Technical Guide
Measuring Slideway Friction: A Pull-Test Method for Machine Ways
Friction on a machine slide is usually argued about rather than measured. A controlled pull test replaces the argument with two numbers worth keeping - the force needed to break the slide away and the force needed to keep it moving - recorded at fixed stations along the travel. This guide sets out the setup, the arithmetic and the record sheet, and is equally clear about what the result does not prove.
Direct answer: pull the slide slowly and record two forces
Measure friction on a machine way by pulling the slide slowly and steadily along its direction of travel with a calibrated load cell or a spring balance in line with the motion, and record two figures at each station: the peak force at which the slide breaks away, and the force that keeps it moving once it has started. Divide each force by the normal load carried on the sliding faces - mass multiplied by 9.81 for a horizontal slide - to obtain an apparent friction coefficient. Repeat at the same stations, in both directions, at the same temperature and with the same oil state, and compare the machine with its own earlier record rather than with a catalogue figure.
That last sentence is the whole method. A pull test produces a comparator, not a material constant. Two readings taken on the same machine, at the same station, with the same instrument and the same lubrication are directly comparable; a reading taken on a different machine, or after the wipers, screw or oil have changed, is not.
Why the test needs two forces, not one
A single drag figure collapses two different behaviours into one number. What an operator feels as a heavy slide may be steady resistance, and what feels light may still hesitate at the start of a move. Separating breakaway from running force keeps those two conditions apart, and their difference is the friction step that feeds low-speed stick-slip.
| Record | What it describes | What it usually points to |
|---|---|---|
| Breakaway high, running close behind it | The slide resists movement consistently | General drag: oil not reaching the face, stiff wipers, covers, or a tight gib along the whole travel |
| Breakaway high, running much lower | The slide releases once movement starts | A friction step at start-up: oil squeezed out at rest, a local tight spot, or a running face that is no longer fully separated |
| Both figures rise together over weeks | The whole assembly is getting heavier | Contamination ingress, oil breakdown, wiper damage or progressive liner wear |
| One station high in both directions | The resistance is local | A high spot, a step in the way, or a chip held under a wiper |
| One direction higher than the other | The slide is being wedged | Gib taper, alignment, or an adjustment taken up at one end only |
What the pull force actually includes
The force at the attachment point is the sum of everything that resists the movement, and the way contact is only one term in it. Before the first reading, decide what the record is going to describe, because a number that mixes two definitions cannot be compared with anything later.
- Way friction at the bearing faces, which is the term the test is usually trying to isolate.
- Wiper drag, which rises sharply when a wiper stiffens, swells with coolant, or packs with chips.
- Gib preload, which adds normal force to the gib face without appearing in the slide mass.
- Screw and nut drag when the drive is still connected, plus any preload in the thrust bearing.
- Way covers, cable chain and guard seals, which add a resistance that changes with position.
Instruments and setup that hold up in a workshop
The instrument matters less than the consistency of its use. Choose one that puts the readings in the upper half of its range, record its resolution, and use the same instrument for the before and after records.
| Method | What it resolves | Use it when | Limitation |
|---|---|---|---|
| Load cell with peak hold and a digital display | Breakaway peak and running force as separate values | The drive can be decoupled and the pull line kept straight | Needs a fixture and a second person to watch the display during the pull |
| Spring balance with a maximum pointer | Breakaway peak directly, running force by eye | A quick before and after comparison on the same machine | Pointer friction, pull speed and parallax all affect the reading |
| Lever or screw pull with a load cell in line | A steady, repeatable pull speed | Heavy slides and stations that need several passes | The fixture has to pull parallel to the way, or it wedges the slide |
| Drive load monitor on a slow traverse | Relative force trend during motion | The axis cannot be decoupled without a major strip-down | Includes screw and guide losses, and the control loop filters the breakaway peak |
Attach the pull at one defined point, in line with the direction of travel and at about the height of the screw axis so that the pull does not rotate the slide into its gib. Mark the stations on a fixed reference - both ends of the stroke and the quarter points are enough for most ways - so that the same point is measured every time. Zero the instrument with the line slack, and check that the reading returns to zero after each pass.
