Transmission Endplay
Dial Indicator Procedure

Endplay is the first number you should take on a core and the last number you should confirm before the pump goes back on. Skip it and you are guessing about the condition of every thrust surface inside the unit. Take it wrong and you will chase a noise that was never there. Here is the bench procedure, the numbers that mean something, and the four ways techs blow the reading without knowing it.

What Endplay Actually Is

Endplay is axial float — how far a rotating member slides fore and aft along its own centerline before something stops it. In an automatic, two different measurements get called "endplay."

Front / Input Shaft Endplay

Total axial travel of the input (turbine) shaft and the drum stack behind it, measured between the pump thrust washer at the front and whatever surface stops the stack at the rear. This gets checked on nearly every longitudinal unit and tells you the most about internal wear.

Rear / Output Shaft Endplay

Travel of the output shaft in the tail housing, controlled by the rear thrust washer, rear bearing, and in some families a snap ring. Less commonly checked, but worth pulling on a unit with driveline vibration or a chewed rear bushing.

Both numbers are the sum of clearances across every thrust washer, bearing, and selective in that load path. That is why they are useful: one measurement, 90 seconds, sums up wear you would otherwise have to tear down to see.

Why It Matters

• Thrust washer wear. Fiber and bronze washers wear thin, and every thousandth they lose shows up as endplay. It is the cheapest early warning of a starved or overheated unit you will get.
• Pump bushing and converter hub. Excess float lets the converter hub hammer the pump bushing. Once that bushing goes oval, pressure leaks past and you get a soft, hot unit that never quite makes line pressure.
• Converter thrust load. A converter under lockup pushes hard forward against the pump. With too much stack clearance, that load lands on the wrong surfaces and eats them.
• Assembly verification. On the build side, endplay proves the correct selective went in and nothing is sitting crooked in a groove. A stack that assembles at zero is telling you something before you ever bolt a pan on.

When You Check It

Twice. Non-negotiable.

Before teardown, as a diagnostic. Core comes in, converter off, unit fixtured. Take the reading and write it on the case in grease pen. That number frames the whole teardown: high tells you where to look first, at spec tells you the wear is somewhere else.

At final assembly, before the pump goes on. Stack the unit with the new washers, set the pump gasket, and verify. This is where a selective gets swapped. Find it wrong after the pump is bolted and torqued and you are pulling it back off.

The Procedure, Step by Step

You need a dial indicator with at least.200 in of travel, a solid magnetic base, and both hands free — meaning the case is clamped or fixtured, not rocking on a bench.

1. Pull the converter. It preloads the stack and hides the float. Converter off, always.
2. Clean the mounting surface. Wire brush a spot on the bellhousing face or case boss down to bare metal. Paint, ATF film, and casting sand cost you magnet grip and add motion you will read as endplay.
3. Mount the magnetic base on that spot, as close to the shaft as the arm allows. Every extra inch of arm is extra deflection.
4. Set the tip on the end of the input shaft, dead center, plunger parallel to the shaft axis. If the shaft end is drilled or chamfered, use the turbine hub shoulder instead. Preload about.050 in so the plunger can travel both ways.
5. Push the shaft fully rearward, hold it, and zero the dial.
6. Lift the shaft forward. Two long screwdrivers under the drum, prying evenly on opposite sides, is the classic method. A short blast of shop air through the front pump feed works on some families and is easier on the parts. Pull straight, not sideways.
7. Read total travel — that is your endplay.
8. Repeat three times and average. A spread over about.003 in means something is moving that should not be, usually the base.

Dial indicator with magnetic base sets

Transmission holding fixtures

Typical Spec Ranges

The factory service manual for your specific unit is the only authority. Specs vary by family, by year, and sometimes by pump casting within a family. The table below is shop rule-of-thumb — useful for judging a core, useless as a build spec.

Ranges that wide reflect how much real variation exists between manuals. Look yours up, then measure.

Reading the Number

High Reading

Over spec means material is missing from the thrust stack. In descending order of likelihood: worn or partially wiped fiber washers, a flattened thrust bearing race, a worn pump bushing letting the whole stack shift, and a converter hub that has worn its own thrust face. A reading.015 in over is not "close enough" — that unit ran metal-to-metal somewhere, and the debris is in your valve body.

Zero or No Movement

Zero is worse news than high, not better. Check the obvious first: converter still in, tip riding a chamfer instead of the shaft. Then the real causes — something binding or cocked in a groove, a snap ring out of its land, or a too-thick selective from a previous build. On a core with genuine zero front endplay, expect galled thrust surfaces from a stack that had no oil clearance to feed.

Snap ring plier sets

Correcting It With Selective Washers

Endplay is not adjusted by torque or by shimming the pump. You change the thickness of one selective washer in the load path — usually the front pump-to-drum thrust washer. The math is direct: measured minus target equals the thickness you add.

Selectives come in kits stepped roughly.005 to.010 in apart. Mic the washer you pull and the one you install — do not trust the color code or the box. Aim for the middle of the spec range, not the edge, so the unit keeps margin after break-in.

Selective thrust washer kits

Feeler gauge sets

Four Mistakes That Give You a False Number

• Tip not parallel to the shaft axis. A tip 20 degrees off-axis reads roughly 6 percent low and gets worse fast past that. If you cannot get square, move the base — do not fudge it.
• Magnetic base on paint, grime, or rough casting. The magnet feels solid and still creeps under pry load. Every thousandth of base movement lands in your reading.
• Checking with the converter installed. It preloads the stack forward and masks most or all of the float. This is the most common way a shop reports "zero endplay" on a unit that is actually loose.
• Mistaking bearing preload for endplay. If the shaft feels springy instead of hitting a hard stop, you are compressing a wave washer or loading a tapered bearing, not reading free play. Push to a definite stop both directions and watch the needle quit climbing.

FAQ

Can I check endplay with the transmission still in the vehicle?

No. The converter is bolted to the flexplate and the bellhousing is buttoned to the block, so you have neither access to the shaft end nor a free stack to move. Endplay is a bench measurement, converter removed.

Do I need a dial indicator, or will feeler gauges work?

Feeler gauges have a place — clutch pack clearance, checking washer thickness — but they cannot measure axial float in an assembled stack. A basic.001 in resolution indicator with.200 in travel and a decent magnetic base does everything here.

My three readings are all different. What now?

Spread means motion in your setup, not in the transmission. Check base grip first, then the arm joints (a loose swivel knuckle is the usual culprit), then whether the case is rocking. Fixture it, retighten every joint, re-measure.

How much over spec is worth rebuilding for?

Endplay alone does not decide it — it points you where to look. A core.015 to.020 in over spec needs front thrust washers and a pump bushing at minimum. Drop the pan and read the fluid before quoting. Excess endplay plus metal in the pan is a full rebuild, not a washer swap.