us a different perspective of the rocket. We were using the technique to investigate what happens inside the lowest bottle during flight. We mounted the camera on a long boom sticking out the side and looking back at the rocket and nozzle.

The full writeup with pictures and video of the flight is here:

I want to fly this one again, but actually remember to put the food colouring in it this time for better contrast enhancement. The fins also need to be painted a darker colour to work better with the camera's auto-iris.

Tuesday, September 08, 2009

Lots of progress, little time to write about it

Here is a quick update as there has been lots of stuff happening in the background but have had little time to do proper write ups on the main website.

Flight Computers
Flight computers V1.7 and 2.0 are coming along well. We are developing two different ones at the same time with different capabilities. V1.7 is aimed at providing a very simple, inexpensive and small solution to parachute deployment. Focus is on least number of components in a lightweight package coupled with a single 3.7g servo and a small 6V battery.

V2.0 is a more advanced version offering expanded functionality over V1.6, with extension modules that are also suitable for pyro rockets. V2.0 also offers a much more flexible programing model capable of sequencing complex operations.

One of the main design criteria for these flight computers is that they can be fully configured in the field without any special external equipment like keypads, displays or a laptop. Both are now using surface mount devices for smaller footprints.

Website Extension
A lot of writeup time has been spent doing work on the website extension. This is mostly why there has been fewer updates on the main site. Currently the plan is to go live with the extension at the end of September. Some of the recent write ups will be available with the extension.

Rocket development
Acceleron V is currently resting until dad gets back from Europe. I've gone and given it a bit of a service clean such as the non-return valve that was showing signs of corrosion, and generally made sure that everything was in order. We need to take two cars full of equipment to the launch site and it takes both of us to set it up and launch it, so it will wait until he gets back.

The new rocket I want to fly this weekend (Sunday) is also close to ready. I'm using the Acceleron launcher, (without boosters) as it already has provisions for the 2 meter guide rail, as the rocket design does not allow us to use our regular launcher.

Other projects
I've also been helping out with prototyping work for a paid rocket project which I can't post about yet, but it has been quite interesting and something different. This has also taken up quite a bit of my time over the last two weeks.

On top of that I am finally working again on the new bigger workshop under the house. I'm hoping the electrician will be able to come in this week for a quote for all the wiring, and then we can put up the wall panels and then it should start looking like a real workshop. :)


Sunday, August 30, 2009

How to make a nozzle tutorial

This weekend we've put together a couple of short video tutorials on how to make a 9mm Gardena nozzle. One of the issues that people have been reporting is that they are unable to get a hold of the older bottle caps with removable seals. The video covers an alternate source of caps that are suitable for making the nozzles:

The second video shows how to resolve an issue with some launchers as not all Gardena (and clones) quick connect fittings are created equal. The problem reported by quite a few people is that the nozzle does not lock all the way down into the launcher. The nozzle appears to be locked in position but releases itself as pressure builds. The solution is fairly simple:

We did not attend the launch at Doonside this week because the weather predictions indicated strong wind and rain. When I woke up in the morning I saw that it was pouring, and decided to skip the hour long drive to the launch site. A number of people attended the launch but said the weather was less than ideal.

The good news is that NSWRA now has permission to launch 2 times a month at Doonside. So the wait time for doing bigger test flights is reduced.

Thursday, August 20, 2009

Acceleron Photos from 4th July

A big thanks goes to Andrew from NSWRA for shooting the following photos of the Acceleron V booster in flight. We didn't get any still photos of the launch on the day as we were busy shooting video. (click on the images for a higher res version)

In the first one you can see the turbulent flow from the red and blue nozzles, whereas the green nozzle has very nice laminar flow (seen further down the plume).

The second photo is my favorite one with the air pulse starting and the sun reflecting from the sustainer.

The last one shows the rocket shortly after staging occurred. The black fins of the sustainer are about to clear the nose of the booster.

I'll post some of the other photos of the flight in next main page update.


Monday, August 17, 2009

Development continues

It's been a while since the last update so I thought I'd post about what we've been up to. In the last few weeks I've been catching up mostly on non-rocket related matters, but development has been progressing.

Acceleron Vc
The rocket is now mostly repaired. The broken ring brace was easily repaired with a bit of epoxy glue, and a new coat of paint. New separator rings have also been made and are currently having the servo and camera fitted. The only thing that needs to be done is for the whole rocket to be pressure tested again and re-assembled.

