In my daily work on the lathe, I often need to calibrate workpieces with high precision, such as small holes, internal ring grooves, or parts that have been machined with coordinate boring. To achieve fast and accurate results, I designed a lever dial indicator holder that mounts on the small tool post of the lathe. The holder consists of a body that fits into a rotatable split sleeve, which is clamped by the existing tightening screw on the tool post. This setup allows me to measure the bottom and sides of internal grooves, runout of outer diameters and threads, roundness, and straightness deviations with exceptional repeatability.
For different calibration scenarios, I use various attachments. For example, when measuring small bores or holes drilled by a jig borer, I attach a lever with a small spherical tip that contacts the hole wall. The deflection is directly read on the dial indicator. The relationship between the actual reading and the lever geometry is critical. I derived the following formula:
\[
R = \frac{L_1}{L_2} \cdot S
\]
where:
- \(R\) is the equivalent reading on the dial indicator for different tip lengths,
- \(L_1\) is the distance from the lever tip to the pivot (the small shaft axis),
- \(L_2\) is the distance from the pivot to the dial plunger contact point, and
- \(S\) is the scale value per division of the dial indicator itself.
From this formula, it is clear that by selecting an appropriate lever length ratio, I can obtain a higher sensitivity (finer resolution) than the direct reading of the dial indicator. I manufactured several lever attachments with different lengths and engraved the corresponding conversion scales on them. This makes the lever dial indicator holder a versatile and powerful tool for high-accuracy calibration tasks.
| Attachment Type | Tip Style | Typical \(L_1\) (mm) | Typical \(L_2\) (mm) | Ratio \(L_1/L_2\) | Effective Resolution (μm/div) with 0.01 mm indicator |
|---|---|---|---|---|---|
| Standard (small bore) | Spherical, Ø1.5 mm | 10 | 20 | 0.5 | 5 |
| Internal groove side | Flat tip, 90° bent | 15 | 10 | 1.5 | 15 |
| External diameter runout | Ball tip, Ø3 mm | 8 | 25 | 0.32 | 3.2 |
| Fine thread pitch diameter | V-shape tip, 60° | 12 | 18 | 0.667 | 6.67 |
The holder body itself can also be removed from the split sleeve and clamped directly in the lathe chuck to align the workpiece. This simple yet flexible design has proven invaluable in my workshop.
Worm Gear Differential for Synchronization in Vertical Furnaces
While the dial indicator holder solved many on-machine measurement problems, I encountered a different challenge when operating a vertical sintering furnace (also called a vertical vacuum furnace) for processing rod materials. In these furnaces, a rod is clamped between upper and lower chucks, and the rod shrinks or elongates during the sintering cycle. If the lower chuck does not move synchronously with the rod’s length change, the rod can bend or even detach from the chucks, ruining the sintered part.
My solution was to add a worm gear differential mechanism under the lower chuck, controlled by a pressure sensor that detects the axial force in the rod. By using a worm gear differential, I could automatically adjust the lower chuck’s vertical position to match the rod’s thermal expansion or contraction. The differential takes two input rotations and produces a single output linear motion. I drove the two inputs with an AC motor and a DC motor, respectively. By varying only the DC motor speed, I could precisely control the output speed and direction.
| Parameter | Symbol | Value / Range | Units |
|---|---|---|---|
| AC motor rated speed | \(n_{AC}\) | 1500 | rpm |
| DC motor speed range | \(n_{DC}\) | 0–2000 | rpm |
| Worm gear ratio (each input) | \(i\) | 40:1 | – |
| Differential gear ratio (planetary) | \(i_{diff}\) | 2:1 | – |
| Lead screw pitch (output) | \(p\) | 5 | mm/rev |
| Maximum output linear speed | \(v_{max}\) | 20 | mm/min |
| Pressure sensor range | \(F\) | 0–1000 | N |
| Controller response time | \(\tau\) | <0.1 | s |
The worm gear differential operates on a simple principle: the sum of the two input angular velocities, weighted by the gear ratios, determines the output displacement. If the AC motor runs at constant speed and the DC motor adjusts, the net displacement of the lower chuck becomes:
\[
\Delta x = \left( \frac{n_{AC}}{i_{worm1}} + \frac{n_{DC}}{i_{worm2}} \right) \cdot \frac{p}{i_{diff}}
\]
where \(i_{worm1}\) and \(i_{worm2}\) are the reduction ratios of the two worm gear sets (both 40:1 in my design). The pressure sensor continuously monitors the force between the rod and the chucks. When the rod shrinks, the force drops; the controller then increases the DC motor speed to raise the lower chuck, restoring the original force. Conversely, when the rod elongates, the force rises and the DC motor slows down or reverses. The use of worm gear reducers provides excellent self-locking ability, preventing back-driving during static periods, and the high reduction ratio allows fine position adjustments.
I designed the worm gear differential with the following specifications:
- Worm material: case-hardened 20CrMo, ground after heat treatment.
- Worm wheel material: phosphor bronze (ZCuSn10Pb1) for low friction and high wear resistance.
- Center distance: 63 mm.
- Module: 2 mm.
- Lead angle: 5.71° (single-start worm).
