BLOG

on

Introduction

If you’ve got deck corrosion around coolant passages, you’re staring at one of the most expensive “small-looking” problems on a cylinder head. That crusty pitting around the water ports doesn’t just look ugly—it can break head gasket sealing, pull coolant into places it doesn’t belong, and turn a routine refresh into a full teardown.

The good news: some corrosion is fixable. The bad news: some of it is a hard “no”—even if the head almost looks salvageable.

1) Why this corrosion happens (and why it clusters near coolant passages)

Coolant passages are a perfect storm: mixed metals, heat cycling, and coolant chemistry. Over time, coolant inhibitors get used up and corrosion accelerates—especially if the engine ran straight water, weak mix, contaminated coolant, or neglected service intervals. Modern coolants are designed to include corrosion inhibitors, but they don’t last forever.

Common accelerators:

  • Low/old coolant inhibitor package → pitting and erosion

  • Galvanic corrosion (mixed metals, poor grounds, bad coolant mix)

  • Chronic overheating → gasket “micro-movement” + sealing surface damage

  • Poor previous prep (aggressive abrasives, gouging, uneven cleanup)

2) First question: is it cosmetic, sealing-risk, or structural?

Before you talk repair methods, classify the damage.

A) Cosmetic staining/light pitting (often OK):

  • Shallow speckling around ports

  • No “track” that leads to a fire ring (combustion seal)

  • No undercutting into the coolant port wall

B) Sealing-risk pitting (maybe repairable):

  • Pits near the gasket’s coolant sealing beads

  • “Channel” corrosion that can create a leak path across the deck

  • Corrosion wide enough that resurfacing alone may not clean it up

C) Structural dealbreaker (usually scrap/replace):

  • Corrosion that undercuts the coolant passage opening (thin, sharp edges)

  • Pitting that approaches combustion sealing areas/fire rings

  • Porosity opened up into the water jacket (won’t pressure test)

  • “Swiss cheese” deck where resurfacing would remove too much material

3) Repair option #1: Proper cleaning + measured resurfacing (best for mild-to-moderate)

If the corrosion is shallow, the fix may be straightforward:

  1. Chem clean / hot tank or ultrasonic (shop process)

  2. Inspect flatness

  3. Resurface to restore a uniform gasket finish

Important: Surface finish isn’t “whatever looks shiny.” It must match the gasket type. MLS gaskets typically need a smoother finish—Cometic notes ~50 Ra or finer, and Fel-Pro publishes guidance by material/application.

When resurfacing alone works:

  • Pitting cleans up without removing excessive material

  • No exposed porosity after cutting

  • Head still within spec for overall height and cam timing considerations

4) Repair option #2: TIG welding + machine (the “real fix” for deep localized pitting)

When corrosion is deep but localized, welding can restore missing material, then the deck gets machined flat. This is common on some engines where corrosion at the sealing surface is a known issue and welding minimizes how much material must be removed.

Best use case:

  • Aluminum heads (typically TIG)

  • Damage confined to coolant port edges/near deck, not a long crack network

  • You have a shop that understands cylinder head welding and post-weld machining

Two non-negotiables after welding:

  • Pressure test (don’t skip this)

  • Resurface to the correct Ra for your gasket style

5) Repair option #3: Epoxy/chemical patching (last resort, narrow use)

Yes, there are epoxies and chemical repair methods people use. They can sometimes help with non-critical external seepage areas or as a temporary/low-stress solution, but they’re not my go-to for a deck sealing surface that must clamp a gasket under combustion-level stresses.

If a “patch” is being considered on the deck itself, ask why. Usually it’s because the head is already beyond economical repair.

6) Dealbreakers: when you stop spending money and replace the head

Here’s the blunt truth: there’s a point where the head is no longer a “rebuild candidate.”

Replace the head if you have:

  • Corrosion that has crept toward combustion sealing/fire ring areas (risk of combustion-to-coolant leak)

  • Deep undercutting around the coolant port that leaves knife-edge metal

  • Porosity that opens into a water jacket after resurfacing

  • A head that won’t hold pressure during testing

  • A head that would require excessive milling to clean up (height/timing issues, poor clamp load distribution)

If you’re seeing any of the above, you’re usually better off with a quality replacement—new or remanufactured—than gambling on repeat failures.