We are also making up a whole bunch of new spliced pairs of 2L bottles (we are processing about 40 bottles) to have spare ones ready in case some of the original pairs fail during testing. These will also be used for the Polaron rockets and new bigger boosters.

New rocket
We have also been building a new smaller rocket (not named yet) to try to better understand how foam and foam generation behaves during flight. This rocket was built small so that we can fly it at our local park. I'll post full details in the next main website update.

I've also been working on an extension to our website that hopefully will go live in the next month or two because there is still lots of information to enter and format.

Tuesday, July 28, 2009

Day 79 Flight day report

We've updated our main website with the latest launch day report here:

Included are pictures from the day and a couple of videos. We also discuss the new mini video camera we bought to replace the broken FCO2, and additional thermal testing we did on black Tornado couplings.

Photo: David Kelleher

I'm still putting together the pyro rocket highlights video which will be ready in the next day or so.

Monday, July 27, 2009

Acceleron Vb CATO

We just had a fun weekend at Doonside. The weather was perfect for launching and so we set up Acceleron Vb with all the modifications for the next test flight. About 10 seconds from launch the booster exploded on the pad. It was quite an impressive boom to say the least. One of the spliced bottles came apart. Although quite a few pieces came off, surprisingly there was little damage caused to the booster and sustainer. Repairs are already under way to get it ready again for the next flight.

We got good high speed footage of the explosion (over the years we have learned to keep the camera rolling while pressurising). Photos and video will be posted on the main website soon.

Other than that we ended up flying Axion V three times with foam and all the flights went really well with good onboard video and altimeter data. We were a little dissapointed that Acceleron V did not get to fly (as a cohesive unit :) ) this time, but at least the repair bill will be under $5.

Monday, July 20, 2009

Modifications Update

During the week we've continued to make the necessary modifications to the Acceleron V rocket.

The new nozzles are now finished and internally polished for better efficiency. The new nozzle seats and fill tubes are completed also. The new pressure switch is about half completed. Depending on how tests go with it this week we may need to use the old Acceleron pressure switch as a back up.

We still need to do pressure tests on the sustainer (today) as it has many new bottles. We will also do additional pressure tests on the new nozzle seats and nozzles.

We made a new fin alignment jig last week that is fully adjustable allowing fins of any dimension to be aligned with the rocket axis while they are glued in place. As a result the sustainer now has 3 new larger fins well aligned. The surface area of the fins is about 3x larger than the previous set.

We are now doing the final preparation for the next launch day coming up this week. The weather outlook for the next few days looks good, so we are hoping that will extend into the weekend.



Monday, July 06, 2009

Acceleron V first flight

We've updated our main website with details of Acceleron V's first test flight here:

The update includes photos and a video of the flight.

Although the flight didn't quite go according to plan, there were a lot of positives we got out of it, and obtained some valuable data about the rocket's performance. Reviewing the altimeter and video data we think we know what happened. We now have a plan for what changes we will need to make by the second test flight in a few weeks time. Luckily they are only minor changes.

Complex rockets take time to sort through the various issues as there are many things that can go wrong. Without flights like these we don't learn what works and what doesn't. Overall we consider the flight mostly a success.

I've also uploaded a highlights video of the pyro rockets from the launch day here:


Tuesday, June 30, 2009

High Speed Videos

Launch site

This weekend we went out to Doonside to have a look at the site and help out with a bit of mowing to clear a new area for the launch pads. The new area is needed because huge piles of dirt have been pushed up against the permanent pads. Despite there being virtually no rain all week, the road was still impassible as the ground is still saturated with standing water especially near the entrance.

The aeroclub that used to fly on the other side of Doonside has been relocated to our area as well due to the construction, so some co-ordination will be needed but there is plenty of room for everyone so that shouldn't be a problem. Currently the next launch opportunity is July 4th so we are hoping to fly on the day.