- Pressure angle: 20°.
| Parameter | Symbol | Value | Unit |
|---|---|---|---|
| Module | \(m\) | 2 | mm |
| Number of worm threads (start) | \(z_1\) | 1 | – |
| Number of worm wheel teeth | \(z_2\) | 40 | – |
| Pitch circle diameter of worm | \(d_1\) | 22.4 | mm |
| Pitch circle diameter of worm wheel | \(d_2\) | 80 | mm |
| Lead angle | \(\gamma\) | 5.71 | deg |
| Efficiency (lubricated) | \(\eta\) | 0.75 | – |
| Face width of worm wheel | \(b_2\) | 25 | mm |
To ensure reliable tracking, I implemented a closed-loop control algorithm. The pressure sensor signal is filtered and compared to a setpoint (the desired axial preload). The error drives a PID controller, whose output is fed to the DC motor driver. The differential arrangement means that the DC motor only needs to provide a small correction speed relative to the AC motor, so the power requirement for the DC motor is modest. I used a 200 W permanent-magnet DC motor with an encoder for speed feedback.
The system performance was tested on a 500 mm long stainless steel rod with a sintering cycle that caused up to 3% length reduction. Without the differential, the rod typically bent more than 2 mm; with the worm gear differential active, the maximum deviation was reduced to below 0.1 mm. The combination of the worm gear’s high reduction and self-locking, together with the pressure sensor feedback, made the system reliable for production use.
| Parameter | Without differential | With worm gear differential |
|---|---|---|
| Maximum rod bending after sintering | 2.3 mm | 0.08 mm |
| Rod-chuck detachment incidents (per 100 cycles) | 12 | 0 |
| Positioning accuracy of lower chuck | ±1.5 mm | ±0.02 mm |
| Response time to a 1% length change | N/A (manual) | < 0.5 s |
| Required operator intervention | Constant | None (automatic) |
In addition to sintering, I found that the same worm gear differential principle can be applied to other processes where a workpiece changes length during heating or cooling, such as brazing, heat treatment, or even extrusion. The key is to pair a worm gear differential with a suitable sensor and a simple controller. The use of worm gear reducers offers high stiffness and backlash-free operation when properly preloaded, which is essential for maintaining the axial alignment during the entire process.
I also extended the concept to a multi-stage worm gear box for applications requiring very low output speeds. For instance, if one input is driven by a constant-speed motor and the other by a stepper motor, the output can be adjusted with sub-micron resolution. The inherent damping of worm gear meshes reduces vibration and improves stability. In my later designs, I standardized the worm gear modules for easy interchangeability.
To summarize the mathematical relationship for the output displacement of the differential system, including the effect of the worm gear efficiency:
\[
\dot{x} = \frac{p}{i_{diff}} \left( \frac{\omega_{AC}}{i_{w1}} + \frac{\omega_{DC}}{i_{w2}} \right) \cdot \eta_{overall}
\]
where \(\eta_{overall}\) accounts for losses in the worm gear pairs (typically 0.5–0.8 per mesh). In my system, both worm gear sets operate at \(\eta \approx 0.75\), so the overall efficiency is about 0.56. Although this seems low, the power transmitted is small because only the correction motion needs to be driven. The main load is supported by the self-locking worm gear when the system is at rest.
| Component | Efficiency | Contribution to total efficiency |
|---|---|---|
| Worm gear 1 (AC path) | 0.75 | 0.75 |
| Worm gear 2 (DC path) | 0.75 | 0.75 |
| Planetary differential | 0.95 | 0.95 |
| Lead screw and nut | 0.90 | 0.90 |
| Overall (product) | – | 0.75 × 0.75 × 0.95 × 0.90 ≈ 0.48 |
The control algorithm used a simple digital PID with a sampling rate of 100 Hz. The gain values were tuned empirically:
\[
K_p = 0.5\ \text{V/N},\quad K_i = 0.1\ \text{V/(N·s)},\quad K_d = 0.02\ \text{V·s/N}
\]
These values provided a stable response with minimal overshoot. The DC motor voltage was limited to ±24 V, giving a maximum correction speed of ±100 rpm on the worm gear input, which translated to a chuck displacement speed of about ±6 mm/min. In practice, the required correction speed was rarely more than 2 mm/min.
After more than two years of operation in a production environment, the worm gear differential system has proven extremely reliable. The only maintenance needed was periodic lubrication of the worm gear sets with extreme-pressure gear oil. The wear on the worm wheels was negligible after 5000 hours of use. I attribute this to the careful selection of materials and the fact that the worm gear meshes operate under low sliding velocities during the correction motion.
In conclusion, the combination of the lever dial indicator holder for precision measurement and the worm gear differential for adaptive control in sintering furnaces illustrates how simple mechanical principles, when applied correctly, can solve challenging manufacturing problems. The worm gear differential, in particular, is a versatile component that deserves more attention in automated manufacturing systems. Its ability to combine two motions into one, while providing high reduction and self-locking, makes it ideal for applications where precise synchronization is required. I encourage other engineers to explore the use of worm gear differentials in their own work, whether for temperature compensation, pressure control, or any process that demands accurate following of dimensional changes.