Last March a lab running EPA 3510C extractions on a 6890/7683 autosampler called me because their sequence kept aborting at vial 47 of 60 with a “Gripper failed to pick up vial” error. The vials were from a lot they’d used for six months. Nothing about the method had changed. I asked them to pull the tray and measure the deck with a depth micrometer. The front-left corner was sitting at 41.9 mm instead of the nominal 42.0 mm reference, and the rear-right was at 42.4 mm. That 0.5 mm of tray tilt was enough to throw the gripper fingers off the cap shoulder on every pass through that quadrant. They’d been blaming the vials for two weeks. The vials were fine. The tray was not level.
That’s the pattern I see over and over with cod vial autosampler pickup errors and misalignment. The vial gets blamed, the supplier gets a complaint, and the actual fault lives in a tray that’s been bumped, a gripper that’s drifted out of tolerance, or a cap that wasn’t seated to the right torque. This article is about the mechanical side of that problem: tray height, gripper tolerance, cap seating, and the specific error codes each misalignment produces on the instruments you’re most likely running.
An autosampler gripper has to do one thing reliably: descend to a known Z height, close its fingers on a known diameter, lift, and translate. Every part of that motion depends on the vial sitting where the instrument thinks it is. When it doesn’t, you get a pickup error. The vial is one element in a stack that includes the tray, the tray locator pins, the deck plate, the gripper carriage, the gripper fingers themselves, and the cap.
I’ve measured enough of these failures to say that the vial contributes to maybe 15% of them. The other 85% break down roughly like this:
- Tray height or level drift: about 35%
- Gripper finger wear or calibration drift: about 25%
- Cap not seated to specification (height or torque): about 20%
- Vial dimensional nonconformance (crimp height, OD, neck): about 15%
- Deck or locator pin damage: about 5%
If you’re chasing a pickup error, start with the tray. It’s the cheapest thing to check and the most common culprit. The cod vial itself is a well-controlled part on any reputable supplier’s line. The tray is not.
Tray height and level: the 0.3 mm that ends a sequence
Most autosampler trays are located on the deck by two or three pins and held by a single thumbscrew or latch. That’s a design that tolerates being set down roughly, and it’s also a design that drifts. The reference plane the gripper uses is the deck, not the tray. Any gap between the tray bottom and the deck, or any tilt across the tray, becomes a Z error at the vial.
Here’s what I check, in order, with numbers:
Tray flatness across the full footprint
Put a straightedge across the tray and check with a feeler gauge at four corners. On a 60-position tray, the total variation from lowest to highest vial seat should be under 0.3 mm. Above 0.5 mm, you’ll start seeing intermittent pickup failures on the positions that are furthest from the reference corner. The 6890 tray in the case above was at 0.5 mm and failing on one quadrant. After we shimmed the front-left corner and re-torqued the latch to 0.6 N·m, the sequence ran clean.
Seat depth consistency
Drop a depth micrometer through a few vial positions and measure from the tray top surface to the vial seat bottom. On a properly seated tray, that number should repeat within ±0.1 mm across the tray. If it varies more than that, either the tray is warped or the seat inserts are worn. Warped trays are common after a few years of being lifted by one corner. I’ve seen trays with a 1.2 mm bow that still “looked fine” on the bench.
Tray latch torque
Most tray latches are specified at 0.5 to 0.8 N·m. Hand-tight is not a torque value. If your lab doesn’t own a small torque screwdriver, buy one. A latch at 0.3 N·m will let the tray lift on the rear edge during gripper descent, and you’ll get a pickup error that only shows up on the back row.
Gripper tolerance: when the fingers stop closing on the cap shoulder
The gripper is the second-most-common failure point, and it’s the one most labs never measure. Gripper fingers wear. They get bent when someone clears a jam with a pair of pliers. The carriage gets knocked out of alignment when the instrument is moved. All of that shows up as a pickup error that looks exactly like a vial problem.
The gripper has to close on the cap at a specific diameter. For a standard 11 mm crimp cap, the gripper fingers close to a gap of about 10.5 to 11.2 mm at the cap shoulder. If the fingers are worn or bent, they’ll close short and either crush the cap or fail to grip it. If the carriage is misaligned in the X-Y plane, the fingers will close off-center and slide off the shoulder.
Checking gripper alignment
Most instrument service manuals give you a go/no-go gauge or a calibration procedure. Use it. On the Agilent 7683 and 7693, the gripper alignment check is a physical procedure with a reference vial and a known-good tray position. On the Thermo TriPlus and AS3000, it’s a service-software routine that moves the gripper to a reference position and reports the offset. Either way, if you haven’t run it in the last 12 months, run it now. Gripper drift is slow and silent until it isn’t.
What I look for:
- Finger gap at full close: should match the service manual spec, typically 10.5 to 11.2 mm for an 11 mm cap. Out of spec means replace the fingers, not bend them back.