High speed video

After seeing the great high speed videos of water rockets made by Dr. Pitan Singhasaneh using a Casio Exilim EX-F1 camera ( We decided to get a high speed camera as well. Unfortunately the F1 is still too expensive here in Australia and so we decided to get it's baby brother the 9MP Casio Exilim FC-100. Overall it's a nice camera with some really cool features. I was really surprised at how small the whole thing is. Other than the 30 fps 1 second burst it can do - which will be great for launches, it can shoot HD video at 30fps, but also high speed video at 210, 420 and 1000 fps. The trade-off is the frame size decreases as fps goes up. The 1000fps is pretty useless due to the tiny frame size but for very specific tasks it may work. For the high speed shots lots of light is necessary otherwise the video is grainy. Both the 210 and 420 fps frame sizes are usable.

I've posted a video of some of the first tests here:

We want to bring the camera with us for launching the big rocket. We have to decide though whether we want to film the launch in HD or high speed. The high speed is likely to be more interesting though.


Monday, June 22, 2009

Launches on Hold

We are currently in a holding pattern with launches for two reasons: The weather here in Sydney has not been co-operating in the last few weeks and the regular launch site at Doonside is currently undergoing construction. The heavy machinery has torn up the access road and with the rain it has made it impassable.

Here are some recent pictures of the site:

We need to get in and help with the working bee to try to restore the site so it is suitable for launches and parking again, but due to the bad weather this was postponed again. The forecast for this weekend looks more promising with the rain stopping in the next day or so.

Acceleron V

During earlier testing we found that a couple of the 13mm nozzles leaked slightly when the rocket was unpressurised. We are only talking a few drops of water. The o-ring grooves were fractionally too deep, and with the narrow o-rings this is more of an issue. When the boosters were pressurised the o-rings would seal properly. Dad decided to fix the issue and machined up a whole new set of nozzles.

We also decided against a guide rail for this rocket as there is no convenient place to secure it. We could attach it to the outside of one of the boosters, but it could cause more harm than good due to the potential uneven drag on that side of the rocket. The nozzles are quite long ~40mm and the fill tubes are quite a tight fit, so in the absence of a guide rail, the rocket straying from vertical could potentially wedge one of the nozzles on the fill tube causing all sorts of chaos.

We decided to machine out the top 30mm of the nozzle to be about 15mm in diameter and the nozzle hole itself by another 0.5mm. This allows the nozzle to pivot more without seizing on the fill tube. The 0.5mm increase to the nozzle gives about 8% more cross sectional area so we will get a little more thrust on take-off.

Static Tests

While we have been waiting for the weather to clear, we have done more static tests on the test stand of the membrane variable nozzle as a follow up to the tests done in March. We are in the process of putting all the data together at the moment so we will post all the results once they are in a meaningful form. All up there were 19 static firings made last weekend. (14 done in March)

Launch Abort Valve

After recent discussions on the Yahoo water rocket forum we decided to have a go at building a simple, remotely operated launch abort valve. The idea was to try to make it from commonly available components, and no need for special tools. The full details of the launch abort valve are here:


Tuesday, June 09, 2009

Testing and progress

I've updated the build thread on the Australian Rocketry forum with some of the latest testing and build progress here:

The booster and sustainer are now mostly complete, still with a bunch of little things left to finish. So far the weather forecast looks good for the weekend so we have our fingers crossed.

I'm not sure how many hours we've put into the rocket so far, but it is more than I had anticipated when we started the build. It is probably approaching around100 hours now. A lot of the time is spent thinking about how to solve particular problems, so hopefully the second time around a build should be much faster.

Thursday, June 04, 2009

3 Years

3 years ago today, dad and I built our first horizontal launcher, hooked it up to a scuba tank and launched our very first coke bottle in the back yard. A few days earlier after seeing a Mythbusters episode trying to disprove the Japanese video of a guy getting launched across the water, we looked up water rockets on the net and discovered a whole world of water rocketeers building all kinds of water rockets. And so the fun started....

Before we started collecting detailed flight data for each flight, we did about 60 launches of basic single bottle rockets. Since we started keeping records there have been a further total of 406 firings -
  • 5 of which have been pyro rocket flights
  • 3 flour rocket flights
  • 67 static tests
  • 331 water rocket flights ( 23 multi-stage/booster flights )
We are still enjoying the hobby, and have a long list of things we still want to try. A big thank you goes to the great water and pyro rocket communities that have prooved to be a very valuable resource in helping to design, build, test and fly rockets.

