Your print says “Electroless nickel per ASTM B733, Type V, SC3, Class 1”. Here is what each of those three fields controls, what the phosphorus range actually buys you, how AMS 2404 differs, and how to write the callout so nobody has to guess.
ASTM B733, Standard Specification for Autocatalytic (Electroless) Nickel-Phosphorus Coatings on Metal, covers nickel-phosphorus alloy deposited from aqueous solution onto metallic products for engineering — that is, functional — use. Not decorative nickel, and not electrodeposited nickel; if you want that, you want electrolytic nickel plating instead.
The designation has three independent fields, and engineers routinely fill in one and forget the rest. Type is the phosphorus content of the deposit — the alloy, and it sets almost every property you care about. Service condition is the minimum thickness. Class is the post-plate heat treatment, if any.
B733 also puts a burden on the buyer. Stress relief before plating, embrittlement relief after, significant surfaces, peening and substrate tensile strength all apply when specified by the purchaser. They are not automatic. A drawing naming only the standard has left those decisions to whoever is holding the part. The electroless nickel process page covers how the bath works; this page is about the callout.
This is the most looked-up table in the standard. Phosphorus is stated as weight percent of the deposit.
| Type | Phosphorus, % by weight | Commonly called | Structure as plated |
|---|---|---|---|
| I | No requirement for phosphorus | Unspecified | Whatever the shop’s bath produces |
| II | 1 to 3 | Low phosphorus | Microcrystalline |
| III | 2 to 4 | Low phosphorus | Microcrystalline |
| IV | 5 to 9 | Mid phosphorus | Mixed; amorphous above roughly 7 % |
| V | 10 and above | High phosphorus | Amorphous |
Types II and III overlap between 2 and 4 %. That is in the standard, not a typo. Type I is the trap: it places no phosphorus requirement on the deposit at all, so a Type I part can arrive anywhere from 2 % to 12 % phosphorus and be compliant. If any property below matters, do not order Type I.
The class is a post-plate thermal treatment. Class 2 is a hardening bake. Classes 3 through 6 are adhesion bakes, sorted by what the part is made of, and Class 3 doubles as the embrittlement relief for steel.
| Class | Temperature | Time | What it is for |
|---|---|---|---|
| 1 | None | — | As-deposited. No heat treatment. |
| 2 | 260–400 °C (500–750 °F) | 1–20 h | Hardening. Class 2 carries a hardness requirement: not less than 850 HK100, or the Vickers equivalent. |
| 3 | 180–200 °C (355–390 °F) | 2–4 h | Adhesion on steel, and hydrogen embrittlement relief. |
| 4 | 120–130 °C (250–265 °F) | 1–6 h | Adhesion on age-hardened (heat-treatable) aluminum alloys and on carburised steel. |
| 5 | 140–150 °C (285–300 °F) | 1–2 h | Adhesion on aluminum, copper and beryllium alloys. |
| 6 | 300–320 °C (570–610 °F) | 1–4 h | Adhesion on titanium alloys. |
Classes 4 and 5 exist because the bake has to sit below what the substrate can take. A T6 aluminum part baked at 190 °C is not a T6 part any more, which is why age-hardened alloys get the 120–130 °C treatment and non-age-hardened ones can take 140–150 °C.
| Service condition | Minimum thickness | Approx. | What it is really for |
|---|---|---|---|
| SC 0 — minimum | 0.1 µm | 0.000004 in | A flash. Contact and specialized electronic use, not corrosion protection. |
| SC 1 — light | 5 µm | 0.0002 in | Indoor, dry, clean. Solderability and conductivity work. |
| SC 2 — mild | 13 µm | 0.0005 in | Indoor with condensation, light handling and mild wear. |
| SC 3 — moderate | 25 µm | 0.001 in | Outdoor exposure, industrial atmospheres, hydraulic and pneumatic hardware. The workhorse. |
| SC 4 — severe | 75 µm | 0.003 in | Aggressive chemical service, downhole and process equipment, sustained wear. |
Thickness and corrosion life are not linearly related, because electroless nickel is a barrier coating and not a sacrificial one. Zinc protects a scratch; nickel does not. A porosity path puts electrolyte on the steel with a large nickel cathode beside it, which pits fast — which is why 25 µm outperforms 13 µm by more than the ratio suggests.
