High-Tear Strength Liquid Silicone (LSR) vs. RTV: When to Upgrade Your Tooling

High-tear LSR and high-tear RTV are not two grades of the same tool. LSR is a heat-cured, injection-molded part. The tool is steel or aluminum. RTV is a room-temperature poured mold rubber. The tool is the silicone.

Shops upgrade at the wrong moment because a datasheet printed a large ASTM D624 number next to the word silicone. Tear on an LSR keypad is why the webbing survives a nick in service. Tear on a platinum RTV glove mold is why the cavity survives a resin pull. Those numbers are not interchangeable, and neither one is a cycle-life guarantee.

Stay on platinum when the job can cure. Tin-cure RTV can be high-tear on paper and still shrink in the library. Peroxide HCR is a millable factory path, not a poured mold. Do not mix those cycle stories.

Two tear numbers, two objects

ASTM D624 (or ISO 34-1) measures how a nicked specimen resists unzipping. Die B and Die C are not one scale. Read the die before you compare a “high-tear” LSR to a mold rubber.

LSR tear is a property of the article you sell: bellows, baby-bottle nipples, face-mask skirts, connector seals, thin undercut parts that must stretch off a core. Heat cure and a tight network are why production LSR often sits above a poured RTV of the same Shore A on tear and tensile. That is a part decision.

RTV tear is a property of the cavity you pour into resin, wax, plaster, or a low-melt. A high-tear platinum 20A skin inside a mother mold is a classic undercut tool. A stiff 40A with modest tear will split at the first negative draft even if tensile looks fine.

If the RFQ says “high-tear silicone mold” and the deliverable is 10,000 LSR gaskets, the buyer asked for the wrong object. If the deliverable is a block mold for twenty urethane masters, LSR injection is the wrong object.

What high-tear LSR actually buys

LSR is two-part platinum addition rubber, meter-mixed, injected into a hot tool, cured in seconds to a couple of minutes depending on thickness and grade. High-tear grades exist specifically so thin sections and undercuts can demold without tearing the part, and so nicked parts in the field do not run.

Upgrade into LSR (from RTV-cast prototypes or from a soft-tool urethane stand-in) when:

• The geometry is the production geometry, including knit lines and gate vestige you will live with.
• Volume makes a steel or aluminum tool cheaper per piece than pouring RTV and casting a stand-in.
• The spec is medical, food-contact, or a documented platinum LSR compound with a post-cure.
• Dimensional life has to hold across lots, not across twenty pulls of a block.
• You need overmolding onto PC, PA, or metal that RTV cannot do as a production process.

High tear does not replace hardness. A 70A high-tear LSR still will not invert like a 20A glove. A 20A high-tear LSR still needs a tool that can fill and vent that viscosity. Buy Shore A, tear (die stated), elongation, and the process together.

LSR does not replace a poured mold for a one-off sculpture, a texture copy, or a master you will throw away. The steel is the cost. The rubber is not.

When poured platinum RTV remains the right tool

RTV-2 platinum high-tear grades exist because undercut masters and resin casting are real work. They win when:

• The “tool” must stretch off a locked master, then rebound.
• You need the cavity this week, not after steel.
• The production article is not silicone. You are molding something else in a silicone cavity.
• Lot size is prototypes, bridge, or texture library work.

Stay with platinum RTV if the master is clean enough to cure. If sulfur clay, amines, or undercured SLA poison the face, high-tear platinum has zero life. Switch to tin for a copy mold, then platinum for the production cavity. Tear rating does not cure inhibition.

Tin-cure high-tear is a short-run and dirty-master rubber. Library shrinkage still happens. Do not archive it as if it were platinum. Do not quote its tear as an LSR part spec.

A hybrid that shops actually run: RTV high-tear glove for the first geometry freeze, then LSR in aluminum when the CAD stops moving. That is an upgrade of tooling, not a swap of tear numbers on the same PO.

The upgrade triggers that are not “more Shore A”

Raise hardness and you often lose the stretch that made the undercut possible. Tear and elongation are the stretch story. Hardness is the cavity-support story.

Upgrade out of RTV tooling when any of these show up on the floor:

• You are replacing block molds faster than you are shipping parts.
• Flash, gate, and knit-line requirements belong on a steel tool, not a poured split.
• The customer’s incoming inspection is DIN ISO 3302-1 / ISO 3302 on a molded elastomer part, not “looks like the master.”
• Cleanroom, extractables, or a named ISO 10993 / USP Class VI compound are on the drawing. Those are LSR or HCR articles, with the device maker still owning the biological evaluation.
• Undercuts on the part need a collapsing core or a stretch-off in production LSR, which is a tool design problem, not a bigger bucket of RTV.

Stay on RTV when the master will change twice more, when the casting resin is the product, or when twenty parts is the entire job.

ASTM D624 is not a cycle count

Vendors print tear because undercuts nick rubber. They do not print “this mold survives 200 urethane pours” as a standard. Resin exotherm, release chemistry, and how you lever the part out decide the production clock. A high-tear LSR part in a polished cavity can still tear if the core has a sharp corner and no draft. A high-tear RTV skin can still split if the mother mold does not support it.

Compare grades only:

• Same cure family (platinum RTV vs platinum RTV; LSR vs LSR)
• Same die (B or C) and same unit (kN/m or ppi)
• Same hardness band, or admit you changed two variables

Do not paste an LSR tear onto an RTV RFQ to “make the mold stronger.” If the mold is tearing, change the RTV grade, the thickness of the skin, the mother mold, or the pull. If the part is tearing at the gate or the undercut, that is an LSR tool and grade conversation.

Sentence for the specification

“If the deliverable is a poured cavity: platinum-cure RTV, high-tear grade, Shore [A], ASTM D624 die [B/C] per grade TDS, vacuum degassed. If the master inhibits platinum, tin-cure copy then platinum production mold. If the deliverable is the elastomer article: platinum LSR, high-tear grade, Shore [A], D624 die stated, in [steel/aluminum] production tooling. Do not substitute LSR tear data for RTV mold life, or RTV mold-rubber data for LSR part specs.”

That sentence stops the most expensive mix-up in this category: buying a drum of mold rubber when the customer needed a press, or cutting steel when the customer needed a glove mold.

FAQ

Is high-tear LSR just a stronger version of platinum RTV mold rubber?

No. LSR is injection-molded and heat-cured in metal tooling. RTV is poured at shop temperature and *is* the tooling. Both can be high-tear on ASTM D624. They are different objects on the purchase order.

When does a shop outgrow RTV block molds for silicone parts?

When geometry is frozen, lot size makes metal tooling cheaper per piece, and the drawing wants production flash, gates, and a named LSR compound. Bridge runs and changing masters still belong in RTV.

Can I raise Shore A instead of buying a high-tear grade?

Usually not for undercuts. Higher hardness fights stretch. Tear and elongation decide whether a nick at a negative draft unzips. Buy hardness for cavity support or sealing force, and tear for nicked geometry.

Does a high ASTM D624 number guarantee long mold life in urethane?

No. Tear is a nicked-specimen test. Urethane exotherm, release, and how the part is levered out set the production clock. Use the number to compare grades in the same family and die, not as a cycle quote.

Should medical silicone parts be prototyped in tin-cure high-tear RTV?

Not as the device material. Tin-cure is a mold rubber for dirty masters and short copies. Medical articles almost always start as platinum LSR or platinum HCR. You can still use RTV as *tooling* to cast a stand-in, if the stand-in is clearly not the device compound.