You approved a sample. It gripped the desk, absorbed a waist-height drop onto tile, and came off the device without a fight. You wrote "50 Shore A" on the drawing and released the order. Six weeks later the production units arrive at the same nominal durometer and feel noticeably stiffer, the corners feel dead rather than springy, and one of your merchants splits a port cutout in the first week. Nothing in the paperwork is wrong. The Shore A hardness spec for silicone protective cases is one of the most confidently written and least well-understood numbers in an accessory drawing, and on its own it does not control the thing you actually care about.
Durometer is a material property. Feel is a structural property. Those two facts explain almost every hardness dispute in this product category.
What a durometer measures, and what it does not
The test method, and the two details that cause disputes
Shore A is defined by ASTM D2240 and ISO 7619-1. A spring-loaded indenter with a specified truncated-cone tip is pressed into the rubber under a defined force; the scale reads how far the material resists penetration, from 0 (very soft) to 100. It measures resistance to a single indentation at one point on one surface. It says nothing about how the part bends, how much energy it absorbs in a drop, how far it stretches before it tears, or how long it holds its shape.
Two details in the test method cause more disagreements between buyers and factories than the compounds ever do.
Reading time. Silicone is viscoelastic — it relaxes under the indenter. An instantaneous peak reading and a reading taken after a 15-second dwell can differ by several points on the same part. ASTM D2240 permits both, provided you state which you used. If your drawing does not say, your factory's QC and your incoming inspector can each be correct and still disagree.
Specimen thickness. The method requires a specimen at least 6 mm thick, and allows plied-up layers to reach it. A protective case wall is nowhere near that. Press a durometer against a 1.8 mm case wall and you are partly measuring whatever is underneath — the bench, the device, your thumb. The number is not valid and it is not repeatable.
That second point has a direct consequence: the factory cannot legitimately durometer your case. What a competent supplier does instead is mould a test slab or button from the same batch of compound, cure it on the same schedule, and measure that. Which means your hardness spec is a specification on the compound, verified on a witness sample — not a measurement of the part in the carton. Write it that way and the whole conversation gets easier.
Why the same nominal Shore A feels different at different wall thicknesses
This is the mechanism behind the disappointed-buyer story at the top of this article.
When you flex a case wall between your fingers, you are not indenting it. You are bending it. Bending stiffness for a plate or shell scales with the cube of thickness. Go from a 1.2 mm wall to a 2.4 mm wall at identical durometer and you get roughly eight times the bending stiffness. A 50 Shore A case at 2.5 mm reads as firm and structural in the hand; the same compound at 1.2 mm reads as floppy and cheap.
The practical consequences stack up fast:
- A tooling change that thickens a wall to fix a short-fill or sink problem also changes the feel of the product, without touching the compound.
- A case that is 2.2 mm on the back panel and 1.4 mm on the side rails will feel like two different materials, because it structurally is.
- Reinforced corners and a raised screen bezel — standard construction on rugged POS and PDA cases — are local thickness features. They change perceived stiffness in exactly the places a reviewer picks the case up.
- Drop energy absorption depends on how much material sits between the impact point and the device, and on how far it can compress before bottoming out. That is geometry, not durometer.
So if the approved sample is the reference for feel, freeze the geometry alongside the hardness. Nominal wall thickness with a tolerance, corner section thickness, rib heights and bezel lip height all belong on the drawing. A factory quoting "temperature and pressure controlled" compression molding for consistent wall thickness is telling you it can hold what you specify — but only what you specify. The CMH guide to writing a product spec sheet a Chinese factory can actually build to covers the drawing discipline in general; this is the case-specific version.
What the tolerance band really means
Ask a factory to hold ±3 Shore A and most will agree. Ask what that number is composed of and the picture changes.
| Source of variation | Typical magnitude | Who controls it |
|---|---|---|
| Batch-to-batch compound variation | A few points | Compound supplier |
| Pigment and filler loading | Can read a point or two harder at heavy loading | Factory / your colour spec |
| Cure temperature and dwell variation press to press | A point or two | Factory process control |
| Post-cure / post-bake schedule | Usually reads slightly harder after post-bake | Factory process |
| Reading method — instantaneous vs 15 s | Several points on silicone | Your drawing |
| Hand-held durometer vs calibrated test stand | A point or two, operator dependent | Test method |
| Measuring a curved or thin surface instead of a slab | Large and unrepeatable | Test method |
Add those up honestly and a commercial ±5 Shore A band on a moulded slab is a realistic commitment. A ±3 band is a genuine tightening that costs process attention. A ±2 band written on a drawing with no stated reading method and no stated specimen is not a specification at all — it is a number that will be argued about at inspection.
There is a second, subtler trap. Heavily pigmented compounds tend to read slightly harder than lightly pigmented ones from the same base. If you run a black case and a light-grey case as one SKU family at "50A," expect them not to measure identically, and decide in advance whether that matters to you or not.
The three properties that matter more than hardness — and get specified less
Hardness is the number buyers write down because it is the one they know. These three cause more field failures on protective cases.
Tear strength
A silicone compound can be soft and still tear easily, or soft and highly tear-resistant. On a case, tearing starts at stress concentrations: the corner of a rectangular port cutout, the edge of a speaker grille, a thin lip that gets stretched over the device on every installation. If you have chosen a soft compound for grip and drop performance and said nothing about tear, you have optimised one property and left the failure mode unconstrained.
Specify tear strength to ASTM D624 (Die B or Die C — state which) in kN/m, and require the value on the compound datasheet. Then radius every internal corner on the cutouts. A sharp inside corner on a moulded elastomer is a crack starter, and it is free to fix in CAD.