The procedure
- Lock out the drive and confirm that the slide can be moved by hand. On a vertical or inclined axis, support or brake the member before releasing anything.
- Clean the exposed way and the wipers, and record whether the wipers were in place or lifted during the test.
- Bring lubrication to its normal working state, then record the oil grade, whether the face is flooded or wiped, and the temperature at the machine.
- Mark the measuring stations on the way or on a fixed reference, and keep the same stations for every later record.
- Attach the pull at the defined point and check that the line runs parallel to the travel.
- Zero the instrument, then pull slowly and steadily until the slide moves. Do not snatch at the line.
- Record the breakaway peak and the force that keeps the slide moving over the first part of the stroke.
- Repeat three passes in each direction at each station, keep the highest peak and the median running force, and note if the passes scatter.
- Write down temperature, oil state, wiper state and any noise or jerking observed during the pass, next to the numbers rather than in a separate note.
- Photograph or copy the record before the machine is stripped, because the baseline cannot be recreated afterwards.
The arithmetic
For a horizontal slide carrying its mass on the measured faces, the apparent friction coefficient is the pull force divided by the normal load: mu = F / (m x 9.81). Two caveats belong with the formula rather than after it. Where the load is shared between several faces, the normal load on each face is a fraction of the weight. Where a gib is preloaded, the normal force on the gib face is not the weight at all, and no pull test can separate that component from the total.
| Line | Worked example | Where the number comes from |
|---|---|---|
| Slide mass carried on the measured faces | 1,250 kg | Machine data or a weighing record |
| Normal load, m x 9.81 | 12,263 N | 1,250 x 9.81 |
| Breakaway force at station 3 | 410 N | Peak-hold reading |
| Apparent coefficient at breakaway | 0.033 | 410 / 12,263 |
| Running force at station 3 | 300 N | Reading once motion is established |
| Apparent coefficient while running | 0.024 | 300 / 12,263 |
| Friction step, breakaway minus running | 110 N, or 0.009 | 410 - 300 |
A record sheet for the travel
Five stations and both directions are enough to show whether resistance is uniform, local or directional. The example below uses the same invented inputs as the arithmetic above, with one deliberately high station to show how a local fault looks in the table.
| Station | Direction | Breakaway, N | Running, N | Difference, N | Observation |
|---|---|---|---|---|---|
| S1, start of stroke | Push | 395 | 290 | 105 | Smooth |
| S1, start of stroke | Pull | 405 | 296 | 109 | Smooth |
| S2 | Push | 400 | 292 | 108 | Smooth |
| S2 | Pull | 412 | 300 | 112 | Smooth |
| S3, mid stroke | Push | 402 | 294 | 108 | Smooth |
| S3, mid stroke | Pull | 410 | 302 | 108 | Smooth |
| S4 | Push | 468 | 340 | 128 | Slight hesitation |
| S4 | Pull | 475 | 348 | 127 | Slight hesitation |
| S5, end of stroke | Push | 398 | 288 | 110 | Smooth |
| S5, end of stroke | Pull | 406 | 294 | 112 | Smooth |
Reading the pattern
The absolute figures are the least interesting part of the record. What earns its place in a rebuild report is the shape of the pattern: whether resistance is uniform, local or directional, and whether the breakaway-to-running gap is stable.
