The choice between SMC pressing and spray transfer for composite battery covers usually gets settled on tooling price, and that is the wrong first question. A battery cover is a sealing surface first and a structural panel second. Its flange has to hold an adhesive or gasket joint flat enough to keep water out for the life of the pack, its fiber content has to stay consistent enough for the crash and dielectric case, and both have to repeat on part ten thousand. SMC pressing wins that argument in series production because a matched steel mold defines both faces at once. Spray transfer stays viable where volume is low, the cover is larger than your press can take, or the design is still moving.
What a battery cover has to hold before process choice matters
Every cover requirement ends up as a tolerance on the sealing path or on the panel, and both processes are judged against the same list.
| Requirement | What it governs | How to specify it |
|---|---|---|
| Flange flatness across the seal path | Adhesive or gasket compression, IP67 joint life | Flatness under clamp load at service temperature, on the seal path |
| Wall and rib stiffness | Panel oil-canning, modal behavior, handling damage | Deflection or modal target on the cover alone and bolted in the pack |
| Flame and thermal response | Occupant warning time, material classification | The rating and enclosure test basis the customer’s homologation cites |
| Dielectric strength | Isolation after humidity aging | Withstand and insulation resistance after damp heat cycling |
| Dimensional stability over temperature | Seal shear, bolt preload retention | Warp and expansion measured across the service range |
Homologation sets the outer boundary. GB 38031-2020 requires that occupants receive a warning at least five minutes before a thermal event in the pack becomes dangerous to the cabin [1]. The cover contributes to that outcome without producing it; cell spacing, the vent path and the thermal barrier do most of the work. What the cover must not do is fail early, feed the fire, or lose the seal while the pack is venting.
Flatness is the requirement most often specified badly. A flange measured on a free, unclamped cover tells you what the molder can hold. A flange measured under clamp load and at service temperature tells you whether the joint seals. Ask for the second, on the seal path, and write the measurement condition on the drawing before you send it out for quotes.

How SMC pressing controls the part in a closed mold
Sheet molding compound arrives as a compounded sheet: unsaturated polyester or vinyl ester resin, calcium carbonate filler, chopped glass, a thickener and an initiator, matured until it can be handled without running. The charge is cut to a pattern, placed in a heated matched mold, and pressed. Material flows outward to fill ribs and bosses while the resin cures, and the tool defines the top face, the bottom face and the flange in the same cycle.
Three consequences follow from that geometry.
The first is thickness control. Because both faces are steel, wall thickness is set by the tool rather than the operator, and the flange that carries the seal comes out as a molded surface. That is the single biggest reason SMC holds a sealing joint on a series program.
The second is filler loading. SMC tolerates high filler content, which lowers the resin fraction, reduces cost per part, and gives the formulator room to reach a flame rating without adding the loading of fire retardants that would otherwise hurt toughness. It also explains why an SMC rating does not travel: the rating belongs to the formulation, not to the fiber or to the process name.
The third is flow. Fiber orientation changes as the charge spreads, so properties are not uniform across the part. Flow fronts meet and form knit lines, and a knit line that lands in a rib root or across the seal path is a weak plane. Charge pattern development is where most of the trial time goes, because a pattern that fills a flat panel will not necessarily fill a deep rib without trapping air.
Where SMC pressing stops is at the press. Required tonnage scales with projected area, so a cover that is 20% longer and 20% wider than the last one needs roughly 44% more clamp force at the same clamp pressure. Undercuts cost tooling money. Thick bosses cure more slowly than the panel around them and can pull a sink mark onto the appearance face, which matters when the customer sees the cover from outside the vehicle.
What spray transfer deposits and what it leaves to the operator
Spray transfer builds the laminate by deposition. A chopper gun cuts roving and throws it into a resin stream, and the mixture lands on the mold; the operator rolls it out to wet the fiber and drive air out. Some shops spray into a closed mold or vacuum-bag the deposit before cure. Either way, the material arrives at the tool rather than being pre-formed into a sheet.
What that buys is real. One mold instead of a matched pair, no press, and thickness built only where the design needs it. Local reinforcement, encapsulated inserts and one-piece covers beyond any realistic press bed are all within reach, and a design change usually costs a new mold surface rather than a new tool set.
What it costs is repeatability, and the cost shows up in two places. The process defines one face. The mold side is smooth and dimensionally set; the back side is whatever the passes produced, so fiber content is set by gun output against pass speed and the rolling that consolidates the laminate varies with the person holding the roller. Void content and resin-rich zones follow the same source. A flange that has to seal usually gets machined afterward, and machining brings each part to the print without narrowing the run-to-run spread that the next part will show.
The second cost is environmental. With styrene-based polyester or vinyl ester resin, open spraying puts styrene into the workspace, so extraction, air monitoring and operator protection are part of the process cost rather than an overhead you can defer. A closed-mold variant with vacuum consolidation reduces that exposure and improves the laminate, but it removes the tooling advantage that made the process attractive in the first place.
Spray transfer fits pilot and bridge covers built while the SMC tool is being cut, annual volumes too low to amortize a steel mold, covers too large for an available press, thick monolithic sections, and repair laminates on returned packs. It fits a series battery cover poorly when the seal path flatness and the fiber content have to repeat on every part without a machining step.

