How to Print and Read a Temperature Tower (Step by Step)

Every spool of filament has a temperature sweet spot. The label says "190-220°C" but that's a range, not an answer. Print too cold, and you get weak layer adhesion and under-extrusion. Print too hot, and you get stringing, oozing, and blobby surfaces. The temperature tower tells you exactly where your specific filament performs best — in about 45 minutes and a few grams of plastic.

I run a temp tower for every new spool I buy. Yes, even if I've used the same brand before. Different colors from the same manufacturer can have different optimal temperatures by 5-10°C because the pigments affect the melt behavior. Let me walk you through the whole process.

What a Temperature Tower Tests

A temperature tower is a single print with multiple sections, each printed at a different temperature. The model is designed so each section includes features that are sensitive to temperature changes:

• Bridging — horizontal spans between two points with no support underneath
• Overhangs — angled surfaces at 30°, 45°, and 60°
• Stringing — thin pillars with gaps that show retraction performance at temperature
• Fine detail — small text or geometric features
• Surface finish — flat walls that reveal layer consistency

Each section is labeled with its temperature, so when the print finishes you can physically inspect each one and compare.

Step 1: Download the Model

Search Thingiverse or Printables for "temperature tower" — there are dozens. I recommend the "Smart Compact Temperature Tower" by gaaZolee on Printables. It's compact (uses minimal filament), includes bridging and overhang tests, and has clear temperature labels.

For PLA, grab the version that covers 190-220°C in 5°C increments. That gives you 7 sections: 220, 215, 210, 205, 200, 195, 190. Most PLA lands somewhere in the 200-215°C range, but you'd be surprised how often the "wrong" temperature is where a filament shines.

Step 2: Set Up in Cura (Change at Z Method)

Here's the setup if you're using Cura:

1. Import the temperature tower STL into Cura
2. Set your starting temperature to the highest value (220°C for PLA)
3. Go to Extensions → Post Processing → Modify G-Code
4. Click Add a script → ChangeAtZ
5. Set "Change Height" to the Z height where the second section starts (usually 8.2 mm for most temp tower models — check the model description)
6. Set the new temperature to 215°C
7. Repeat: add another ChangeAtZ script for each section — 210°C at 16.4 mm, 205°C at 24.6 mm, etc.

The exact Z heights depend on your specific model. Each section is typically 8.0-8.2 mm tall. Count the sections, check the layer height, and calculate. With 0.2 mm layers, each section is 40-41 layers.

Step 3: Print Settings

Use your normal print settings for everything except temperature. The point is to isolate temperature as the only variable:

• Layer height: 0.2 mm (standard)
• Speed: 50-60 mm/s (your normal speed)
• Retraction: Keep your current settings — don't change them for this test
• Bed temp: Your normal PLA bed temp (60°C is standard). If you're testing PETG, check your PETG bed temp settings
• Cooling: 100% fan after layer 3 (normal PLA settings)
• Infill: 10-15% is fine — this is a test, not a structural part

The print should take 35-50 minutes depending on your speed and the tower model. Use minimal filament — usually 8-15 grams.

Step 4: Reading the Results

This is the critical part. Once the print finishes and cools, inspect each section systematically. Here's what to look for:

Bridging

Look at the horizontal bridge on each section. A good bridge is flat, with no drooping in the middle. Too hot and the middle sags visibly. Too cold and the filament doesn't stick to the anchor points and you get a messy gap. Pick the section where the bridge is flattest.

Stringing

Check the gaps between the small pillars. You want zero strings (thin filament hairs) between them. Higher temperatures cause more stringing because the melted filament is less viscous and oozes during travel moves. If every section strings, your retraction settings also need tuning — but that's a different calibration.

Overhangs

Look at the angled surfaces. The 45° overhang should be clean at almost any temperature. The 60° overhang is the real test — too hot and it droops, too cold and the layers don't bond properly and you get curling. Find the temp where the 60° overhang is smoothest.

Layer Adhesion

Try to snap each section by bending it. Higher temperatures produce better layer adhesion because the new layer bonds more thoroughly to the one below. If you're printing functional parts that need strength, prioritize the section that's hardest to break — even if the surface finish isn't quite as pretty.

Surface Finish

Inspect the flat walls. Consistent, smooth layers with no blobbing or rough patches indicate good temperature. Too hot and you'll see shiny spots and blobs. Too cold and you'll see matte, slightly rough texture from incomplete inter-layer flow.

Interpreting Conflicting Results

Here's the real-world scenario: 210°C gives the best bridging, but 215°C gives the best layer adhesion, and 205°C has the least stringing. Which do you pick?

For functional parts (brackets, mounts, tools): prioritize layer adhesion. Go with the higher temperature.

For visual/display models (figurines, decorative items): prioritize surface finish and minimal stringing. Go with the lower temperature.

For general purpose: pick the temperature in the middle of the range that performs well across all categories. Usually that's 5°C above the best stringing temperature.

Once you find your temperature, write it on the spool with a marker. Seriously — future you will thank past you. And pair this with the right filament from our PLA brands guide to get consistently excellent results.