How to Use Small Study Models to Solve a Tricky Roofline or Tower
If a roof edge, ridge, spire, or tower connection is hard to judge from drawings, build a small, focused study model before making the full presentation model. For an architecture student or hobbyist working from a plan and elevations, the task is to compare the difficult geometry at a manageable scale. Keep the model simple enough to revise: test the footprint, heights, slopes, and joins first, then add only the details needed to read the form.
Decide what question the model should answer
Write down one question the model needs to resolve. For example: Does the tower sit centered on the roof ridge? Does the roof plane meet the tower at the intended height? Does a projecting eave make the tower look too short? A study model is most useful when it tests a specific design decision; it need not represent every part of the building. The University of Melbourne’s model-making guide distinguishes process models, which help examine particular decisions, from presentation models made for a more finished audience-facing result. Design and Assembly, Fab Lab
If you have more than one uncertainty, rank them. Start with the one that affects the other measurements. A tower’s position and height, for example, should be settled before you spend time on finials or surface patterns. This order is an applied workflow: resolving the main geometry first makes later decorative choices easier to evaluate and prevents detail from disguising a basic proportion problem.
Collect matching plan and elevation information
Use at least a plan view and the elevations or sections that show the roof and tower. A roof plan tells you the footprint, ridge direction, hip lines, and tower location; an elevation or section shows heights and slopes. If the drawings disagree or omit a needed dimension, mark that uncertainty rather than silently guessing. You can still make a model to compare plausible options, but label the assumed value so it is not mistaken for a confirmed dimension.
Check that any printed drawing is actually at its stated scale. The University of Manchester’s model-making account describes scaling CAD drawings to a required model size, printing plan and elevation references, and checking the print scale before construction. Making the U.o.M. Snowglobe Model A quick check is to compare one known dimension on the print with the drawing’s stated dimension. If a 10-metre wall should measure 50 millimetres at 1:200, but the print measures 45 millimetres, do not use it as a cutting template until corrected.
Choose a scale that exposes the problem
Select a scale that makes the roof and tower large enough to handle, while keeping the test inexpensive and quick to rebuild. Scale means model length divided by real length: at 1:100, one metre becomes 10 millimetres; at 1:200, one metre becomes 5 millimetres. Convert all dimensions using the same ratio, including the roof overhang, tower footprint, and vertical heights. Mixing scales between plan and elevation can produce a model whose parts appear to fit while representing incompatible dimensions.
For a first pass, make the roof and tower as simple blocks or folded planes. Choose card, boxboard, or foam core according to the shape and level of precision needed. The University of Melbourne’s materials guide describes boxboard as low cost and easy to glue and paint, corrugated cardboard as suitable for sketch models and bases, and foam core as a lightweight sheet material. It recommends a sharp blade and several gradual cuts when cutting foam core to avoid tearing its inner layer. Materials, Maker Spaces Knowledge Base
Convert roof pitch into a model slope
If the roof pitch is given as rise over run, use that ratio to calculate the vertical rise for the half-span or other horizontal run shown in the section. A 7:12 pitch rises 7 units for every 12 units of horizontal run. For a 15-foot run, that gives 105 inches, or 8 feet 9 inches, of rise; the Pennsylvania Department of Education illustrates this proportion method in its carpentry teaching material. Determine the Rise and Run Using Roof Pitch Theory
Then scale the rise and run by the same model ratio. As an illustrative calculation, suppose the model is 1:100 and the roof run is 15 feet. At 1:100, 15 feet becomes 45.72 millimetres; the 8-foot-9-inch rise becomes 26.67 millimetres. These are calculated example values, not dimensions for a particular building. The slope stays the same because both dimensions have been reduced by the same factor. Use the actual drawing dimensions for your own model.
Be precise about what the pitch notation means. A common roof pitch ratio gives vertical rise per stated horizontal run; it is not the sloping roof-surface length. If your drawing provides an angle instead, use the angle to lay out the roof plane, or derive the rise and run from that angle. Do not infer a pitch from a perspective view alone when a section or dimensioned elevation is available.
Build the model in layers
Start with a flat base and mark the building footprint, ridge line, and tower footprint. Make the main roof planes removable or assemble them with light temporary joints so you can adjust their relationship. Build the tower as a separate piece. This lets you slide or rotate it against the roof, test its position, and see where the junction lands without rebuilding the whole model.
Make one roof plane at a time from the scaled outline. For a simple gable, cut two matching slopes and join them at the ridge. For a hip or multi-plane roof, use plan and elevation together to identify where edges converge; label each piece before cutting. Add a tower mass and any ridge cap or eave overhang only after the primary roof and tower align. Keep a second or third version of the part if competing options are worth comparing side by side.
The Manchester workshop account recommends marking pitch guides clearly and checking them as work progresses; its tower model used staged roof pitches and was corrected after a plan-reading error was noticed. That example supports a practical habit: draw the intended reference line on the material, cut slightly conservatively, and compare the piece against the plan before removing more. Making the U.o.M. Snowglobe Model
Inspect the junction from several views
Look at the model from above, from each elevation, and at eye level from the directions where the roofline matters most. Compare the physical model with the same views in the drawings. Ask specific questions: Is the ridge in the right direction? Does the tower footprint overhang or sit inside the roof footprint as intended? Do the roof edges meet the tower at the same height on both sides? Does the eave hide too much of the tower base? A single attractive perspective can conceal a misaligned plan or an incorrect height.
Use a ruler or calipers to check the model’s major dimensions, and note the drawing dimension beside each check. If a mismatch appears, identify whether it comes from scale conversion, an incorrect cut, or an unresolved drawing assumption. Fix one cause at a time. When you alter a part, recheck any connected edges: a change to the tower width can change the roof intersection, while a change to the roof rise can move the point where a tower cap meets the ridge.
Decide what to change—and what to leave out
A small study model is a decision aid, not proof that the full-scale construction will work. It can reveal proportion, alignment, and visual continuity, but a simple card model does not test structural capacity, waterproofing, drainage, or construction details. Keep conclusions within what the model can show. If the question is about how the form reads, compare the form; if it concerns full-scale performance, the model alone is insufficient.
Record the selected option with a short note or mark-up: which roof pitch or tower position you chose, what drawing information supports it, and which assumptions remain open. Then carry the settled dimensions into the next drawing or model. Add small details only when they help answer a remaining question—for instance, a cap or eave thickness may matter if it changes the apparent junction. The aim is to finish with a clear geometric decision and a compact record of why it was chosen.