Bench Testing Transmission Solenoids: Resistance and Flow

A solenoid pack comes across the bench off a 68RFE that was dropping to limp. Every coil ohms dead-on spec. The customer gets the pack back, it goes in the truck, and the same complaint walks through the door six weeks later.

That is not bad luck. It is the predictable result of using one test on a part that has two separate ways to fail. A transmission solenoid is an electromagnet bolted to a hydraulic valve. The ohmmeter looks at the electromagnet. Nothing in that reading tells you whether the valve inside is sealing.

Sonnax states the limit outright in the instructions for their own solenoid test manifold kit: the manifold "can prove or disprove internal mechanical sealing ability only; it does not test the integrity of the electrical coil and contacts," and they note that while internal sealing problems are the leading cause of failure for that solenoid family, electrical testing should also be carried out. Read that in both directions and you have the argument for a real bench procedure. Neither test covers the other.

---

Test 1: Resistance, and what it is actually good for

Resistance is not a quality test. It is an identity check and a gross-fault check, worth doing first because it takes fifteen seconds and catches open coils, shorted windings, and the solenoid somebody put in the wrong hole.

That last one matters more than people expect. On the 45RFE / 545RFE / 68RFE pack, six of the seven solenoids read the same, so the ohmmeter separates the pressure control solenoid from everything else and nothing more. Sonnax's published identification data:

Two things to bank from that table. First, the rest state column is the part techs skip, and it is what makes a no-move complaint make sense: multi-select and underdrive are normally applied, so a dead circuit to those two does not look like the others. Second, the overdrive solenoid was eliminated in the 2010 through 2018 version of this valve body, and case connector pin 7 — overdrive solenoid control — is not used on those. If you are pin-probing a late pack against an early diagram you will chase a circuit that does not exist.

At 1.3 ohms your meter leads and the contact resistance at the terminals are a real percentage of the reading. Short the leads together, note the value, subtract it. Skip that and a good solenoid reads 1.7 and goes in the scrap bin.

Temperature is part of the spec

Copper resistance rises with temperature, and the change is measurable at the temperatures a transmission actually lives at. Garrett Herning of Hydra-Test USA, writing in ATRA's GEARS, puts it plainly: the resistance of a solenoid coil will measurably change based on temperature, room temperature compared to 200 F. Published specs are room-temperature specs. A solenoid you pulled out of a unit ten minutes ago will read high, and a solenoid that only misbehaves hot will read fine on a cold bench. That gap is where most "it tested good" comebacks live.

Test 2: Load test with an ammeter

Circuits that pass a resistance check and fail under load are a known category. As Gregg Nader of the Sonnax TASC Force wrote in Transmission Digest, load testing stresses the whole circuit the way it works in operation, and it is the fastest and most effective way to diagnose an electrical circuit problem. On the bench, energize the solenoid through an ammeter and read the current it actually draws.

The math is Ohm's law and the numbers are easy to hold in your head. A 24-ohm solenoid draws about half an amp. A 12-ohm solenoid draws about one amp. To back into exact circuit resistance, measure your actual supply voltage and divide by measured current: 13.88 volts divided by 1.13 amps equals 12.2 ohms of circuit resistance. Compare that to the solenoid spec and the difference is everything else in the path — terminals, leads, connector.

The mistake that costs you a solenoid

Do not hang 12 volts directly on a low-resistance solenoid. Nader is specific about it: resistance in PWM and EPC solenoids can be as little as 2 ohms, and a 12-volt circuit through a 2-ohm solenoid draws 6 amps — enough to power the headlights, and enough to damage wiring and supply circuits. Every solenoid in that 68RFE table except the pressure control is well under 2 ohms.

The fix is a current-limiting probe. Nader describes a test light modified by replacing the bulb with a common 10-ohm, 10-watt resistor, used as a lead extension between the ammeter and the circuit. Add 10 ohms to a 2-ohm solenoid and you have 12 ohms total, so a 12-volt supply now moves 1 amp instead of 6. Measure current through the probe, apply Ohm's law to get total resistance, then subtract the 10 ohms you added to get the circuit value. Build one out of a dead test light and a resistor from the parts drawer and it lives in your bench drawer forever.

Reading the current signature

If you have a low-current probe and a scope or a graphing meter, the current trace tells you more than the number. Herning's breakdown of the two solenoid families is the frame to work from: on/off solenoids are usually greater than 10 ohms and draw around 0.4 to 0.7 amps; regulating solenoids are generally less than 10 ohms and can range from 0 to just over 1 amp.

On a regulating solenoid, the real test is a sweep. Ramp current from zero to max — roughly 1 to 1.3 amps — while watching outlet pressure, then ramp back down. What you are hunting for is not a single bad reading:

• Dead spots. A range of current where outlet pressure stops responding.
• Hysteresis. Different outlet pressure at the same current value going up versus coming down. Some is normal. A lot means the valve is dragging in its bore.
• Poor repeatability. Run the sweep three times. If the curve does not land in the same place, the part is not controllable, whatever it ohms.