Monday, June 01, 2009

Rings, fins and other things

We had quite a busy weekend working on the rocket. The following has been completed:
  • The ring brace is now fiberglassed, sanded and painted
  • The wiring connections to the servos and pressure switches are complete
  • The parachute deployment mechanisms are complete
  • The fins are complete
  • The sustainer is finished
  • The flight computers have been attached
Still left to do:
  • Painting
  • More pressure testing of various components and bottle assemblies.
  • Simulations
  • Attaching various things like parachutes and the ring brace to the bottles.
  • Mounting a camera and altimeter in the booster
  • Bunch of little fiddly things
  • Full integration testing of the entire system at 20 psi
  • Launch abort loop and hook
One of the main problems we've had is improving the yield of the splices at the 130psi range. Dad pressure tested the 9 spliced pairs to 130psi for 2 minutes during the week. Out of the 9, five were good. Two had very minor leaks of a few drops and a couple had more significant leaks. Dad had a go at sealing them from the inside with silicone and we are waiting to re-test them on Wednesday. We have about 7 spliced pairs currently that we need to test and get 4 good ones out of those. Otherwise there are still 2 good ones we have in the Polaron rocket should we need them.

There are still minor things to finish during the week, but most of the long lead time things are done.
The new ring brace can be seen in the image below (red with white stripes). The red Velcro strips around the bottles and lower section are only temporary at this stage to hold the rocket together as we work on it. It makes it easy to remove bits as needed and put them back.

In the photo below the rocket is still unpainted, but you can see the flight computer compartment on the right, and one of the parachutes on the left.

The new sustainer is a little shorter (about 6 inches) than the photo posted earlier in the month because we decided to put a Robinson coupling in the lowest bottle to generate foam in flight for a longer and more visible burn.


Tuesday, May 26, 2009

Acceleron V stager testing

We've updated our main website with more details and photos of the Acceleron V build. There is short video that shows a couple of the pressure tests of the staging mechanism.

There is still lots to do before the next NSWRA launch. We're currently working on the parachute deployment mechanisms. We decided to use two of our existing parachutes for the booster so we will save some time there.

Wednesday, May 20, 2009

A little bit of pyro

We bought our 10 year old son a bigger pyro rocket for his birthday this year. He wanted something a little bigger than the Thunderbee he has flown several times on a 1/2A3. We bought him the Praetor from Suburban Rocketry.

Slowly over the last few weeks he actually built most of it himself, with a little help to get things lined up from dad. He did learn quite a bit about how a pyro rocket gets put together and the different materials and construction techniques needed compared to water rockets. He also learned quite a bit about the painting and extensive sanding process. It was just as much interesting for me to see how pyro rockets are put together. He chose the following colour scheme and spray painted it also mostly himself:

We had some decals from old model planes so we put those on to give the rocket a bit more character. We're hoping to launch it this upcoming NSWRA launch event on a C6-5.

Monday, May 18, 2009

Staging mechanism assembly

This week we have been continuing with the staging mechanism of the Acceleron V booster. We've also made the non-return valve that fits inside the long PVC pipe. The valve was adapted from the old Acceleron booster that was mounted in a bottle cap.

More information and photos of the build can be found on the Australian rocketry forum here:

It hasn't all been smooth sailing though. After test assembling the base plate in the launcher with the actual booster segments fitted with their nozzles we discovered that the central mechanism couldn't go down far enough to lock into the mechanism. Not wanting to machine up a whole new central nozzle, we decided to put a thick plastic washer underneath it on the base plate, and using somewhat longer screws to hold it all in place was enough to resolve the issue.

The other issue we had was while trying to fit the hose on the end of this central nozzle we found it a little difficult, so we decided to machine down the hose connector a little bit to make it easier. While machining the small thin tube caught on the knife and unceremoniously bent, and was damaged. While trying to straighten it, as suspected it snapped off. Dad came to the rescue and machined off what was left, then proceeded to cut a thread into the mechanism and made a new separate hose connector component that simply screwed into the end. We epoxied it in place and all was good again.

We've also had discussions about how we are going to make the top bracket that holds the boosters together. It's going to involve steam.... but more on that later. :)

Monday, May 11, 2009

Building continues

We have been continuing with the development of the booster. Here we test assembled some of the components to check for clearances and get an overall idea of how it will go together. The nice thing is that it just fits inside the house. (one inch from the ceiling)

One thing that was very evident is how high the top payload is, and we are going to have to move the flight computer down, or at least extend the Arm button down where it can be reached.