7) What to do before the new/repaired head goes on (so it doesn’t happen again)

Deck corrosion around coolant passages is often the symptom. The cause is frequently the cooling system.

Do this on every install:

  • Flush the system properly (not just drain-and-fill)

  • Use the correct coolant type and mixture

  • Replace suspect grounds/straps if galvanic corrosion is suspected

  • Fix overheating at the source (fans, thermostat, radiator flow, cap pressure)

  • Match gasket type to surface finish (MLS especially)

Conclusion

Deck corrosion around coolant passages is one of those problems that punishes guessing. Mild pitting might clean up with proper machining. Deeper localized damage can often be saved with welding + surfacing + pressure testing. But once corrosion becomes structural—undercut ports, porosity into water jackets, or damage near combustion sealing—it’s usually a replacement call.

If you’re unsure whether your head is repairable or a dealbreaker, the fastest path is usually swapping in a quality replacement head and returning your core.

In-stock replacement options:

Check these articles:

on

If you’ve ever torqued a cylinder head “exactly to spec” and still ended up with a blown head gasket, warped deck, or mystery coolant loss… here’s a hard truth: dirty threads ruin torque accuracy. And the culprit is often hiding in plain sight—the bolt holes in the block.

Torque wrenches don’t measure clamp load. They measure resistance. If the threads are packed with rust, oil sludge, old threadlocker, coolant crust, or metal debris, your wrench can click at the right number while the head is clamped wrong. That’s how “followed the manual” turns into “why is it leaking?”

1) Torque spec isn’t clamp load (and dirt changes everything)

When you torque a head bolt, the goal is consistent clamp force across the head gasket. But torque is only an indirect way of getting there.

Dirty threads increase friction and create false resistance—so your torque wrench reaches spec before the bolt has stretched properly. Result: low clamp load, uneven sealing, and high risk of gasket failure.

On the flip side, contamination like oil pooled in the hole can create hydraulic pressure and artificially increase bolt tension (or crack the block in extreme cases). Either way, you’re not getting “true” torque.

2) The 5 common “thread killers” inside cylinder head bolt holes

Here’s what usually lives down there:

  1. Old oil and sludge (especially on high-mileage engines)

  2. Coolant residue (common on engines with coolant-passing head bolts)

  3. Rust and corrosion (stored blocks, marine, winter vehicles)

  4. Old threadlocker or sealant

  5. Metal debris (from machining, drilling, or previous bolt damage)

Any one of these can throw off your torque reading enough to matter—especially on MLS head gaskets and modern torque-to-yield (TTY) bolts.

3) The “bottomed out” bolt problem: the silent torque lie

A bolt hole full of crud shortens usable thread depth. That can make a head bolt bottom out early.

Your torque wrench climbs fast, you hit spec, and you think you’re golden… but what actually happened is the bolt stopped moving. The head didn’t clamp—your wrench just fought a dead stop.

This is one of the nastiest failure modes because it looks “correct” on paper.

4) How to clean bolt holes properly (the right way, not the lazy way)

You want clean, dry, consistent threads—without removing good material.

Best practice steps:

  1. Blow out loose debris (compressed air—wear eye protection)

  2. Brake clean / solvent flush to break oil and sludge

  3. Chase threads with the correct tool (more on this below)

  4. Flush again and blow dry

  5. Test-fit a clean bolt by hand (it should thread smoothly to depth)

Pro move: Put a rag around the hole before air-blasting so you’re not peppering the cylinder bores and deck with grit.

5) Thread chaser vs. tap: don’t “cut” when you should “clean”

A lot of people reach for a standard tap. That can be a mistake.

  • Thread chaser = reforms and cleans existing threads with minimal metal removal

  • Cutting tap = removes material, changes thread fit, and can reduce holding strength

If the threads aren’t damaged, use a chaser. Save cutting taps for repairs where threads are truly compromised.

6) Watch out for wet holes: when bolts need sealant

Some engines have head bolt holes that break into coolant passages. In those cases, cleaning matters even more—and you must follow the service procedure for sealant.

General rule (always confirm for your engine):

  • If the manual calls for thread sealer on specific bolts, use it.

  • Don’t “upgrade” to random threadlocker unless the OEM procedure says so.

7) Re-torque habits can backfire on modern engines

Old-school advice says “torque it, heat cycle it, re-torque it.” That does not apply universally today.

