
A probe with a hex surface screw is a thermowell-style probe that is held into its port by a nut rather than by the thread alone, and that single difference changes three things at once: how it seals, how it is installed, and how it will be removed. It is a small design difference and it produces a genuinely different part.
Deep drawing followed by CNC finishing on critical faces, at +/-0.02 mm on machined seats and sealing faces flat to drawing, with finer bands confirmed from capability data before quoting. So the tight band’s destination is the seats – and on this part the seats are the packing gland, which is the feature that makes the design work.
Why a nut, when a thread already holds the part in
- A taper-sealing threaded thermowell works by interference. The threads wedge, the seal is made along the flanks, and the joint is sealed by the tightening. That is a good seal and it is not removable – because the flanks have deformed and sliding them apart tears them.
- A packed or O-ring-seated probe with a nut seals at a gland instead, and the thread only carries the clamping load. The seal is a separate element, it can be replaced, and – this is the important one – the joint is made and broken many times without destroying anything.
- And the surface screw – the external hex – is what lets the nut be applied and removed without fouling the port. So the hex is a functional feature on a part where a hex would otherwise be an assembly convenience, and its dimension is a torque specification rather than a handle dimension.
So the design question on this part is: is the joint meant to be broken? If it is, the nut-and-gland arrangement is correct and the thread only has to hold the load. If it is not – a permanent installation – a taper would have been simpler and more reliable.
The gland, which is the hard feature
On this design the gland is what makes the seal, and it has three requirements that a general tolerance does not cover.
- The gland’s geometry sets the seal’s compression. A groove that is too deep lets the packing bottom out before the thread’s preload is reached; one that is too shallow compresses the packing so hard that it extrudes into the clearance gap between the probe and the port and fails there. So the groove’s depth and width are functional and they are set by the packing’s properties, not by the metal.
- The probe’s bore diameter in the gland is a clearance, and it is a thermal path. A packing gland normally has a small bore above the gland so the packing’s compression is not carried into the measurement – and that step creates a thermal discontinuity. So the step’s position and the material above it are both accuracy parameters.
- And the packing’s material and hardness are requirements. Packing against a stainless steel probe at temperature is a compression set question; on a probe that is removed and refitted repeatedly, a packing that has taken a permanent set will leak the next time.
The hex, as a torque interface
- Width across flats, as a fit to the wrench, with the corner condition stated – and the tolerance tightened enough that the assembly torque it transmits is consistent. A hex that varies in width varies the torque needed to reach the same preload.
- The hex’s strength, because the nut has to be run down the thread and stopped, and a hex that rounds under that will damage the port. So on a small probe the hex’s height and the material’s hardness matter more than on a large fitting.
- And the material pairing with the nut, which on stainless steel is the galling question again. A stainless steel nut on a stainless steel probe, or on a stainless steel port, will gall during assembly and the joint will be difficult to break. A dissimilar grade pairing or a dry film lubricant solves it, and it costs one line on the assembly instruction.
The probe itself, and the three things it must do
- Carry the process. Where the process is pressurised, the probe’s wall is a pressure boundary and the gland’s position along the wall is a structural question as well as a sealing one – the pressure acts on the area between the gland and the tip.
- Conduct the temperature. And on this part there is a specific complication: the gland’s packing is a thermal break, and the nut is a heat sink. So the assembly’s nut – which is usually brass or steel and much larger than the probe – conducts heat away from the probe’s base, which increases the error at the probe’s tip. On a high-accuracy probe this is a real and often-missed contribution.
- And survive the process thermally. A packed joint at temperature is a joint with two materials at two temperatures, and the differential expansion between a stainless steel probe and a steel or brass nut is a load on the thread.
A packed probe’s third material job is specific to the gland
Materials held include carbon steel, 304 and 316 stainless steel, aluminium and brass, with material substitutes proposed only with written approval. On this part the material has three jobs and the third is specific.
- 316L for the probe on the chloride logic, and the L grade where there is welding. On a probe that may be welded into a fitting, that matters.
- And the nut and the packing are separate material decisions. A brass nut is the common choice because it will not gall against stainless steel and it deforms rather than damages. A steel nut is stronger and will gall. And on a small probe the nut’s material is what sets whether the joint can be broken.
- The pairing inside the gland. A stainless steel probe with a stainless steel gland and an O-ring is one arrangement; a stainless steel probe with a filled PTFE gland is another, and the filled gland has a temperature limit that the probe itself does not. So on this part the assembly’s temperature limit may be set by the gland rather than by the probe, and that should be stated.
Packing needs a prepared gland, and plating gives it nothing to grip
Natural, plated or coated per the print, with the agreed finish logged and finish thickness and adhesion checked on the first article. Three items.
- The gland and the probe’s bore through it, unmasked – the packing has to grip a prepared surface and a plated gland will not hold the same compression.