Where the two processes diverge on the same cover
The table below compares the two routes on the points that decide a cover program. It compares mechanisms, not the two quotes sitting on your desk.
| Decision point | SMC pressing | Spray transfer | Do not infer |
|---|---|---|---|
| Sealing flange | Molded by matched steel on both faces | Mold face set, back face and flange thickness from passes, often machined after | That a machined flange matches an as-molded flange in run-to-run spread |
| Fiber content | Set at compounding, checked by burn-off on a sample | Set by gun output against pass speed | That one coupon’s glass content represents the whole cover |
| Cycle and labor | Press cycle with one operator loading a charge | Laydown and rolling, labor scales with cover area | That a spray cycle measured on a small panel transfers to a full cover |
| Flame performance | High filler loading reaches a rating with less additive | Higher resin fraction usually needs more retardant loading for the same rating | That an SMC rating carries over to a spray laminate |
| Capital | Matched mold plus press tonnage on projected area | Single mold, no press | That the lower tool quote is the lower project cost |
| Late design change | Charge pattern rework plus mold change | Cheaper mold change, sometimes a new surface only | That flexibility compensates for a tolerance the process cannot hold |
What the table cannot tell you is how your specific cover behaves. Rib depth, insert layout, the number of drops on the seal path and the customer’s test basis all shift the answer, and they shift it in ways a comparison of process names never captures. Treat the table as the list of questions to put to both suppliers, not as the verdict.
One fiber-related point deserves its own line. Reinforcement changes the process choice for structural parts, and the reinforced RIM family sits closer to SMC on clamping force than an unreinforced RIM system does.
When the cover design needs fiber reinforcement and a fast closed-mold cycle without SMC-level clamping force, the reinforced RIM family is the next comparison to make. <RRIM Vs RIM Which Process Suits Your Structural Part> covers where reinforcement changes the process choice for structural parts.

Tooling cost, press capacity and where each route stops paying
Tooling price is a one-time number. Part cost is where the routes separate, and it separates on labor, scrap and the machining step rather than on material price. An SMC quote carries an amortized matched mold, press time, and a charge pattern that has already been through its trial iterations. A spray quote carries laydown and rolling hours per part, flange machining, extraction equipment, and the scrap that comes from a laminate that must be inspected rather than assumed.
The crossover sits wherever the annual volume stops covering the mold. That point depends on your mold budget, your available press time and the loaded labor rate in your plant, so a rule of thumb borrowed from another factory will usually point the wrong way. The honest version is a two-line calculation: mold cost divided by annual parts, compared against the labor and scrap difference per part. If the mold cannot be paid back inside the program life, the SMC route is not the cheap one.
Press capacity usually decides before cost does. Projected area against available tonnage sets the largest cover you can mold, and a short-tonnage press shows up later as short shots at the far rib or porosity in the flange rather than as an obvious failure at the press.
A third route is worth a look before either is fixed. Resin transfer with a fiber preform gives you two defined faces from a liquid resin system, which suits a cover whose flange tolerance is tighter than a hand-laid laminate can hold.
Closed-mold liquid processes sit between the two routes discussed here, because a preform plus injected resin gives you two defined faces without a sheet charge. <SRIM Explained Structural Reaction Injection Molding Basics> covers how structural reaction injection molding handles fiber loading and flow before the resin enters the mold.
The flange, the adhesive bead and the fill behind the cover
The cover decision does not end at the mold. Two dispensed materials sit on either side of it, and both inherit the tolerance the process left behind.
On the sealing side, a two-component polyurethane adhesive or sealant is laid down as a bead and compressed by the clamp load. Bead volume and clamp together set the bond-line thickness, so the dispense rate has to track the robot path or the bead necks and starves a section of the seal path. A ratio error between the two components shifts cure speed and adhesion, and it does so quietly: the part passes leak test at the end of the line and fails in the field after thermal cycling.
Behind the cover, polyurethane potting or gap filler fills the space around modules and around the thermal interface. Shot weight per cavity, fill path and air entrapment decide whether that fill is uniform, and the number of cavities decides the pump size.
The fill behind the cover is a metering problem before it is a chemistry problem, because cavity count and shot size drive the pump choice and the cycle. <How To Calculate Shot Weight In Polyurethane> covers how shot weight is calculated for a given part and mix ratio.
On the dispensing side, Haifeng Polyurethane Machinery builds two-component platforms that hold dynamic metering within ±0.3% of full range and store formulas per part number, which is what makes a single line practical when it runs three cover variants in a week.