Pictured along with the rocket is a standard sized kid. Simulation says that the booster alone would lob him about 4 meters into the air. :)


Thursday, May 07, 2009

Acceleron V construction

Most of this week has been spent making the base plate for Acceleron V. Here it is partially assembled:
The sustainer air supply and retention mechanism still needs to be machined, which will be done soon.

The bottle splicing is also complete, although the bottles still need reinforcing around their necks. They also all need to be pressure tested. We made a number of spare spliced bottle pairs should any of these leak. We will give all the splices at least 1 week to fully cure.

We used PL for the splicing and Sikaflex for sealing the joints as we had leak issues when making spliced pairs for the Polaron rocket a couple of months back.


Thursday, April 30, 2009

Launch Report update

We've posted an update from the last launch day on our main website here:

There are a number of photos from the day as well as some more design details of the Acceleron V booster build. We are pretty much now focused on getting the Acceleron booster completed. While we are aiming to have it ready for launch at the next NSWRA launch event, it may likely slip until the next one.

Monday, April 27, 2009


We only flew a couple of smaller water rockets this weekend at the NSWRA launch day event, mostly due to the high wind conditions. A few of the pyro rocketeers lost their rockets somewhere over the adjoining suburb, so we decided to call it quits a little earlier too.

I'm putting together the launch report at the moment which should be available in the next few days. It also includes a progress update on the Acceleron V booster.

We did launch a couple of small pyro rockets with the boys and safely recovered those. During the week we bought a bigger pyro rocket kit from Suburban Rocketry so we are now building it with the boys in between all the other rocket stuff. I can see we'll probably be building a few of these as they are fun to build, but the cost of motors is pretty high when compared to air and water. :)


Sunday, April 19, 2009

Acceleron V modelling

This weekend we spent a bit of time modeling Acceleron V to scale in 3DS Max in order to figure out the spacial relationships of the new release mechanism and the position of the staging mechanism. We now have a clear idea how to build the base support mechanism and what clearances are needed. The images below include the second stage, although the colour scheme will most likely be different.

The length of the second stage has not been determined yet but will be such that the entire two stage rocket is stable in flight.

The rocket is designed to be launched from our drop away booster launcher.

Detail of the release mechanism and the location of the staging pod. The pod contains one of our flight computers and uses a standard 9mm Gardena release mechanism to release the second stage. Depending on the length of the second stage, the pod can be moved further up the rocket.

Fabrication of the booster segments continues, and we now have a materials list for the release mechanism so we can start construction on that as well. The staging pod is already finished. I'll post actual build pictures next time.

We've also been preparing our rockets for the next NSWRA launch day. There are no major new rocket changes although there are some experiments we want to run in getting more stable descent video as well as flying the whistling nozzle. After the last crash the Axion rocket has been rebuilt.

Saturday, April 11, 2009

Acceleron V Redesign

We've been collecting 2L bottles from friends and family over the last couple of months to make up more spliced pairs for the new Acceleron booster. After we had a few leak problems with the last spliced-pair batch we're now trying the sikaflex glue as a sealant in the joint area but PL premium as the main glue that holds the splice together.

Some of the 2L bottles ready for processingBottles cleaned, trimmed and curled ready for sanding.

All up Acceleron V has three segments in the cluster and each segment uses 3 x 2L spliced pairs giving a total volume of almost 33 liters. Up until now we have been using 10mm nozzles on the Acceleron boosters but with the larger volume and larger weight of water we have decided to go with the larger 13mm nozzles we use with our drop away boosters. Including the second stage, the total liftoff weight is now going to be approaching 16 kilos. Switching to the new nozzles would have also meant that we would have to machine new nozzle seats for the old Acceleron launcher.

Instead we decided to adapt the booster launcher to work with the Acceleron rocket. The old Acceleron rocket uses a speargun release mechanism, so we are replacing this with custom made aluminium gardena nozzle mounted on the baseplate that holds the cluster together. This nozzle will fit into the existing Gardena central release head.