| Pattern in the record | What it points to | Next check |
|---|---|---|
| All stations high, gap small | General drag rather than a friction step | Oil delivery at each station, oil grade, wiper condition and cover drag |
| One station high in both directions | A local geometry or contamination fault | Contact pattern and straightedge at that station, and the wiper covering it |
| One direction consistently higher | The slide is being wedged in one direction | Gib clearance at both ends of the travel and the taper contact |
| Breakaway much higher than running | A friction step at start-up | Oil state at rest, gib preload and local tightness at that station |
| Readings climb over weeks with unchanged load | Progressive contamination, oil breakdown or liner wear | Running face condition, wiper sealing and oil cleanliness |
| After relining, worse than the baseline | Something in the rebuild changed the oil path or the contact | Whether the oil grooves were reproduced in the liner, finish of the running face, contact pattern |
Engineering judgement: what the numbers are allowed to decide
- Compare the machine with itself. Absolute coefficients differ with mass, wiper drag, screw preload and oil grade, so a figure from another machine says nothing about this one.
- Agree the acceptance target before the rebuild: a lower breakaway force, a smaller breakaway-to-running gap, and evenness across the stations. Without a baseline the after-reading has nothing to be measured against.
- Do not turn the apparent coefficient into a material claim or a pressure limit. Pressure is a separate calculation on the real bearing area, set out in the slideway liner load and pressure guide.
- Test with the lubrication working. A dry pull measures a different system and cannot be compared with a flooded reading.
- Watch the pull for a sawtooth pattern rather than a steady rise. A force that builds and releases repeatedly is the signature of the friction step described in the stick-slip guide, and no single average figure captures it.
- Confirm that the oil grooves still feed the face after relining. A liner bonded over a groove blocks delivery and produces a reading that climbs from one end of the stroke to the other; the slideway lubrication and oil groove guide covers the groove patterns and oil matching.
- Repeat the test at the same stations with the same instrument after the rebuild. The checks in the accuracy after relining guide use the same station-based record, so the two sets of readings can be read together.
What the test cannot tell you
A pull test measures resistance to motion. It does not show where the liner is bearing, how much pressure each face carries, whether the way is straight, or whether the finished geometry will hold position. Those questions need their own checks: spotting paste for contact, a straightedge or indicator for geometry, and a pressure calculation for the bearing area. Use TEBAK Tuschierpaste for the contact check, and treat the pull record as one column in the rebuild file rather than as the verdict.
Frequently asked questions
Can slideway friction be measured without a load cell?
Yes. A spring balance with a maximum pointer gives a usable breakaway peak, and the running force can be read by eye while the slide moves. Keep the attachment point, the pull direction and the stations identical between records, and treat the result as a comparison rather than as a calibrated coefficient.
Should the test be done dry or with lubrication?
With lubrication in its normal working state, because that is the condition the machine runs in. Record the oil grade, whether the face was flooded or wiped, and the temperature, so that a later reading taken under different conditions is not mistaken for a change in the machine.
Why is my calculated coefficient higher than the material data sheet?
Because the pull includes wiper drag, gib preload, screw losses and cover resistance, and because the normal load on a preloaded gib face is not the slide weight. The calculated figure is an apparent coefficient for the assembly, not a property of the liner.
Does a low pull force prove the liner is healthy?
No. A low force shows low resistance to motion in the condition tested. It does not show full contact, correct pressure distribution or a sound bond. Read it together with the contact pattern and a pressure check before accepting the repair.
How often should the record be repeated?
At least three times in a rebuild: before strip-down as a baseline, after cure and finish machining, and after the first period of production. After that, repeat it at planned maintenance so that a rising trend is caught before it becomes a positioning problem.
Send the record with the specification
For a specification review, send the machine make, model and axis, the way type and bearing width, the slide mass and how the load is shared, the stations used, the breakaway and running force at each station in both directions, the instrument and its resolution, and the oil grade, oil state and temperature at the time of the test. Add the contact pattern and the pressure calculation if they exist, because the liner decision depends on all three. The TEBAK range covers the Blue TEBAK slideway liner, the Green TEBAK slideway liner and the other colours in the same compound, the matching two-part epoxy adhesive and the Tuschierpaste used to check contact. Tell us the finished thickness and width the job needs, and we will confirm the closest stocked size against the record you send.