Compression set
A case grips because the material is held in tension or compression against the device. Compression set measures how much of that recovery is permanently lost after being held deformed at temperature — ASTM D395 Method B, reported as a percentage after a stated time and temperature.
This is the property behind "the case has gone loose." A terminal sitting in a warm charging cradle overnight, every night, for a year, is a compression set test you did not run. A PDA in a vehicle mount in a hot climate is a harsher one. Cure state matters here: an under-post-baked peroxide-cured silicone typically has worse compression set than a properly post-baked one, which is why the oven schedule is worth asking about even though it never appears on a quote.
Elongation at break and modulus at 100%
Elongation tells you whether a lip can be stretched over the device without permanent deformation. Modulus at 100% elongation — the stress required to double the length of a test piece — correlates better with the effort a user feels installing and removing the case than durometer does. If your complaint is "too hard to get on and off," modulus is the number to move, not Shore A.
Choosing a target, and the trade-offs on either side
Protective case compounds commonly sit in the 40–70 Shore A band, and the choice is a set of trade-offs rather than a right answer. Treat the following as directional relationships to discuss with your supplier's engineering side, not as a specification to copy.
| As hardness goes down (softer) | As hardness goes up (firmer) |
|---|---|
| Grip and anti-slip performance improve | Grip drops; texture has to do more of the work |
| Energy absorption in a drop improves for a given wall | Impact is transmitted more directly to the device |
| Installation and removal get easier | Retention improves; the case stays on under load |
| Thin lips and cutout corners become more tear-prone | Thin sections survive handling better |
| Dimensional stability and edge definition soften | Bezel lips and button features hold shape better |
| Buttons feel mushy and travel further | Button actuation feels crisper but takes more force |
| Surfaces attract lint and dust more | Surfaces stay cleaner in the field |
A common resolution on rugged terminal cases is a firmer shell with softer, thicker corner bumpers — achieved either as a single compound with local geometry changes, or as a two-material build with a hard PC or ABS frame and a silicone or TPU overmould. A factory that runs both silicone and plastic lines and lists TPU, PC and ABS alongside its silicone grades can price that route as one part rather than as an assembly, which is worth asking about before you push a single-material compound to an uncomfortable extreme.
Writing it down, and the conversation to have with engineering
A hardness spec block you can paste into a drawing
Write it out in full. Ambiguity here is expensive later.
- Compound: [silicone grade / family], colour to [Pantone or approved master swatch].
- Hardness: 55 Shore A ± 5, per ASTM D2240, 15-second delayed reading, measured on a 6 mm moulded slab from the same compound batch and cure schedule as the production lot.
- Nominal wall thickness: back panel [x.x] mm ± [0.x], side rails [x.x] mm ± [0.x], corner bumper section [x.x] mm ± [0.x].
- Bezel lip height above screen plane: [x.x] mm ± [0.x].
- Tear strength: ≥ [value] kN/m per ASTM D624 Die B, per compound datasheet.
- Compression set: ≤ [value] % per ASTM D395 Method B, 22 h at [temperature] °C.
- Reference sample: golden sample dated [date], retained by both parties; production feel to match reference sample within the tolerances above.
- Lot verification: one slab per compound batch, results reported with the shipment documents.
- Cutout corner radii: minimum R[0.x] internal on all port and grille openings.
- Rejection basis: hardness outside band on the witness slab, OR wall thickness outside band at any of the [n] measurement points marked on the drawing.
Item 9 is the one that quietly saves the most money. Item 7 is the one that ends arguments, and it only works if both sides physically hold a sealed, dated golden sample — a practice covered in more depth in the CMH note on pre-shipment inspection in China.
Common ground with your supplier's engineering team
A supplier that carries an in-house engineering group — mould design, structural analysis, material selection — can usually be pushed past a catalogue answer if you ask the right question. WJM Silicone in Longgang, Shenzhen is the kind of case manufacturer this applies to: silicone compression and plastic injection lines under one roof, silicone, TPU, PC and ABS on the materials list, a 12-person engineering team handling 3D modelling, structural analysis and material selection, and secondary operations including surface treatment and printing done in-house.
What that means for you practically is that "what durometer should I use" is the wrong opening question. "Here is the drop height, the surface, the cradle it lives in, the ambient temperature range, and the approved sample — what compound and wall combination gets me there" is the right one, and it is a question an engineering team can actually answer.
What to ask the supplier next
- What Shore A range can you mould and hold in production for this compound family, and what tolerance band will you commit to in writing?
- Do you measure hardness on a moulded slab from the same batch, or on the part? Instantaneous or 15-second reading?
- Is the durometer calibrated, on a test stand or hand-held, and what is the calibration interval?
- How often is hardness checked — per batch, per lot, per shift — and at what sample size?
- Can you supply the compound datasheet with tear strength (ASTM D624) and compression set (ASTM D395) values, not just hardness?
- Do you post-bake this compound, and what is the schedule? What is the hardness shift before and after post-bake?
- Will a black and a light-coloured version of this compound measure the same, and by how much might they differ?
- What nominal wall thickness and tolerance is the current tool cut to, at the back panel, side rails and corners?
- If the spec moves from 50A to 60A after tooling, what happens to the shrink factor and therefore to fit on the device?
- Will you hold a sealed, dated golden sample and report per-lot hardness results with the shipping documents?
Question 5 separates factories that buy compound on a datasheet from factories that buy it on price alone, and it is a useful proxy for how the rest of the quality control process is run.