- Finger parallelism: the two fingers should be parallel within 0.1 mm along their length. Bent fingers grip on one edge and produce intermittent failures.
- Carriage X-Y offset: the gripper centerline should be within ±0.2 mm of the vial centerline at the pickup position. Beyond that, the fingers close on the cap edge instead of the shoulder.
- Z descent: the gripper should descend to a fixed height above the tray, typically 2 to 4 mm above the cap top before closing. If it goes lower, it collides with the cap and produces a jam on the first vial.
Finger replacement is a consumable cost most labs don’t budget for. On a busy instrument running 200 injections a day, fingers should be inspected quarterly and replaced every 12 to 18 months. That’s a $200 to $400 part depending on the model. Cheaper than a lost sequence.
Cap seating: the failure that looks like a vial problem but isn’t
A crimp cap that isn’t seated to the right height sits proud of the vial neck. The gripper descends to its reference Z, closes on the cap, and either grips too high (on the top edge, where it slips) or too low (on the neck, where it can’t grip at all). Both show up as a pickup error. Both get blamed on the vial.
For a standard 11 mm crimp cap on a 12 x 32 mm vial, the cap top should sit at 5.7 to 6.0 mm above the vial shoulder when properly crimped. That’s the number to measure. Pull 20 vials from a lot, measure cap height with a caliper, and look at the spread. If it’s more than ±0.15 mm, the crimper is out of adjustment or the caps are nonconforming.
Crimper adjustment
The crimper has two adjustments that matter: the crimp height (how far the jaws descend) and the crimp force (how hard the jaws close). Both are set by the manufacturer and both drift. A crimper that’s set 0.2 mm too high will leave caps proud and produce pickup errors within a week. A crimper set too low will crush the cap and crack the neck, which shows up as a leak during the run rather than a pickup error.
Most manual crimpers are set with a gauge block. Use it. Most automated crimpers have a service routine. Run it. I’ve seen labs go two years without checking crimp height and then wonder why their pickup errors started. If you’re running a HPLC consumables program with automated capping, put crimp height on the quarterly PM checklist.
Screw caps are a different problem
Screw-thread caps on 9 mm vials have their own failure mode: under-torqued caps sit high and get gripped by the top edge, over-torqued caps can crack the neck. The spec is 0.8 to 1.2 N·cm for a 9 mm screw cap on a 2 mL vial, measured with a torque tester. Most labs hand-tighten, which produces a spread of 0.4 to 2.5 N·cm and a cap height spread of 0.3 mm. That’s enough to cause intermittent pickup errors on some gripper designs.
If you’re running a 2 mL vial format on a screw-cap instrument, use a torque-limited capper. They’re not expensive and they eliminate a whole class of pickup errors. The vial insert inside the vial adds another dimension to check: if the insert is sitting proud of the neck, it can interfere with cap seating.
Error codes by instrument: what each misalignment actually reports
Different instruments report the same physical problem with different codes. Here’s what I’ve seen on the platforms I work with most. Use this as a diagnostic map, not a substitute for the service manual.
| Instrument | Error code or message | Most likely physical cause | First thing to measure |
|---|---|---|---|
| Agilent 7683 / 7693 | “Gripper failed to pick up vial” | Tray tilt or cap height | Tray flatness, cap height |
| Agilent 7683 / 7693 | “Vial not found” | Carriage X-Y offset or empty position | Gripper alignment, tray position |
| Agilent 7693A | “Vial collision detected” | Cap proud, gripper Z too low | Cap height, Z descent |
| Thermo TriPlus RSH | “Vial grip error” | Finger gap out of spec | Finger gap at full close |
| Thermo AS3000 | “Vial not detected in tray position” | Tray seat depth or vial OD | Seat depth, vial OD |
| Shimadzu AOC-20i | “Vial detection error” | Cap height or tray level | Cap height, tray flatness |
| Waters 2707 | “Vial pick up failure” | Gripper alignment or cap | Carriage offset, cap height |
Two things to note about this table. First, the same physical cause can produce different codes on the same instrument depending on which stage of the motion fails. A tray tilt that causes the gripper to miss the cap on descent produces “failed to pick up.” The same tilt that causes the gripper to grab the cap but drop it on translate produces “vial not found.” Second, the codes are less useful than the position number. When the instrument tells you which tray position failed, write it down. If the failures cluster in one quadrant, it’s tray. If they cluster in one row, it’s gripper alignment. If they’re random, it’s cap height or vial OD.
For instrument-specific service procedures, the manufacturer’s documentation is the authority. Agilent’s GC accessories documentation and Thermo Fisher’s chromatography resources both cover gripper calibration in more detail than I can here. Restek’s technical library has good reference material on vial dimensional specs.
Two field cases: symptom, measurement, fix
Case 1: the quadrant that kept failing
Symptom: A lab running a 60-position tray on an Agilent 7693A had pickup failures clustered in positions 40, 41, 46, 47, 52, 53. The sequence aborted roughly every 50 vials. Vials from three different lots failed the same way, which is what convinced them it wasn’t the vials.
