Following the successful flights of the Titania parallel-stage rocket at IRW99, I wanted to make another, more spectacular parallel-stage model. Meanwhile, on the newsgroup uk.tech.rocketry, there arose a thread about building a scale model of a British missile. So I decided to build either a Bloodhound or a Thunderbird.
Steve Moores of SERFS, who also wished to build a Thunderbird and who had thoroughly photographed two specimens, sent me copies of the photos. Here is one at the REME Museum.
The first flight of the sustainer alone took place during Boglob, a flying event at Comer Farm, on 29/4/00. The non-existent tracking system successfully locked onto a non-existent enemy aircraft. More accurately, the rocket, underpowered by a single D12-3 and flying into a fairly strong wind, weathercocked severely. But two days later, another D12-3 sent it on a near vertical flight into a calm sky. One large fin was damaged when the heavy nose-cone swung back on too short a shock cord.
By the time of UKRA2000, the boosters were complete and I took the whole model along, primarily to show it off. I wasn't planning on flying it except in the unlikely event of near zero wind. But on the evening of Friday 2/6/00, near zero wind is exactly what was there. The full stack with boosters attached was prepped and placed on the pad. As it was more powerful than a single D12 (due to there being a D12-5 in the sustainer and C6-0s in the booster) it was moved to the high power area, from which it executed a perfect launch. The boosters separated as intended, leaving the sustainer to cruise on up. A "recovery crew" of volunteer children picked up the boosters while I found the sustainer a little way into an adjacent corn field.
The next flights were at International Rocket Week 2000 The nose had now been repainted white, in accordance with two of Steve Moores' photos. Dave Hart supplied this picture of the rocket launching.
Thunderbird returned to International Rocket Week in 2006. The nose cone
had been reshaped slightly to make it more accurate. Photos:
On the pad.
Boosters have burned out and are starting to separate.
Boosters have now fully separated.
Booster attachment system
After examining the photos and making a guess as to how the real Thunderbird's boosters were attached and detached, I designed the model's system along roughly similar lines.
At the rear of the sustainer are rectangular posts with lugs made from dowelling. Each booster has a pair of tubes which lock onto these lugs, keeping the tail end in place. A balsa assembly on which the tubes are mounted pushes against the posts on the sustainer, transferring thrust. There are eight posts in total to engage the four boosters.
At the front, the boosters are held in place by a collar. The collar is split in half, one half being fixed to each of two boosters mounted opposite each other on the sustainer. Each half has a lug at the tip. The other two boosters each have two short tubes on the detaching nose which lock onto these lugs so that when all four boosters are attached, the collar is intact and the boosters can not fall off.
There are also four slots in the sustainer body. Each booster has a small balsa block which fits into a slot, so that when the boosters are locked down by the collar, these blocks and slots prevent them from moving around, up or down the sustainer. This ensures that, should one booster ignite late or not at all, it will not be left behind.
When the booster motors burn out, they generate forward pressure, which blows the noses off the boosters. As the collar is held together by lugs on the noses of two boosters, this effectively breaks the collar. The boosters can now drop off the back, allowing the sustainer to continue upwards.
Fin span (sustainer only): 202mm
Fin span (with boosters): 338mm
Photos of fins, scaled so you can print them and use them as templates.
Photos by Steve Moores.
Before printing these photos, set your printer's orientation to "Landscape".
Large fin on sustainer: from the training round at Coventry.
Aft fin on sustainer: from the training round at Coventry. The spike on the leading edge is not present on the service round at the REME museum, so I did not include it on my model.
Booster fin: from the service round at the REME museum.
VCP data file
Get the VCP file here!
1. To simulate the aft sustainer fins, I used two entries ("Fin Unit 2" and "Fin Unit 3").
2. "Fin Unit 4" is actually the boosters' bodies and cones. VCP can't handle parallel tubes so I pretended they were "cardboard cutouts".