AMS 2404, Plating, Electroless Nickel-Phosphorous, is the SAE aerospace material specification for the same coating. It is not a competing version of B733. It controls a different set of things.
B733 is a deposit specification: it classifies what you get — alloy, thickness, heat treatment — and points at test methods. AMS 2404 is an aerospace process specification: it leans on solution control, lot acceptance, periodic testing and pre-plate condition, in the way an aerospace plating flow-down expects. The differences that catch people out:
AMS 2404 has no phosphorus type. It allows a broad nickel-phosphorus range — nominally about 1 to 13 % — and gives you no field in which to specify the band you need. If phosphorus matters, say it separately or cite B733 alongside. SAE’s low-phosphorus companion, AMS 2405, is the document that constrains phosphorus downward.
The class numbers do not correspond. AMS 2404 has four, all thermal: Class 1 no post-plate treatment except embrittlement relief; Class 2 a hardening bake at 450 °F and above; Class 3 an adhesion bake near 375 °F for non-heat-treatable aluminum; Class 4 near 250 °F for heat-treatable aluminum. An AMS 2404 Class 2 is not a B733 Class 2 — the hardness floor alone is lower, 800 HK100 rather than 850. Default thickness works differently too: AMS 2404 sets a minimum that depends on the substrate and has moved between revisions, so read the revision your drawing cites rather than carrying a number across from memory.
If a drawing cites both, treat AMS 2404 as governing process and quality and B733 as the vocabulary for the deposit, and reconcile the class numbers on the purchase order — “AMS 2404 Class 1, B733 Type V, SC 3” is unambiguous; “Class 2” alone is not. If it cites neither and says only “electroless nickel”, you have specified nothing: no phosphorus, no thickness, no heat treatment. Three shops will hand you three different coatings, all of them honest.
Everything else on the callout is bookkeeping. This is the engineering choice, and there is no range that wins on every axis. High phosphorus wins on corrosion. Low phosphorus wins on as-plated hardness and on caustic exposure. Mid phosphorus is the default because it loses to neither by very much.
| Low — Type II/III, 1–4 % P | Mid — Type IV, 5–9 % P | High — Type V, 10 %+ P | |
|---|---|---|---|
| Structure | Microcrystalline | Transitional | Amorphous — no grain boundaries |
| Hardness as plated | Highest of the three | Middle | Lowest of the three |
| Hardness after Class 2 bake | High | High | High — all three converge near 1000 HV |
| Acid corrosion | Poor | Good | Best. Chloride, CO2, H2S service. |
| Alkaline / caustic | Best | Good | Weakest of the three |
| Magnetic response | Ferromagnetic | Weak; falls off with phosphorus | Non-magnetic |
| Solderability | Good; best electrical conductivity | Good | Good |
| Deposition rate | Slower | Fastest | Slower |
Hardness. As-plated nickel-phosphorus sits around 500 to 600 HV, and hardness falls as phosphorus rises. A Class 2 bake collapses the difference: all three reach roughly 1000 HV. So if the part is going to be baked anyway, “low phosphorus for hardness” has already been settled by the oven. Where low phosphorus still wins is a part that cannot be baked.
Corrosion, and why it flips. The amorphous high-phosphorus deposit has no grain or phase boundaries, so there is nothing for an acid to attack preferentially — published salt-spray comparisons put it well ahead of anything at 9 % or below. Reverse the pH and the ranking reverses: phosphorus is attacked in strong alkali, so a low-phosphorus deposit is the one that survives a caustic wash or a high-pH process stream. High phosphorus called out reflexively for a part that lives in caustic is the most common error on this specification.
