Product manufacturing uses a wide range of processing centers

A 96-position tray that ran clean for six months started throwing “No Vial Detected” errors on positions 61 through 72, roughly every third injection, on a Thermo TriPlus RSH. Nothing had changed in the method. The vials were the same lot. The closures were the same lot. What had changed was the tray: a junior analyst had swapped in a spare tray carrier and seated it without checking the front locating pin. The carrier sat 1.8 mm high and 0.6 mm forward. The gripper was picking the cap flange, not the neck. Every symptom looked like a closure failure. None of it was. This is the pattern I see more than any other on autosampler pickup and misalignment calls: the closure gets blamed, and the alignment does not.

What follows is the field version. Tray height, gripper tolerance, cap seating, and the specific error codes each misalignment produces on the instruments most of us run. If you are chasing intermittent vial closures autosampler pickup error and misalignment faults, this is the diagnostic order that actually finds the root cause.

The gripper on a modern autosampler does not have much room to be wrong. On an Agilent 7693A or a PAL3 RTC, the gripper jaws close to a nominal gap and the vertical stroke is fixed. The stack that has to line up is roughly:

• Tray carrier seat (top of the instrument deck)
• Tray plate thickness and flatness
• Vial overall height (typically 32.0 mm ± 0.25 mm for a 2 mL 12 × 32 mm vial)
• Cap crimp or screw depth, which sets the flange plane
• Gripper jaw vertical position at pickup

Every one of those has a tolerance. Stack them and the worst case on a 12 × 32 mm vial with a crimp cap can be 0.8 mm of vertical drift at the flange plane. The gripper does not care about your intent. It cares about the flange. If the flange sits outside the jaw window, you get a pickup error, a dropped vial, or a bent needle on the next injection. The closure is blamed because it is the part you can see. The tray is the part you cannot.

This is why I tell procurement managers to stop buying closures against a nominal drawing and start buying against the stack. The vial closures on your bench are one dimension in a chain of five. Tightening the closure spec without checking the tray and the gripper is spending money in the wrong place.

Tray height is the first thing to measure, not the last

Most labs check the closure first because it is easy to swap a cap. I check the tray first because it is the fastest measurement and the most common culprit. Put a straight edge across the top of the tray plate and measure to the deck at four corners. You are looking for parallelism, not absolute height. A tray that is 0.4 mm out of parallel across a 96-position plate will pass on positions 1 through 40 and fail on 80 through 96, because the drift accumulates with distance from the locating pin.

The numbers that matter on a 96-position tray

On a standard 96-position 12 × 32 mm tray, the plate should sit flat to within 0.2 mm across the full footprint. The locating pins should seat with no perceptible rock. The front edge of the tray should sit within ±0.3 mm of the instrument’s reference datum. If you can slide a 0.5 mm feeler gauge under any corner, the tray is not seated and you have found your pickup error before you have touched a single closure.

I have seen trays warp from autoclave cycles, from being stacked in a cupboard under other trays, and from a single drop onto a benchtop. Polypropylene trays creep. A tray that was flat at installation can be 0.6 mm out of flat after 18 months of daily use and wash cycles. That is enough to throw a gripper.

The field case that started this article

Symptom: “No Vial Detected” on a Thermo TriPlus RSH, positions 61 through 72, every third injection. Measurement: straight edge across the tray showed 0.9 mm of lift at the front-right corner. The locating pin was seated but the rear clip was not engaged. Fix: reseated the tray, confirmed 0.15 mm flatness at all four corners, ran 200 injections with zero pickup errors. The closures were never the problem. The tray had been reinstalled after a cleaning cycle and the rear clip had been missed. Two minutes of measurement saved a week of closure investigations.

Gripper tolerance is tighter than most people assume

The gripper jaw window on a typical robotic autosampler is sized to the cap flange with a small margin. On a 11 mm crimp cap, the flange diameter is nominally 11.0 mm. The gripper jaw opening at pickup is typically sized to 11.4 to 11.6 mm. That is a 0.2 to 0.3 mm per-side margin. It is not generous. It is barely enough.

