September 2022. I handle motion and controls purchasing for a mid-size packaging operation — three lines, roughly 40 motors and actuators through our maintenance budget every year. I have been doing this since 2016. In that time I have logged 11 significant buying mistakes that cost us somewhere north of $14,000 in rework, delays, and spare parts we never needed. The mistake I am about to describe is number seven. Our team keeps my error log on the shared drive as training material.
The project looked routine. Line 2 used a pneumatic pusher to index cartons into a case packer. The air supply to that station was bleeding roughly $900 a year in leaks and compressor load, and the cycle tuning was a constant headache. We decided to replace it with an electric linear actuator. Straightforward call.
Same week, our HVAC contractor flagged two failing damper actuators in the cleanroom air handler. They quoted Johnson Controls VA7810-HGA-2 electric actuators. That quote landed on my desk in the same email thread as the line retrofit. Remember that detail. It matters in about 900 words.
I asked the line engineer for the required speed. He said, “About two inches a second — whatever that is in metric.” 50 mm/s. I wrote it down. Then I made the mistake.
The Assumption I Did Not Check
I went looking at linear actuator datasheets and filtered for 50 mm/s. Plenty of units hit that number. The one I picked — a 12V DC unit with a ball screw — listed 52 mm/s at no load. Good enough, I thought. Ordered it. (This was September 2022, when standard lead times on this class of actuator were still four to six weeks, so there was no room to course-correct quietly.)
Here is the thing about “how fast can a linear actuator move”: it is not one number. It is a curve. The figure on the first page of the datasheet is almost always the fastest point on that curve — no load, nominal voltage, ambient temperature, fresh grease.
What I should have checked was the load-speed curve. Our carton pusher needed about 45N of continuous thrust to index the pack. At 45N, that actuator’s speed dropped to something closer to 28 mm/s. And that was before we factored in the start-stop acceleration in a 40-cycle-per-minute duty.
I assumed “52 mm/s rated” meant “52 mm/s under our conditions.” Did not verify. Turned out the datasheet number was the ceiling, not the floor plan. The spec was fine on paper. Put another way: fine on paper is not fine under a 45N load.
The Model Number Mix-Up
This is where it got embarrassing.
Our procurement assistant — good person, still learning the product lines — pulled the actuator quote and the HVAC quote into the same purchase request spreadsheet. When she needed a reference for “what speed an electric actuator runs at,” she grabbed the Johnson Controls VA7810-HGA-2 spec sheet.
That is a damper actuator. Ninety degrees of rotation, travel time measured somewhere in the tens of seconds (I want to say 30 to 90 depending on the model, but do not quote me on the exact figure), low torque, built for air valves and butterfly dampers. It has nothing to do with a linear thrust application.
But the speed figure on that sheet — measured in seconds per 90 degrees — got written into the comparison notes. For a week, two of us were comparing a rotary damper actuator’s angular travel time against a linear rod’s stroke speed. We were using the same word and meaning completely different things. “Fast” for a damper actuator means “opens before anyone notices.” “Fast” for a linear actuator means the rod is physically moving millimeters per second.
We caught it when the electrical contractor asked why a VA7810-HGA-2 was referenced on the line retrofit drawings. He thought we had lost our minds.
I should add that this was not helped by the Johnson Electric versus Johnson Controls brand confusion. Johnson Electric industrial motors and Johnson Controls HVAC actuators share a last name and nothing else. A new hire on the team once asked why we were ordering air dampers from the motor supplier. If you have both in your supply chain, put a note on the folder. Seriously.
The Gear Motor Question Nobody Asks
While all that was unfolding, the servo drive on our sewing machine station — a servo motor sewing machine running a decorative stitch pattern on the fabric liner — started throwing position faults. Different project, same week. That is how maintenance goes.
The fault traced back to the gear motor we had specified upstream the previous quarter. I picked it on output torque and ignored backlash and reflected inertia. Technically, the gear motor was within torque spec. But its backlash was adding roughly 0.4 degrees of positioning error at the load. On a stitch pattern that stacks 12 stitches into 3mm, 0.4 degrees shows up as a visible defect.
The question everyone asks about a gear motor is “what is the torque?” The question they should ask is “what is the backlash, and what does it do to my positioning tolerance at the load?”
(Should mention: the gear motor itself was fine. We specified the wrong one for the tolerance. Different kind of expensive.)
The Three-Day Fix
We caught the speed problem during first-try commissioning, which was luck. The actuator moved the cartons — just not at cycle rate. We were running 34 cycles per minute against a 40-cycle target. That is a 15% shortfall on a line that was already capacity constrained.
We ordered the next size up with a 10mm lead screw instead of 5mm. That roughly doubles the speed at the same motor RPM, at the cost of available thrust. Our thrust requirement had margin, so the trade worked. Took three days to get the unit and redo the mounting.
Cost: about $1,800 in expedited shipping, overtime, and the catch-up plan we ran on the backlog. The original actuator now lives on a shelf as a reminder. We have caught at least four mismatched specs with the updated checklist in the past 18 months — two of them would have halved throughput if they had shipped.
What I Actually Learned About Linear Actuator Speed
The formula is simple, and I ignored it
Linear actuator speed comes from screw geometry:
Speed (mm/s) = [Motor RPM ÷ Gear Ratio] ÷ 60 × Lead (mm per revolution)
A 3,000 RPM motor through a 5:1 gearbox with a 5mm lead screw runs at about 50 mm/s on paper. Swap in a 10mm lead screw and you are at 100 mm/s with half the thrust capability. You can do this arithmetic before you buy anything.
No-load speed is not your speed
The datasheet number is the no-load ceiling. Under load, motors slow down — how much depends on the torque-speed curve, the driver, and the supply voltage. Plan with the loaded number.
Acceleration eats your cycle time
A 40-cycle-per-minute application gives you 1.5 seconds per cycle. If 0.2 seconds is acceleration and 0.2 is deceleration, you are only at full speed for the middle. The peak speed that matters is often 10 to 20% higher than the average speed you need.
Different actuator families, different rules
HVAC damper actuators, industrial linear actuators, and servo position loops all describe “speed” differently. Rotational travel time, stroke speed under load, actuator bandwidth. If you find yourself comparing numbers across those three categories, stop and check the units. I learned that the hard way.
The Checklist We Use Now
I maintain the pre-purchase checklist. Here is the actuator section:
- Required average speed under expected load — not no-load speed
- Peak speed needed to hit cycle time, after accounting for accel and decel
- Load-speed curve at actual thrust demand
- Screw lead, gear ratio, motor RPM — written out as a calculation, not a vibe
- Duty cycle and the continuous thrust the actuator will actually see
- Mounting orientation, side load, and whether the rod is guided or freestanding
- The supplier’s “speed” spec type — no-load, full-load, or average
- Whether you have accidentally referenced a different actuator family’s datasheet (yes, this is on the list now)
The checklist costs about 20 minutes per purchase. Cheaper than $1,800 and three days.
If you are picking a linear actuator for the first time and the datasheet lists a single speed figure, treat that number as marketing, not engineering. Ask for the load-speed curve. If the supplier does not publish one, that silence tells you something.
And if your purchasing team is juggling multiple actuator types across different product lines — HVAC, linear motion, servo systems — build the cross-reference document on purpose. Because the moment two datasheets sit in the same spreadsheet, somebody will compare numbers that do not mean the same thing.
That somebody was me. I have the invoice to prove it.