I really do not recommend doing a clustered shuttle. The thrustlines and the thrust levels have to be balanced out properly. And they all have to ignite just right, or the wide spacing of the engines apart from each other will make the model veer off at liftoff.
For my own models that I tested out thru the years, I only did one that had a true cluster in the SRBs for thrust, about 1/100 scale with BT-55 SRBs. It used a North Pacific (now Guillows) foam orbiter, and to balance out the mass of the orbiter I shifted the noseweight inside the ET nose to the -Z axis (Orbiter is on the +Z axis), to balance it out. This allowed the SRBs to have the engines in the center of the SRBs. I used a D12 engine in each SRB. I wanted to see how a truly clustered shuttle with main power in the SRBs would work out. I knew the risk. I figured if it worked well 10 times, I might go for it for a serious model. Well, on flight #1, both D12s ignited. But they did not ignite exactly at the same time, so there was a thrust curve mismatch. Attached is a B&W photo showing it. Hard to tell in the pic, but one D12 had produced more thrust than the other D12, and caused the 3/16 launch rod to bend as the shuttle went up the rod. As it started to leave the rod, the bend un-did itself, which caused to shuttle model to do a strange whip roll, and then when it settled out it was climbing at a 45 degree angle. That was enough of a close call for me, as I said I was looking for it to work 10 times, and it came close to disaster on the first try.
Years later I tried another approach. F engine in the bottom of the ET, but with Token A motors in the SRBs. The SRBs were rigged to be sepped, so the A3s also were used for a time delay and ejection charge for the SRBs. Well, flight #1, the F in the ET chuffed, so it did not take the model up for a couple of seconds. But, the A3s fired too. Their thrust was enough to push the model up the rod a couple of feet. Then it fell back down and hit hard, breaking an SRB aft mount. So when the F in the ET did finally fully ignite and carried it into the air, one of the SRBs was loose, only being pulled by the front end. Since the stability was based on a big fin on each SRB, that played hell with he stability, it went nuts.
I tried to fix the problem of that approach. Same model, but sitting on a switch mounted to the pad (a N.C. switch held "open" when pressed). When the F in the ET lit, and pushed the model up, the switch would close and fire current through extended leads going to flashbulb ignitors in the SRB A3s. So, the launch sequence was ignition of the F, move up the rod an inch, then ignition of the A3s. But, I never got both A3s to ignite. Most of the time only one lit, and at least once, zero lit. So of course this also meant that the "dead" SRBs free-fell to the ground since they counted on the A3 ejection charge to deploy a chute (The ultimate model I ended up building has no engines in the SRBs . There is a wind-up timer that starts running when the SRB seps, then about 2 seconds later it triggers a rubber band driven system to push off the nose and deploy the chutes).
The only reason to even bother with the Token A3s was for scoring flight points in a contest. I just decided it was not worth the risk. So, I just adopted the single engine method, with one engine in the ET, offset towards the orbiter.
BTW - even with a single engine in the ET, the offset distance is key. The heavier the orbiter weighs compared to the rest of the stack, the more the engine mount needs to be offset towards the orbiter. This is even more critical for a minimized fin system such as I used. The old Estes shuttles plug-in fin system is more tolerant of an imbalance, to a point. Of course the Estes shuttle also was done presuming a certain range of mass for their orbiter.
For my models, I had to find out where to locate the engine mount in the ET the hard way. Build pretty much the whole model, except for gluing on the aft dome and of course not installing an engine mount. I rigged up a temporary bulkhead taped into the front end of the ET tube, and this temporary bulkhead had a clear plastic window glued underneath, with some grid lines. Separately, I rigged up a pointed dowel to be mounted upright on a table, with the point of the dowel standing high enough that the entire shuttle stack could be placed onto the point of that dowel, with the point visible thru the clear window. So, I used that whole rig to find out where the lateral balance point was on the model, as regards the Z axis (+Z towards orbiter, -Z opposite orbiter). I added a 1/8 balsa standoff to the -Z axis of the ET, near the top, and allows a piece of thread to hang down off of it, with a blob of clay at the bottom. So, that was a plumb-bob. I shifted the whole stack along the point of the dowel until the thread line (plumb bob) was parallel to the ET tube. When I did find that spot, I marked it on the clear window, and used that distance to determine where the center of the engine mount would be.
Now, I left out a couple of details to help get that idea across. Now that (hopefully) you see how and why, heres what I left out. In the above procedure, the ET nose was not in place, in order to allow seeing the window. Well, I had to account for the missing mass of the ET nose in that step. So, I weighed the ET nose (INCLUDING any noseweight it would end up having), and added simulated weight (a bunch of pennies taped in a roll) to the dead center of the front bulkhead. And, I guessimated the mass of the missing aft dome and aft centering rings that would be added later, plus the glue mass, and mass of paint and any details that would be added to the ET/SRBs. The grand total of the guesstimated mass that would be added later, was all added to the mass added to the center of the temporary front bulkhead.
I should also say that when I did that, the orbiter was already completed, with radio gear, so it weighed what it would for flight. This made for a tricky situation when I wanted to try out a new fin system in 1988, but did not have radio gear in the orbiter. The rest of the shuttle stack was still what I had built in 1984, I replaced the 1984 Estes type fins and tested out one fin on each SRB at 45 degrees. Since the radio gear was not in the orbiter (in another model by then), the ET would have needed a different thrustline for the engine mount. But, it would have been too much of a hassle to re-do the ET mount. So, what I did was add dead mass to the orbiter to bring it back to what it used to weigh.
The above image is the 1988 test flight of the 1/72 model that proved out the single fin on each SRB stability. The bright spot on the bottom of the ET, near the orbiter, is the exhaust flame from the virtually smokeless exhaust of the old Aerotech F15 engine that it flew on.
Now, it gets even a bit more complicated if the orbiter is supposed to glide. The up elevons will tend to try to make the shuttle stack pitch nose-up on boost. The ideal solution to that is to rig the elevons so they can be flat for boost and come up for glide. But that may not be practical, such as for using the North Pacific/Guillows orbiter. So, for those it could help for that perfect lateral balance I described above, to be made a bit imperfect, so that the thrustline would be a hair closer to the orbiter so as to cause the stack to pitch down a little bit to counteract the pitch-up force of the elevons. But, that could also be addressed by positioning he shuttle so that the belly of the orbiter faces into the wind, so that the wind will try to pitch the model into it from weathercocking (as usual), to counter the pitch-up for the elevons. Also, if the fins are more like the old Estes shuttle, or the Dr. Zooch shuttle fins, it will not be affected as much (it is the one fin on each SRB type fin system that is really sensitive).
- George Gassaway