Magnetic response. Deposits go effectively non-magnetic once the structure goes amorphous, and sources put that transition anywhere from about 7 % to 11 % phosphorus. If non-magnetic is a requirement rather than a preference — sensor housings, instrument bodies, medical device hardware near imaging, RF and microwave parts — call Type V and put a maximum relative permeability on the drawing rather than trusting the type number.
Stress and fatigue. Deposit stress varies with phosphorus, and published sources genuinely disagree about the low-phosphorus end. What is consistent: high-phosphorus deposits from a fresh bath are compressive or near neutral and drift tensile as the bath ages, and a tensile deposit is the one that cracks and takes the substrate with it. Either way, treat electroless nickel as a fatigue debit on a cyclically loaded part. One published set of figures on a high-strength low-carbon steel: endurance limit 342 MPa bare, 260 MPa with 25 µm of electroless nickel, 176 MPa after a 400 °C hardening bake. Hardening the coating made fatigue worse, not better. Shot peening before plating is the usual countermeasure, and B733 has a field for it.
Solderability and weldability. All three ranges solder with a mildly activated rosin flux, and low phosphorus is the pick where contact resistance matters too. Welding is the opposite story: phosphorus contaminates the weld pool and gives porosity and hot cracking, so weld first and plate after. Where the sequence cannot change, mask the weld prep and say so on the drawing.
There is no current, so there is no current distribution, so there are no high-current-density edges and no starved recesses. The reaction is autocatalytic: it proceeds at the same rate on every wetted, catalytic surface. A bore, a blind hole, an internal thread, a cross-drilling and the outside diameter all build at the same rate.
The consequence for tolerancing is what engineers actually care about. With a racked electrolytic deposit you get a distribution: you specify a minimum on a significant surface, accept whatever the edges do, and machine or mask around it. With electroless nickel you get a number. Subtract the build from the machining dimension and hold a bilateral tolerance on a plated bore. The deposit grows on both walls, so a bore closes by twice the thickness and a diameter grows by twice the thickness.
Two caveats that are not marketing: deep blind holes and long small-bore tubes still need solution exchange, and a feature that never wets — trapped air in an upward-facing blind hole — does not plate at all. Both are fixturing conversations that belong at RFQ. Our design guide covers what else to put on a plating drawing.
The hardening bake precipitates nickel phosphide out of the deposit. Hardness roughly doubles, from 500–600 HV as plated to about 1000 HV, which puts electroless nickel in the same conversation as hard chrome for wear. B733 Class 2 allows 260–400 °C for 1 to 20 hours and requires at least 850 HK100; the peak lands near 400 °C for about an hour, with 340 °C for four to six hours a common lower-temperature route.
Three costs, all to be checked before the bake goes on the drawing. Corrosion resistance: hardening a high-phosphorus deposit crystallises the amorphous structure that made it corrosion resistant, and the volume change opens microcracks, so you cannot have maximum hardness and maximum corrosion resistance in the same deposit. Substrate temper: 400 °C is above the tempering temperature of much hardened steel and well above what a heat-treatable aluminum alloy tolerates — which is why B733 splits Classes 3 through 6 by substrate. Fatigue: the hardening bake makes the fatigue debit larger, not smaller.
Electroless nickel adheres well to properly prepared metal and not at all to an oxide. Almost every adhesion failure is a preparation failure, and it shows up late — days later, or at the first thermal cycle, not at final inspection.
Aluminum cannot be plated directly; it re-oxidises in seconds. The route is a zincate, an alkaline immersion that strips the oxide and lays down a thin zinc film which the hot nickel bath then displaces. A double zincate — zincate, strip in nitric, re-zincate — is more reliable across the widest range of alloys: most of the etching happens on the first pass, so the second film is thinner and more uniform. Alloy chemistry matters. Silicon and copper constituents etch at different rates from the matrix, and over-etching opens capillary voids around them that trap zincate and bleed it into the hot bath, giving pits and skip plate. High-silicon castings and high-copper 2000-series alloys should be flagged at RFQ, not discovered.