Many modern setups use:

  • TTY bolts (one-time use)

  • Angle torque procedures

  • MLS gaskets that don’t want repeated clamp disruption

If you re-torque TTY bolts, you can overstretch them and lose clamp force later. The safer play: new bolts (or studs) + clean holes + correct sequence.

8) The quick checklist before you torque a cylinder head

Use this every time:

  1. Deck and head surfaces clean and dry

  2. Bolt holes cleaned + chased

  3. Holes fully dry (no pooled oil/coolant)

  4. Correct bolts (new TTY if required)

  5. Correct lube/sealant per spec

  6. Torque sequence followed exactly

  7. Torque wrench verified / known-good

That checklist prevents a ridiculous amount of “brand new head gasket failed” stories.

Conclusion

Cylinder head installs live and die by clamp load. And the fastest way to sabotage clamp load is ignoring the crud hiding in the bolt holes. Dirty threads ruin torque accuracy because your torque wrench can’t tell the difference between friction and stretch—and your head gasket definitely can.

Clean the holes. Chase the threads. Follow the correct lube/sealant procedure. Then torque it like you mean it.

If you’re rebuilding, upgrading, or replacing a head, don’t gamble on a “maybe it’ll seal” install.

on

A cylinder head can be “rebuilt,” look spotless, and still nose over at RPM like it hit a soft limiter. That’s usually not ignition, fuel, or a “bad tune.” It’s often simple physics: the valvetrain can’t control the valve anymore. And the fastest way to catch that before it costs you an engine is valve spring testing—because “new springs” or “rebuilt head” does not automatically mean “correct pressures.”

If the springs are wrong (or installed wrong), the valves can float—they stop following the cam profile, bounce off the seat, and the engine loses power, starts misfiring, or worse.

9 Reasons a “Rebuilt” Head Can Still Float Valves

1) Seat pressure is too low (even if the spring is “new”)

Seat pressure is what keeps the valve planted on the seat and under control as RPM climbs. If it’s soft, the valve starts to loft, bounce, and float. Summit Racing’s help center breaks down why installed height and spring rate directly determine spring pressure—and how that ties to avoiding valve float.

2) Installed height wasn’t actually checked (or changed during the rebuild)

Installed height is the distance the spring sits at when the valve is closed. Change retainers, locks, valve length, seats, or shims—and installed height changes. That changes seat pressure, whether anyone intended it or not.

3) Open pressure doesn’t match your cam lift and RPM

Open pressure is what controls the valve at max lift and during the violent acceleration phase of the cam lobe. Engine Builder Magazine notes spring selection revolves around closed pressure, open pressure at max lift, and the rest of the clearance stack-up.

4) Coil bind clearance is too tight (or unknown)

Coil bind isn’t “maybe.” If the spring stacks solid at full lift, something gives—usually not the spring. Even if it doesn’t fully bind, running too close to bind spikes loads and kills stability. Summit explicitly calls out coil bind height and avoiding coil bind as critical.

5) Springs lose pressure over time (and some lose it fast)

Not all springs age the same. Heat cycles, quality, and material matter. Real-world measurements show “same part number family / replacement spring” can come in with noticeably different seat pressures at the same installed height.

6) Retainer-to-seal / retainer-to-guide clearance wasn’t verified

You can have “good pressures” and still crash parts at lift: the retainer hits the seal/guide before full lift, the valve stops, and everything goes unstable. Engine Builder calls out retainer-to-guide clearance as part of the required selection/verification set.

7) Spring rate isn’t right for the valvetrain weight

Heavier valves, retainers, locks, and pushrod/rocker dynamics need more control. Spring rate (lbs/in) is the “how fast pressure rises with lift” factor—and it changes everything about stability. Summit highlights how spring rate + installed height = pressure.

8) The rebuild used “catalog springs” instead of matching the cam and goal

A lot of rebuilds are built to stock assumptions. Add cam, add RPM, add boost, add heavier valves—those assumptions die. If your build is not stock, you want pressures verified against your actual lift, target RPM, and component weight.

9) Nobody actually tested the springs after assembly

This is the big one. You can do everything “right” and still end up wrong if you never measure. Spring testers exist for a reason: verify seat pressure at installed height and open pressure at your actual lift—not “what the box says.”

Quick “Do This” Checklist: How to Verify Before You Install

If a shop can’t tell you those numbers (or won’t), you’re gambling.

Want the “bolt-on and go” option?