- The thread, masked or specified post-finish with an allowance.
- The exterior, where a passivated stainless steel probe is the right answer and a plating is generally not – because a plated probe that gets scratched in service will corrode behind the plating and the seal’s surface will have changed.
The fourth flagged dimension is a gland diameter, inside the part
+/-0.02 mm on machined seats and sealing faces flat to drawing. On this part the flagged dimensions are four, and one of them is a diameter in the gland that is easy to overlook because it is inside.
- Sealing faces: flatness and perpendicularity, measured as two values, against a named datum.
- Thread: class and pitch diameter, gauged, and the sealing method stated – and on this design the thread does not seal, which is the whole point, so the statement should say so.
- The gland’s bore diameter, as a clearance to the packing, held. It is the diameter that decides how much of the packing’s compression is carried in metal rather than in the seal.
- The hex: width across flats as a fit, with the corner condition.
- And the requirements that are not dimensions: the packing’s material and hardness, its compression percentage, the assembly torque, and whether the joint is intended to be broken. Those four determine whether this part works as designed, and on a packing gland all four are missing more often than any dimension.
A packed joint can only be verified by assembling it
First articles verify the whole flow before the run is released. On this part two inspection items are not dimensional and both matter.
- A functional leak test on every part, at the specified torque, because on a packed joint the seal is made by assembly and the only way to verify it is to assemble it.
- A break-and-refit check on a sample – assemble, leak test, disassemble, reassemble, leak test again. On a probe whose whole point is that it can be removed, the second seal is the one that matters and it is the one that gets skipped.
- Dimensional: the sealing faces, the thread, the gland’s bore, the hex.
- And a functional temperature check where an element is fitted, since the packing’s compression and the nut’s heat sinking both affect the reading.
Removed and refitted in service, and how often?
Is this probe going to be removed and refitted in service, and if so how often? That question decides whether the packed design is worth its extra complexity over a simple taper – and on a probe in a process that has to be sampled or inspected regularly, it usually is. It also decides the packing’s material, because a joint broken occasionally can take a packing that takes a permanent set, and a joint broken every week needs one that does not. And there is a second question that is specific to this arrangement and that is worth more than it looks: how much of the probe’s thermal error is the nut adding? Because the nut is a large mass of metal in contact with the probe’s base, and it conducts heat away from exactly the point where the temperature gradient is steepest. On a probe where the nut is brass – a common and sensible choice for the galling reasons – the thermal penalty is larger than with a steel nut of the same size. So the nut’s material is chosen for the joint’s serviceability and it costs accuracy, and on an accurate probe that trade should be made consciously rather than by default. Where the accuracy matters more than the refit frequency, a longer probe or a thermally isolating nut boss is often the better answer than choosing a different nut.
This part belongs to our Sensor Housings range. The sensor housings pages collect the process notes, common materials and inspection points behind every product in the family.
Specifications
| Item | Stainless steel 316 sensor probe with hex surface screw |
|---|---|
| Brand | AUNOV |
| Designing | Send drawings, or send a sample to be measured – DFM notes are returned before any tooling is cut; the same drawing is used for every repeat order. |
| Tooling | Tooling is built and held under your part number, so die cost stays transparent, and the die remains dedicated to your part. |
| Samples | Approval samples of the base ship with dimensional data plus material certs. |
| Materials | Materials held include carbon steel, 304 / 316 stainless steel, aluminium and brass; material substitutes are proposed only with your written approval. |
| Surface finish | Natural, plated or coated – the print decides, and the agreed finish is logged; finish thickness and adhesion checked on the first article. |
| Processing | Deep drawing first, CNC finishing on critical faces after; first articles check the complete flow before we release the run. |
| Equipment | Hydraulic presses for drawing, with CNC turning for finishing; secondary ops run on CNC turning centres, inspection on the CMM. |
| Tolerance | +/-0.02 mm on machined seats, sealing faces flat to drawing; finer bands confirmed from capability data before quoting. |
| Delivery terms | CIF or DDP, with FOB available, T/T or L/C, with D/P accepted on repeat business; each volume tier carries its own quoted price. |
Related Products
Frequently Asked Questions
Why does a hex-surface screw change the design?
The probe is held by a nut on a hex rather than by thread alone, which changes how it seals, how it is installed and how it is removed – three things at once.
How are the sealing seats held?
Machined seats and sealing faces are finished to +/-0.02 mm with flatness to drawing, because on this part the seats are the packing gland that makes the design work.
What grades is the probe tube made from?
304 and 316 stainless steel, drawn then CNC finished on the critical faces, with the grade chosen for the process it sits in.
Can the hex be formed rather than machined?
The hex can be formed or machined depending on the volume and the seat tolerance, with the choice confirmed from first-article capability data.
Request a Quote for This Part
Send a drawing or sketch — get material advice, tooling cost and a per-piece price within 24 hours (working days).