Because this launcher's minimum nozzle spacing does not allow the three segments to be up against each other like in the old Acceleron rocket, we are going to separate the cluster segments further from each other. This will require a new baseplate for the cluster. This again will be made from aluminium and reinforced with fiberglass. By separating the cluster segments we will be able to lower the second stage release mechanism pod further down in between the segments allowing the second stage to also sit lower in the booster. This will help partially solve one of the problems we've been having with the old Acceleron rockets, namely the stability of the second stage on the end of the staging mechanism. The length of the Acceleron booster and the location of the staging mechanism prevented us from making the second stage any longer. If the second stage had been much longer there was the danger of snapping the rocket at the staging mechanism during the boost phase. By lowering the second stage between the segments, we can use the segments themselves as guide rails for the second stage. This will allow us to build a second stage around the length of an Axion or Hyperon rocket.

The second stage will be pressurised through the central release head allowing it to use a different pressure to the 1st stage. The Acceleron parachute(s) will now be located in the space between the spliced pairs.

With the second stage supported by the segments, it will also mean that the second stage can have a boat tail for better drag efficiency.


Friday, March 20, 2009

Parachute Detachment Mechanism

In chatting to Mat G. this week, he described how his rocket got stuck in a tree flying one of the computers. I thought that the FC should at least be able to try to help in such a situation. Snagging a parachute on trees is a very common problem, especially when the wind carries the rocket beyond the designated range.

We put together a Parachute Detachment Mechanism prototype (DetMech) that allows the rocket to automatically let go of the parachute cord after a predetermined time.

The full details of the mechanism are available here:

We want to test fly it next to see how the clamp holds up under high speed deploys. ... yup just another excuse to fly more rockets. :)


Wednesday, March 18, 2009

FC V2.0 Development

With the last of V1.6 sold yesterday, we are using the proceeds of the sales on the development of the next version of the FC. The new version is quite different and includes a lot of new functionality some of which is intended for use on pyro rockets.

We are also aiming to reduce the overall footprint and weight of the FC by using surface mount components. Circuit design is about 2/3rds done but the software spec for this version is only about 1/4 complete. The core is still based around the PIC16F628A but will have a few more support chips to offload some of the functionality.

The expected time to completion is likely to be around 6 months as there is a lot of new code to write, but that may be pushed back further depending on time allocation priorities on all other projects currently under way. I'll post progress updates and more details when significant milestones are achieved.


Monday, March 16, 2009

Flour Rockets

We've had a bit of fun this weekend with the kids launching rockets powered with flour rather than water.

The full write up of the days events is here:

Included are photos and a highlights video.


Saturday, March 14, 2009

Variable Nozzle Testing

This weekend we pulled out the static test stand and continued with more testing. Before the tests we spent a couple of hours making the flexible nozzle membranes and the tools to make them.
Dad machined up the stamp tool that cuts the membrane out of the rubber sheet to the exact size (the brass component), and then the hole punch alignment tool, that allowed us to center the hole punch in exactly the middle.

We ran about 14 tests with different materials and different sized holes. We haven't had a proper look at the data yet, as that will again take a couple of weeks to analyze and write up. One interesting surprise we did get was during one particular test the nozzle made a pretty loud whistling noise. We might fly that one on the next launch day just for fun.

We'll post the full data on the nozzles when the analysis is complete.

Tuesday, March 03, 2009

Day 73 - Highlights

We've posted an update on our main website with the highlights from day 73. The update includes photos and a video of the days' events.

.... now it's back to the workshop....

Sunday, March 01, 2009

2 great flights, 2 crashes

We had an overall great day at NSWRA's launch event on Saturday. We got to fly the newly rebuilt Polaron VIb rocket that hadn't flown since May last year. Both flights were almost perfect with very nice vertical flights, and no spin which made for great stable ascent video. Although we had tested the bottles to 130psi, we only launched at 120psi to be on the safe side. During testing earlier in the week, we had 3 out of 4 spliced-pairs leak. So only the bottles that had withstood the full 130psi were used in the rocket. We were very happy with the performance of the rocket, and were glad that the stability improving changes we made had been effective.

While other NSWRA members were drilling holes in the clouds (literally) with their impressive pyro rockets we were busy taking core samples near the launch pad. One was the FTC rocket and the other was Axion II. Full details of the crashes will also go in the update.

Saturday, February 21, 2009

Progress update

We've posted an update on main site that describes some of the things we have been working on lately. At the moment we are trying to get two big rockets (Polaron VIb and Polaron VII) completed by the NSWRA launch event next week, as well as the FTC rocket. Currently they are still in pieces, but should be assembled in the next few days.

The update is here:

Thursday, February 12, 2009

Flight Computer V1.6 Details

We've put up the full details of the latest flight computer on our main website here:

There have been a number of changes since version 1.5:

  • The altimeter power connector has been removed. The altimeter can still be powered via the second servo connector. If both servos and altimeter are used then the second servo can be connected to the altimeter's pass through connector to connect the second servo motor.
  • The PCB is now double sided resulting in less weight and 25% smaller area. The PCB is also only 27mm wide allowing it to be used inside T-8 and T-12 FTC tubing.
  • The timing parameters have been changed to represent seconds directly rather than the more confusing offset/multiplier technique used in V1.5. Although the direct setting reduces the range of values possible, they should cover the vast majority of situations as used in the real world. The timing for both D1 and D2 phases can now be set between 0.1 and 99.9 seconds in 0.1 second increments.
  • The setting of the two delay parameters in the Normal mode have been removed and the FC now simply requires one press of the ARM button to ARM it. This was done to make it simpler for the rocketeer to use in the field, and prevent accidental timing changes.
  • The servo control pulse width range has been extended to allow driving some non-standard servos their full range.
HTML clipboard
These are now available for purchase. (Includes FREE worldwide delivery!)


Sunday, February 08, 2009

FTC nosecone success

Well nosecone attempt #4 (see below) worked out as planned.

This is the end result of 8 layers of nylon stockings with polyester resin, one thin coat of 5 minute epoxy, thin coat of spray on putty sanded back for a smooth finish and a couple of coats of gloss paint. Ideally the polyester resin will be replaced completely with epoxy. I will update the nosecone build procedure with the next main website update.


Saturday, February 07, 2009

FTC Nosecones

Since we are rebuilding the FTC payload, we've decided to replace the Kinder surprise egg nosecone with a more streamlined one. I machined up the nosecone mold out of a solid piece of plastic and polished it.

The nosecone mold prior to final polishing.

Attempt #1 - Heated the base of a small PET bottle and forced it over the nosecone mold and tried to heat shrink it over the top. Heat gun was too close and the base started crystalizing. The nosecone ended up flying quite nicely into the rubbish bin.

Attempt #2 - Heated the base of another small PET bottle more gently this time. This was only a slightly better result than the previous one. Before long he joined his friend in the bin. It just could not be shrunk down enough to the 30mm diameter.

Attempt #3 - Coated the mold with release agent and impregnated fiberglass matting with resin and then started wrapping it on the mold, but as you can imagine a nosecone isn't easy to round off with flat matting. Before long I had a nice mess on my hands with lots of bubbles, folds, and the rounded front was anything but. After another 5 minutes of trying to correct it, I finally pulled the whole mess off the mold and guess where it went?

Then I emptied the bin ready for more nosecones.

Attempt #4 - I went to the local supermarket and bought a couple of pairs of pantyhose (nylon stockings) and a shoe polishing brush. Okay I did get strange looks at the checkout counter as they were the only things I purchased. Little did they known those stockings will do over 200km/h one day. Instead of the release agent, I stretched the thumb section of a disposable rubber glove over the mold and tied it off. This made sure there were no creases anywhere. I then proceeded to stretch 8 layers of pantyhose over the top of that again tying it off at the bottom. I then simply poured the fiberglass resin all over it and worked it in with a brush and with my gloves. The nosecone is now drying so I will know in the morning if it worked. Otherwise it looks pretty good.

... Oh and the shoe polishing brush I bought earlier I had cut into about 20 strips each with 5 sets of bristles and use them as disposable brushes for the fiberglass. At 10c per strip they can be thrown away after each job.

I'll post results of attempt #4 when it dries and assuming I can get it off the mold.


Wednesday, February 04, 2009

FTC rocket upgrade

We've been working on a number of rocket related projects this week. We started redoing the FTC payload bay that includes several improvements. We are replacing the split nosecone made from FTC to one made from Fiberglass. The fiberglass is a lot stiffer, but still lightweight and does not collapse like FTC does when it is cut in half.