Now consider the closure. A crimp cap that is not fully seated sits higher on the neck. A screw cap that is under-torqued sits higher. A screw cap that is over-torqued can deform the septum and push the flange down, but it can also gall the thread and sit cocked. Any of those moves the flange plane by 0.3 to 0.8 mm, which is the entire margin and then some.

Gripper alignment checks you can do in ten minutes

• Park the gripper over a known-good vial in position 1 and photograph the jaw-to-flange gap with a phone on a tripod. You want a symmetric gap on both sides.
• Run the same check at position 96. If the gap is asymmetric at 96 but symmetric at 1, the tray is out of parallel, not the gripper.
• Check the gripper jaw for wear. Jaws with a Delrin or PEEK insert wear at the contact face. After 50,000 picks, a jaw can be 0.2 mm wider than spec. That is half your margin gone.
• Check the vertical stroke. The gripper should descend to a fixed Z at pickup. If the Z is drifting, the jaw closes on the cap shoulder instead of the flange, and the vial lifts cocked.

The instrument service manual will give you the nominal jaw gap and the nominal Z. Use those numbers. Do not eyeball it. A 0.2 mm error at the jaw is a 0.2 mm error at the flange, and the flange is what the gripper sees.

Cap seating is the variable the closure supplier controls, and the lab often undoes

Here is the part that gets missed. A crimp cap and a screw cap both have a correct seating plane. The closure supplier controls the cap and septum dimensions. The lab controls the crimp or the torque. If the seating is wrong, the flange plane is wrong, and the gripper error follows.

Crimp caps: the crimp is the seating

A 11 mm aluminum crimp cap on a 12 × 32 mm vial should be crimped to a specific crimp diameter. On a Wheaton or a similar manual crimper, the correct crimp diameter for a standard 11 mm cap is 11.0 to 11.2 mm measured across the crimp skirt after crimping. Under-crimp (below 11.0 mm) leaves the cap loose and the flange high. Over-crimp (above 11.2 mm) deforms the cap and can crack the neck. Both produce a flange plane that is out of tolerance, and both produce pickup errors on a tight gripper.

I have walked into labs where the crimper had never been checked. The crimp height adjustment had drifted, and every cap on the bench was crimped to 10.7 mm. The caps looked fine. They passed a visual. They failed the gripper on every 40th vial because the loose cap rocked on the neck and the flange moved by 0.4 mm at the moment of pickup.

Screw caps: torque is the seating

For 9 mm screw thread vials (the short thread GPI 9 mm format), the correct torque is 0.6 to 0.8 N·m on a calibrated torque driver. Below 0.6 N·m the cap is not fully seated and the flange sits high. Above 0.8 N·m you start to compress the septum beyond its recovery and you risk cracking the neck on thin-wall vials. The HPLC vial neck wall thickness matters here: a 1.0 mm wall vial will take 0.8 N·m all day, a 0.6 mm wall vial will not.

Torque drivers drift. A driver that was calibrated to 0.7 N·m at installation can be 0.5 N·m after a year of daily use. That is enough to leave caps high and produce exactly the pickup error pattern this article is about. Calibrate quarterly. It takes five minutes.

The error codes each misalignment produces on common instruments

This is the table I wish I had when I started. The error text varies by firmware version, but the pattern is consistent. Match the code to the misalignment, not to the closure.

Two things to note. First, the same physical misalignment can produce different error text on different firmware. Do not chase the text, chase the measurement. Second, a “vial crash” or “gripper jam” is almost always a high flange, which is almost always a crimp or torque problem, which is a lab process problem, not a closure supplier problem. The chromatography consumables supplier can only control the cap dimensions. The seating is yours.

The second field case: a torque driver that had drifted 0.2 N·m

Symptom: intermittent “Gripper failed to pick” on a PAL3 RTC, roughly one in 60 vials, always on the same tray but not the same position. Measurement: torque driver checked against a calibrated reference showed 0.5 N·m, not the 0.7 N·m it was set to. The caps were under-torqued and sitting 0.3 to 0.4 mm high. The gripper jaw was within spec. The tray was flat to 0.1 mm. The closures were within drawing. The driver was the fault.

Fix: recalibrated the driver, re-capped the tray, ran 500 injections with zero pickup errors. The lab had been about to switch closure suppliers. The driver had been in service for 14 months without calibration. That is the pattern: the closure gets replaced, the driver stays, and the problem comes back in six months when the new driver drifts too.