What to test on the first covers before releasing the design
Write the acceptance criteria before the trial, not after the parts arrive. These checks cover the points where the two processes actually differ, and each one answers a question the drawing alone cannot.
- Flange flatness on the seal path, clamped, measured at service temperature rather than in the inspection room.
- Wall thickness map on a sectioned cover, taken at the flange, the rib roots and the panel center.
- Glass content by burn-off on samples cut from the center, the edge, a rib root and the flange, not from one convenient trim scrap.
- Cross-section void content at the rib roots and along the seal path.
- Flame coupon cut from the actual production part, tested to the rating the drawing claims.
- Insulation resistance and dielectric withstand after damp heat, on a finished cover.
- Leak test using the production adhesive at the production dispense rate, not a laboratory bead.
- Warp on a bolted assembly through the thermal cycle the pack will see.
State what a pass covers. One cavity, one material lot and one post-cure cycle are inside the result; a second cavity, a reformulated resin or a change of roving supplier starts the record again. A supplier who reports a flame coupon from a laboratory plaque rather than a production part has tested the formulation, not your cover, and the difference is worth a sentence in the purchase specification.

What to send before you fix a dispensing layout
Send the cover drawing with the seal path flatness and its measurement condition, the cover material and the process route you chose, the adhesive or sealant chemistry with the required bead profile, the dispense speed range the robot can hold, the cavity count, the polyurethane potting or gap filler volume per pack with its mix ratio, the line takt, and the pack standard your customer’s homologation cites. Those fields are enough to size a metering system against the cover instead of against a brochure figure.
Which molding route you take stays with you and your molder. The metering and dispensing scope is what we specify, and it starts from the shot and the bead rather than from the machine model. Email Info@chinahaifeng.com or message WhatsApp 86 13566296633 with the drawing and the fill volume, and we will come back with the pump, ratio and control configuration that matches.
FAQ
Can a spray transfer cover pass the same flame test as an SMC cover?
Only on its own evidence. Flame performance follows the formulation, and a spray laminate carries a higher resin fraction than a filled SMC sheet, so it usually needs more retardant loading to reach the same classification. Cut the coupon from the production part, not from a laboratory plaque, and test to the rating the drawing claims. An SMC rating from a different supplier, a different resin or a different filler package tells you nothing about your laminate.
Is spray transfer cheaper at low volume?
On tooling, yes, because one mold and no press beat a matched steel pair. On part cost, it depends on how much flange machining and inspection the cover needs, and on the labor rate for laydown and rolling. Compare mold cost divided by annual parts against the labor and scrap difference per part. When the program is short, the tooling saving usually wins; when the flange has to hold a tight seal path on every part, the machining step can erase it.
How flat must the sealing flange be?
The number comes from the joint, not from an industry default. Gasket compression, adhesive bond-line thickness, clamp load and the temperature the pack sees in service together set the allowable deviation, and the tolerance should be stated on the seal path only. Specify it clamped and at temperature. A flatness figure quoted on a free part is a molding capability statement, and it does not predict whether the joint seals once the cover is bolted down.
Can a spray design be converted to SMC later?
Expect a redesign rather than a tooling swap. SMC needs draft, a charge pattern and a rib layout that fills without trapping air, and the flange moves from a machined surface to a molded one, which changes the tolerance stack on the seal. The flame evidence also has to be rebuilt for the new formulation. Conversion is common between prototype and series, but it belongs in the program plan from the start, not in a change request after the pilot.
Where does RTM or SRIM fit between these two?
It sits between them on both tooling and repeatability. A preform plus injected resin gives two defined faces and a controlled fiber loading without the clamping force an SMC charge needs, which suits covers with deep ribs or a tight seal path. Tooling costs more than a spray mold and less than a fully developed SMC program in many cases. The trade is a preform step that has to be shaped and placed, and that step is where the process earns or loses its cycle time.
References
[1] GB 38031-2020, “Electric vehicles traction battery safety requirements,” Standardization Administration of China, 2020.
If you’re interested, check out these related articles:
Measuring Polyurethane Foam Resilience A Practical Test Guide
SRIM Explained Structural Reaction Injection Molding Basics
What Is Spray Polyurethane Foam Composition And Application
How To Calculate Shot Weight In Polyurethane