Because we used a piece of FTC to form the half-shells, their diameter is also the right size for fitting over the FTC tube. The V1.5 flight computer used is being replaced with the smaller V1.6, and the batteries are being replaced with four AAAAs. The net result is a more compact payload section that is a little lighter than the predecesor. We are also creating smaller fins for the rocket to reduce the overall weight near the tail.

We have also tested an all-Sikaflex splice this week with the glue failing at around 130psi. The glue had given way, as both the sleeve and the bottle that was glued inside the sleeve were undamaged. The glue was left on both surfaces. This means that this glue probably should not be used for pressures above 100psi. We have a number of spliced pairs using this glue already made for the Acceleron rocket, so we will have to replace them with PL splices. We can still use them on other test rockets so we will hang on to them.

Fiberglassing is definitely a messy business but I'm getting used to it a bit more now. Last week I bought a proper respirator with organic gas filters for $45 and its really effective and I can't smell any of the resin when working with the fiberglass.

As we are working on a lot of spliced pairs at the moment, there is quite a bit of sanding to do of the surfaces to be glued. We always used to do it by hand, but we've made it a little easier now by getting a piece of PVC pipe and cutting a pair of long slits in it. We then take two rectangles of sandpaper and put them back to back and place them in the slot. The pipe then goes in the lathe. When it spins the sand paper makes a kind of S-shape and makes it much faster to sand the outside and inside of the bottles. It reduces sanding time by about 1/3 and also is alot easier on the hand strain.

Soldering up the flight computers has also been performed in the background in spare time, and documentation continues to be worked on.


Wednesday, January 28, 2009

V1.6 PCBs have arrived

We received the PCBs yesterday from the manufacturer. I was very impressed with their quality, and breathed a sigh of relief when all the components fit. :) We'll definitely be using this manufacturer again. I was keen to see if the layout design would work but unfortunately we had a blackout and so I had to hold off on soldering the first unit until after dinner.

Breathed another sigh of relief when the FC powered up first go. The software is pretty much done, just doing a few more minor tweaks to the servo timing routines and testing it with a range of servo motors.

The version shown above has the assemtech G-switch soldered in and it's set up for 9V operation. With the battery clip the FC weighs in at 18 grams.

Next up we are going to do some flight tests with them, and I'll replace the V1.5 FC in the FTC rocket with one of these to see how it performs under higher G-loads.

The documentation and manual still needs to be completed, but that is also close to being done.


Monday, January 26, 2009


Last weekend we went out to Doonside with some of the NSWRA rocketeers but only equipped with lawn mowers and brush cutters. All the tall grass is now gone from the launch area so it is a lot nicer to move around and set up. Phil also installed permanent launcher bases which will make setting up much easier and there is less gear to bring each launch day.

While we wait for the new flight computer PCBs to arrive (Apparently they have now been shipped) we have been continuing to build the Acceleron V booster. We have mostly been testing and reinforcing the 2L spliced pairs. The booster uses 9 of them. We splice a pair together and then hydro test it up to about 80 psi. If there are no leaks we then add the neck reinforcing rings and wait a few days for them to dry, before testing again to 130psi. This way if the original splice fails at the 80 psi we don't waste time and bottles reinforcing the necks. The neck reinforcements are necessary as at about 110psi stretch marks begin to appear around the neck of the 2L bottles.

We are making a few spare spliced pairs in case any need replacing. The spare ones can later also be used for the new bigger Polaron boosters. A couple of the pairs are all-sikaflex splices so we are keen to see how they hold up.

I have also had my first go at fiberglassing. I wanted a simple project initially to learn the technique and so decided to make a ring fin that should be able to withstand a bit of punishment during tail-first landings under parachute. It was quite easy to do but I can see I need to pay a bit more attention in getting all the bubbles out and work a bit more at getting the resin in all the way through. But overall I was pretty happy with the results. The ring fin is 37 grams, 0.8mm wall thickness and uses 3 layers of fiberglass matting. It is significantly stiffer than a PET ring.

Here is the ring fin after the first rough sanding. I'll spend a bit more time making it smoother before painting it. The ring fin will be attached with CF struts to the main rocket body.

The next few fiberglass projects will likely to be fin strut attachment supports for the pressure chamber, and reinforced balsa fins. When I get a bit more comfortable working with it, I'll have a go at reinforcing some pressure chambers.