A troubleshooting checklist in the order I actually run it

When a pickup error lands on my desk, I run this sequence. It takes about 40 minutes and it finds the cause on the first pass about 80% of the time.

1. Tray flatness: straight edge across the plate, four corners. Spec is 0.2 mm. If out, reseat and re-measure.
2. Tray seating: locating pins engaged, rear clip engaged, front edge within ±0.3 mm of datum.
3. Vial height: caliper 10 vials from the suspect tray. Spec is 32.0 ± 0.25 mm for a 12 × 32 mm vial. If out, you have a vial lot problem, not a closure problem.
4. Cap seating plane: crimp diameter 11.0 to 11.2 mm, or torque 0.6 to 0.8 N·m. Check the tool, not the cap.
5. Gripper jaw gap: measure at position 1 and position 96. Symmetric and within manual spec. Check for jaw wear.
6. Gripper Z: confirm the descent stroke matches the service manual. Drift here shows up as a cocked pickup.
7. Closure dimensions: only now do you measure the cap. Flange diameter, flange thickness, septum flush. Compare to the supplier drawing.
8. Lot traceability: if steps 1 through 7 are clean and the error correlates to a closure lot change, you have a supplier issue. If it does not correlate to a lot change, you do not.

Notice that the closure is step 7, not step 1. The closure is the easiest thing to blame and the last thing that should be measured. The analytical testing data that proves a closure lot is out of spec is worth having, but only after you have ruled out the tray, the seating, and the gripper. Otherwise you are testing the wrong part.

What to specify so the pickup error does not come back

The fix is not a tighter closure spec. The fix is a controlled stack. Here is what I put in a closure purchase specification when the application runs on a robotic autosampler:

• Vial overall height: 32.0 mm ± 0.25 mm, measured on a calibrated height gauge, 30 pieces per lot.
• Cap flange diameter: 11.0 mm +0.0 / −0.1 mm for 11 mm crimp, 9.0 mm +0.0 / −0.1 mm for 9 mm screw.
• Cap flange thickness: 1.0 mm ± 0.1 mm.
• Seating plane: flange to vial shoulder distance, 2.5 mm ± 0.15 mm on a seated cap.
• Septum flush: no more than 0.1 mm proud of the cap top.
• Crimp diameter after crimping (in-house): 11.0 to 11.2 mm, checked on the first 5 caps of every session.
• Torque (in-house, screw caps): 0.6 to 0.8 N·m, driver calibrated quarterly.

That spec is not exotic. It is what a robotic autosampler needs. The Agilent and Restek autosampler documentation both publish the vial and cap dimensions their grippers are designed around. Use those numbers as the starting point and add the seating plane, which the instrument documentation does not cover because it assumes the lab controls it. The lab does control it. Most labs just do not measure it.

One more thing on the tray side. If you run a 96-position tray on a robotic system, buy a spare tray and keep it flat in a dedicated rack. Trays creep. A tray that has been in service for 18 months is a suspect. Swap it out, measure the old one, and if it is out of flat, retire it. A tray is cheaper than a week of failed runs.

The decision rule

When a pickup error or misalignment fault appears, measure the tray and the seating tool before you measure the closure. If the tray is flat to 0.2 mm, the crimp or torque is in spec, and the gripper gap is symmetric at both ends of the tray, the closure is your suspect. If any of those three is out, the closure is innocent and the fix is in your lab. The error code tells you which of the three to check first: “vial not found” points at the tray, “gripper jam” points at the seating, and “failed to pick” points at the gripper. Chase the measurement, not the part. The closure supplier will sell you a new lot either way, and it will not fix a tray that is 0.9 mm out of flat.

References & further reading

Related resources from our lab consumables network:

• Vial Closures at vialsclosures.com
• Analytical Testing at testinganalytical.com
• Hplc Vial at vialforhplc.com
• Chromatography Consumables at chromatographyconsumables.com

External references and standards:

• Agilent chromatography consumables and columns
• Thermo Fisher chromatography and sample handling
• U.S. EPA SW-846 test methods for waste and materials
• Restek technical resources for chromatography