As Herning puts it, worn solenoids show a difference in outlet pressure at a given current. That is the failure that puts a truck in limp with no hard electrical fault stored, and it is invisible to every meter test above.

Test 3: The flow side

Now the half that the meter never sees. A solenoid valve has to seal when it is closed and pass oil when it is open, and the way it dies is that the bore or the seat wears and it starts leaking in the closed position. The coil is untouched. Resistance is perfect. The circuit it feeds just never builds the pressure it is supposed to.

Vacuum testing with a manifold

The cleanest shop-level version of a flow test is vacuum, because it returns a number instead of an opinion. Sonnax's solenoid test manifold kit (part no. 95430-VTK) covers the Ford 6R60, 6R75, 6R80 and 6R100 family along with ZF 6HP19, 6HP21, 6HP26, 6HP28, 6HP32 and 6HP34, and it runs off their vacuum test stand kit VACTEST-01K. The published procedure is four steps:

1. Clean and dry the solenoids before testing. Residual fluid seals a bore that would leak dry and hands you a false pass.
2. Insert the solenoid into the correct manifold. The manifolds are color-coded — blue, yellow, orange — to the solenoid types they fit. If the cap color on the solenoid has faded past recognition, test-fit it into each manifold until one matches.
3. Attach the vacuum test line from the stand to the manifold quick-connect fitting.
4. Apply vacuum and read the gauge.

The pass/fail is published and it is tight: a good solenoid holds over 21 in-Hg, and any reading below 20 in-Hg indicates a compromised solenoid that should be replaced. There is no in-between to argue about. That is the kind of number you can put in front of a customer.

Your numbers are your numbers

One caution before you write pass/fail values on the wall for units without a published spec. Sonnax is direct that test results vary with how the stand is set up, the maximum vacuum capacity of the pump, and in some cases the altitude the shop sits at, and that gauge and calibration orifices greatly influence readings. Their recommendation for a valve body pump is 3 cfm. Their recommendation for standards is that you establish your own, by keeping a record of results for each valve body at each tested location and comparing over time.

After thirty units you know what a healthy one reads on your stand, and a part that comes in ten percent low is a decision instead of a guess.

Wet air testing when you do not have a manifold

Without a manifold set you can still separate a sealing valve from a leaking one. Wet air testing fills the cavity with ATF and pushes that fluid with regulated air — the working range is roughly 30 to 60 psi. A worn bore lets fluid escape past the valve instead of building the force to move it, and the leak shows up as visible fluid and audible air where it should be dead quiet.

The tradeoff is honest: minimal tools, and the result is visual and audible, but it is not measurable and interpreting it depends on the operator's experience. Calibrated air test stands report a measurable percentage of loss instead. For an occasional unit, wet air makes the call. For volume work, it does not.

The hot test that catches the intermittents

Every test above happens at room temperature, and a large share of solenoid complaints are heat-dependent. Cold, the unit shifts fine. Hot, it drops to neutral or limp. A cold bench pass on that part accomplishes nothing.

ATRA's GEARS published the fix in a 2013 VW 09G case study — a car shifting 1-2-3 then neutral when hot, no codes, no help from a TCM reflash — and the method transfers to any solenoid you can pull:

• Mark each solenoid and its location on the valve body first, so everything goes back where it came from.
• Heat a pan of ATF to 200 F on a hot plate and drop the solenoid in. Heating it is what lets you confirm the solenoid sticks hot.
• Push the valve toward the spring side with a small screwdriver, then tip the solenoid end over end. You should feel the armature travel freely from one end to the other.
• If it does not move, the solenoid bushings are worn or sticky. On that VW, the armature was found stuck in the N282 solenoid bushing, which confirmed the complaint.

Use pliers and a rag at that temperature, not fingers. Five minutes with a hot plate and a drain pan finds the failure a $30,000 machine passes cold.

Putting it together on a 68RFE pack

A complete bench pass on a solenoid pack, in the order that wastes the least time:

1. Ohm every coil cold. Zero out your lead resistance first. Six should land near 1.3 ohms and the pressure control near 3.5. Anything open, shorted to the housing, or off pattern is done.
2. Load test each circuit through a current-limiting probe. Confirm the current draw matches what the measured resistance and supply voltage predict.
3. Sweep the pressure control solenoid if you have the equipment. This is the one that regulates, so this is the one where dead spots and hysteresis show up as a real driveability complaint.
4. Flow or vacuum test the mechanical side, clean and dry, against a published spec if one exists for the family or against your own recorded baseline if it does not.
5. Hot-soak anything the customer described as intermittent and check armature freedom at temperature.
6. Log the results with the unit's build sheet. That record is what turns a warranty argument into a document.

The pack that fails at step 4 after passing steps 1 and 2 is the whole reason to run the procedure. That part would have gone back in the truck.

Related reading

• How to test a shift solenoid with a multimeter — the in-vehicle version of the resistance test
• Valve body cleaning and inspection — what to do with the casting once the solenoids are off
• 68RFE weak spots and fixes — where this unit fails besides the solenoid pack