If you’re tired of chasing mystery RPM breakup, start with a cylinder head that’s built to a higher standard—and buy from a place that backs it up.

(And if you’re already holding a “rebuilt” head you don’t trust, ask us what to measure—we’ll point you at the numbers that actually matter.)

Conclusion (with CTA)

“Rebuilt” is not a spec. Valve spring testing is. If your engine is floating valves, the head may look perfect while the spring setup is quietly wrong—seat pressure, open pressure, installed height, coil bind clearance, and retainer clearance are the difference between clean pulls and valve bounce.

Ready to stop guessing? Head over to our Shop for cylinder heads you can build around, and use our Contact page if you want help choosing the right setup for your RPM and cam.

on

Introduction

If a machine shop refuses to resurface your cylinder head, it’s rarely them “being lazy” or trying to upsell you. Most of the time, it’s a risk decision: they’ve spotted a condition where resurfacing won’t fix the real problem—or could create a bigger one (repeat head gasket failure, valvetrain geometry issues, or a comeback they’ll eat). This is what’s actually going on behind the counter, and how to respond like a smart buyer—not a hopeful gambler.

1) It’s already too thin (you’re at or past minimum spec)

Resurfacing removes material. Every head has a minimum thickness limit, and once you’re close, a “cleanup cut” can push it into the danger zone—reduced clamp load behavior, distorted sealing, and geometry changes. This is one of the most common legit reasons shops decline the job.

2) Warpage is beyond what resurfacing can safely correct

A mildly warped head? Often fixable. A severely warped head? You may need too much material removed to get it flat again (which circles back to thickness limits). Many shops will measure and refuse if it’s outside a reasonable correction window.

3) They suspect cracks, porosity, or a casting that won’t pressure-test

A head can look “fine” and still be junk once it’s hot, pressurized, and stressed. Shops that know what they’re doing don’t want to machine a head that fails crack detection or pressure testing afterward—because then resurfacing was wasted money and time. (Good rebuild operations pressure test as a standard step.)

4) The gasket you’re using demands a specific surface finish they can’t guarantee

This is the sneaky one. MLS gaskets are picky: if the surface finish is too rough, it can seep or fail—even if the head is “flat.” Cometic, for example, commonly calls for ~50 Ra or finer for MLS sealing. Fel-Pro also publishes Ra guidance ranges (different targets for cast iron vs aluminum). If the shop’s surfacer/process can’t reliably hit the right Ra, the safest answer is “no.”

5) The head has prior machining history (and you’re out of “future cuts”)

If the head has been resurfaced multiple times, the shop may refuse because there’s not enough material left for a proper correction—or because they can’t trust what’s been done previously (unknown cut angle/finish). That’s not paranoia—that’s experience.

6) Overheating damage isn’t just warpage—it can be softening or structural distortion

Especially on aluminum heads, overheating can change material properties or distort critical areas beyond the deck surface. The deck can be made flat, but the head can still be “wrong” where it counts (valve seats, guides, cam bores, etc.). Shops avoid resurfacing when they suspect deeper heat trauma.

7) Valve seat / guide issues mean resurfacing alone won’t solve your problem

If compression is leaking past valves, or seats are receded, a resurfaced deck won’t stop misfires, low compression, or hot spots. A responsible shop may refuse a “surface-only” job because it won’t fix the root cause—and you’ll blame the shop when it still runs poorly.

8) The head design is tricky (OHC timing geometry, multi-cam alignment, specialty castings)

On many overhead-cam engines, taking material off the head changes timing relationships and can require additional corrective steps (or special knowledge). Some shops simply choose not to own that liability if they don’t do that platform often.

9) Liability + comeback math: the job isn’t worth the risk

This is the blunt truth: resurfacing is relatively cheap compared to the cost of a comeback. If they think there’s a meaningful chance your head will fail after machining (or the gasket seal will repeat-fail due to finish/spec issues), the smart business move is refusing the job.

Conclusion

When a shop refuses to resurface, they’re usually telling you: “We don’t think this ends well.” The fastest way to waste money is forcing a surface cut on a head that’s too thin, too warped, cracked, overheated, or mismatched to your gasket’s finish requirements. The fastest way to win is treating resurfacing like one step in a measured plan: verify thickness/flatness, confirm crack/pressure integrity, and match surface finish to gasket type.

If your machine shop tapped out, don’t guess your way into a second teardown.