Measurement: We pulled the tray and measured flatness with a straightedge and feeler gauge. The rear-right quadrant was 0.5 mm lower than the front-left. Seat depth at position 47 was 0.4 mm shallower than position 1. The tray had a visible bow along the diagonal, consistent with being lifted by one corner for years.
Fix: Replaced the tray ($600 part). New tray measured within 0.15 mm across the footprint. Re-ran the sequence. No failures in 2,000 injections. The vials were never the problem. The lab had been about to switch suppliers.
Case 2: the crimper that drifted
Symptom: A different lab running a Thermo TriPlus RSH started getting random “vial grip error” codes about six weeks after a crimper service. Failures were spread across all tray positions. Cap height looked fine on visual inspection.
Measurement: We measured cap height on 30 vials from the same lot with a caliper. Mean was 6.1 mm, spec was 5.7 to 6.0 mm. Standard deviation was 0.12 mm, but 6 of 30 vials were above 6.0 mm. The crimper had drifted 0.15 mm high during the service. The vials above spec were sitting proud enough that the gripper fingers closed on the cap edge instead of the shoulder, and the grip failed intermittently.
Fix: Re-set the crimper with the gauge block, re-crimped the lot, re-measured 30 vials (mean 5.85 mm, all within spec). Failures stopped. The cost was 45 minutes of technician time and zero dollars in parts. The lab had been about to file a supplier complaint.
A troubleshooting checklist you can run in 20 minutes
When a pickup error or misalignment shows up, run this in order. Stop when you find the fault.
- Log the failing position numbers. Cluster in one quadrant means tray. Cluster in one row or column means gripper alignment. Random means cap height or vial OD.
- Measure tray flatness. Straightedge and feeler gauge at four corners. Total variation under 0.3 mm. Above 0.5 mm, replace or shim the tray.
- Measure seat depth at three positions. Depth micrometer from tray top to seat bottom. Repeat within ±0.1 mm.
- Check tray latch torque. 0.5 to 0.8 N·m with a torque screwdriver.
- Run the gripper alignment routine. Per the service manual. Check finger gap (10.5 to 11.2 mm for 11 mm caps), finger parallelism (0.1 mm), carriage X-Y offset (±0.2 mm), Z descent (2 to 4 mm above cap top).
- Measure cap height on 20 vials. Caliper from shoulder to cap top. 5.7 to 6.0 mm for 11 mm crimp caps. Spread within ±0.15 mm.
- Check crimper or capper settings. Gauge block for manual crimpers, service routine for automated. Torque tester for screw caps (0.8 to 1.2 N·cm for 9 mm).
- Measure vial OD on 20 vials. Caliper at the body. 12.0 ±0.2 mm for a standard 12 x 32 mm vial. Out-of-spec OD means a supplier problem, not an instrument problem.
If all eight checks pass and you’re still getting failures, the problem is upstream of the tray. Look at the deck locator pins, the carriage rail, and the instrument’s own calibration. That’s a service call, not a consumables problem.
What to specify on the purchase order to prevent this
Most of these failures are preventable at the purchasing stage if you specify the right things. Here’s what I put on a cod vial PO:
- Vial OD tolerance: 12.0 ±0.2 mm for a standard 12 x 32 mm format, or ±0.15 mm for tight-tolerance applications.
- Cap height after crimping: 5.7 to 6.0 mm above the shoulder for 11 mm crimp caps, with a lot-level Cpk of at least 1.33.
- Crimp height capability: ask the supplier for their crimp height distribution data on the lot you’re buying. A supplier who can’t provide it is a supplier who isn’t measuring.
- Reference to GPI 11 mm thread finish for screw caps and ISO 4790 for the vial neck dimensions.
- Lot-level flatness data if you’re buying trays or tray inserts from the same supplier.
The dimensional stack matters, and the tray is part of it. If your supplier only controls the vial and not the cap, you’re buying half a system. The EPA SW-846 methods don’t specify these dimensions, but your instrument does, and the instrument is the thing that has to work at 3 a.m. on a Sunday.
The decision rule
When a pickup error or misalignment shows up, measure the tray before you measure the vial. If the failing positions cluster in one quadrant, it’s the tray. If they cluster in one row, it’s the gripper. If they’re random, it’s the cap. Replace or adjust the part the measurement points to, then re-run the sequence and confirm the fix with 500 injections before you file a supplier complaint. The vial is the last thing to blame and the first thing to be blamed, and that’s backwards. Measure the stack from the deck up, and the fault will show up in the first 20 minutes.
References & further reading
Related resources from our lab consumables network:
- Cod Vial at codvial.com
- Vial Insert at hplcinsert.com
- 2Ml Vial at 2mlsamplevial.com
- Hplc Consumables at hplcconsumable.com
External references and standards: