Linear guide selection is where a lot of automation projects quietly lose precision. Teams lock in stroke and speed first, then pick a rail size that "usually works", and only discover at assembly that the carriage does not fit, the accuracy is not there, or the rail starts knocking long before its rated life. Choosing a linear guide rail properly comes down to three questions: how accurate does the machine have to be, how does the load actually travel, and where does the rail sit. This guide walks through the three steps we use at Zhongbai, and explains how our HG, EG, RG, MG and WE series are positioned.
1. Answer three questions before you open a catalogue
Write these three sentences down first. Everything downstream is derived from them.
- How accurate? A pick-and-place transfer and a micro-level inspection stage are different problems. The accuracy target decides the precision grade and preload, and it also decides whether your mounting surface has to be ground.
- How does the load travel? Load sitting directly on top of the carriage behaves very differently from load cantilevered off the side or hanging underneath. Same weight, different arm, different result.
- Where does the rail sit? Horizontal table, vertical column, or an inverted gantry — the mounting orientation changes the machining of the datum surface, the fixing direction, and how serviceable the axis will be.
None of these need numbers, but they decide which parameters you go and look for. Skip them and you end up comparing rated load figures that do not describe your application.
2. Turn the working conditions into a table
Fill in the inputs first, then use the right column as your screening criteria.
| Selection dimension | Input you must confirm | How to judge |
|---|---|---|
| Stroke | Effective stroke, overall installed length, available space | Stroke comes first; then fix rail length and carriage count. Tight envelopes favour low-profile series. |
| Load | Load weight, centre of gravity position, direction, off-centre condition | The further the centre of gravity from the carriage, the larger the moment. Never look at weight alone. |
| Speed & acceleration | Target speed, acceleration, cycle rate, duty profile | Frequent starts and stops make you verify life and re-lubrication intervals together. |
| Precision | Repeatability, positioning accuracy, structural rigidity | Work backwards from machine accuracy, not forwards from rail accuracy. |
| Mounting | Orientation, datum surface finish, clearance for motor and sensors | Inverted and side-mounted axes need the moment load checked separately. |
| Environment | Dust, cleanliness, noise, protection, maintenance access | Dusty areas need wipers; cleanrooms need low-sling lubricant. |
Once the table is filled, the candidate range usually narrows to one or two series. If a row stays blank, the machine design is not settled yet — finish that before selecting rails.
3. Load and moment: rated load is not usable load
Catalogue ratings are measured under defined conditions and normally have to be discounted for real use. Two things decide whether a rail size is actually adequate:
- Direction. A linear guide carriage can take load up, down, left and right — but the allowable value is not the same in every direction. A side-loaded axis cannot simply borrow the downward rating.
- Moment. When the centre of gravity sits away from the carriage centre, you get moments about three axes. Moment overload is the most common cause of early rail failure, and it is harder to spot than plain overloading because the axis can still feel smooth by hand.
In practice we estimate the worst-case static combination first, then re-check against the dynamic condition with real acceleration. Only when both keep a margin do we confirm the model against the catalogue. Our linear guide rail product page lists the six dimensions — stroke, load, speed, precision, mounting, environment — as a self-check before you get to model numbers.
4. Precision and preload: work backwards from the machine
Precision is the parameter most often over-specified. Over-specifying does not just cost money: it raises the machining requirement on the mounting surface, makes assembly harder, and can even introduce stick-slip. A workable sequence:
- State the machine's required repeatability and positioning accuracy.
- Subtract the error contributed by the drive train — belt tension variation, screw lead error, and so on.
- The remainder is what the rail is allowed to contribute; pick the precision grade from that.
- Then decide whether you need zero clearance or a preload class, based on rigidity requirements.
Zhongbai supplies HG, EG, RG, CG, SE, MG and WE series. It helps to consider the rail and the ball screw together: the screw drives, the rail guides, and if their accuracy grades are mismatched the assembly ends up limited by the weaker one.
5. Mounting: horizontal, side-mounted, inverted and multi-axis
The same rail behaves differently depending on how it is fixed. The common arrangements:
- Horizontal — rail on a flat datum, load pressing down. Simplest load case and by far the most common.
- Side-mounted — rail fixed to a vertical face or column, carriage carrying a moment. Squareness of the datum and the fixing method matter much more.
- Inverted — rail fixed overhead with the load hanging below, so the dead weight becomes a continuous overturning moment.
- Inclined — between horizontal and side-mounted; resolve the load by the actual angle.
- Twin parallel rails — here the accuracy that matters is not the individual rail but parallelism and height matching. Mismatched datums cause binding, which shows up as resistance that varies along the stroke.
- Multi-axis assemblies — on XY, XYZ and gantry structures, the rigidity of the base axis often sets the dynamic behaviour of the whole machine.
Three rules apply in every case: locate the reference side against a shoulder and leave adjustment on the other; tighten bolts in a diagonal sequence in gradual steps to avoid locked-in stress; and push the carriage through the full stroke by hand afterwards to confirm there is no binding or sudden change in resistance. For long-stroke structures where mounting datum design is the critical issue, the same thinking applies to our gear rack module range.
6. Series positioning and where to get data
| Series | Positioning | Typical use |
|---|---|---|
| HG | Heavy duty | Machine tools, heavy transfer, gantry structures |
| EG | Low profile | Automation equipment with limited mounting space |
| RG | Torque resistant | Off-centre loads and higher moment conditions |
| MG | Miniature | Small inspection equipment, precision stages |
| WE | Wide type | Single axis needing higher anti-overturning capacity |
| CG / SE | Additional series | Scope and parameters per catalogue and engineering drawing |
Full model ranges, mounting dimensions and accuracy grades are listed in the catalogue. Download the Zhongbai product catalogue to compare series pages, or send us your completed working-condition table and our selection engineers will match the model to your structure.
Keep the sequence in order: confirm the three questions, fill the parameter table, estimate load and moment, derive precision and preload from the machine accuracy, then check the structure against the mounting orientation. Every later step then has an input from the step above, so any change can be traced through its consequences. Doing it the other way round, model first and working conditions later, usually means solving assembly problems by re-machining the structure.
Related reading
- Belt-driven vs ball screw linear modules: a selection comparison
- Gear rack and pinion vs ball screw for long-stroke linear motion
Ready to check a configuration? Contact our selection engineers with your stroke, load and accuracy target — we will come back with a matching rail and module combination, plus engineering drawings and parameter tables.