Stainless steel is the other one. Its passive film is the reason it is stainless and the reason nothing sticks to it, so it needs activation — acid activation or a nickel strike — immediately before plating, with no rinse-and-wait in between. Get that wrong and the deposit looks perfect and lifts later. If the part wants a passive film rather than a coating, that is passivation on stainless steel.
On both substrates the adhesion bake is part of the process, not an extra — B733 Class 4 or 5, AMS 2404 Class 3 or 4. It relaxes deposit stress and drives off hydrogen absorbed at the interface.
Electroless nickel generates less hydrogen at the part than a cathodic electroplating process does, but the bath is hot, acidic pickles precede it, and hydrogen still enters the steel. On hardened steel that is enough to matter.
B733 draws the line at 1000 MPa ultimate tensile — about 31 HRC. Before plating, parts above that strength that were machined, ground, cold formed or cold straightened after heat treatment get a stress relief, schedules from Specification B849. After plating, they get an embrittlement relief bake per Guide B850, starting preferably within 1 hour and not more than 3 hours of plating — a tighter clock than most engineers expect, and tighter than the four hours AMS 2404 works to.
The catch is the wording: in B733 both are triggered when specified by the purchaser. Put substrate and hardness on the drawing and call the bake out — “4340, 45 HRC, bake for hydrogen embrittlement relief per ASTM B850” — and the requirement exists. Leave it off and, read strictly, it does not. Very hard steel is a written-agreement conversation before the parts ship, not after.
A complete callout names the standard with its revision, then all three fields — type, service condition, class — then what the standard leaves to the purchaser: significant surfaces, substrate and hardness, bake, and any property you are actually buying.
Electroless nickel per ASTM B733-22, Type V, SC 3, Class 1. Phosphorus 10–12 % by weight. Significant surfaces include all internal bores and cross-drillings. Thickness to be verified in the 12 mm bore.
High phosphorus for chloride and H2S. Class 1 because a bake would crystallise the deposit and give back the corrosion resistance the part is being plated for. Naming the internal features makes the uniformity claim testable.
Electroless nickel per ASTM B733-22, Type IV, SC 3, Class 2 (min 850 HK100). Substrate 4140, 42 HRC. Bake temperature not to exceed part tempering temperature; confirm before processing. Stress relieve prior to plating and bake for hydrogen embrittlement relief per ASTM B850.
Mid phosphorus, hardened. The temper limit is stated because the Class 2 window runs to 400 °C and this part cannot see it. Both heat treatments are called out because B733 makes them the purchaser’s to specify.
Electroless nickel per ASTM B733-22, Type V, SC 2, Class 4. Substrate 6061-T6; double zincate pretreatment. Deposit to be non-magnetic; relative permeability ≤ 1.01. Solderable per assembly drawing.
Class 4 keeps the adhesion bake at 120–130 °C so the T6 temper survives. The permeability limit is written as a number because “non-magnetic” is not a specification requirement anywhere in B733.
Gleco Plating has been finishing metal in Texas since 1979, family-owned, with plating plants in Rowlett and McAllen. Electroless nickel is what we run when a dimension inside the part has to be right.
Typical work: hydraulic and pneumatic bodies, valve components and manifolds for industrial customers; enclosures, connector bodies and heat-transfer hardware for power generation and data center programs; instrument and housing work for medical device customers where non-magnetic and cleanable matter more than appearance. If you are quoting the western side of the Metroplex, electroless nickel plating for Fort Worth covers the same line, and our locations page has both plants.
Where electroless nickel is the wrong answer we will say so. If the requirement is thick, ductile, low-stress nickel for build-up or salvage rather than a uniform thin barrier, that is sulfamate nickel, which runs at our McAllen plant. For decorative or general-purpose nickel on simple geometry, electrolytic nickel is usually cheaper per part.
What we ask for at RFQ, because it changes the routing: type, service condition and class as written; a phosphorus range if the type band is not tight enough; substrate alloy and hardness; which surfaces are significant, including internal ones; the substrate’s temper limit if the part is to be baked; and whether it has to be non-magnetic, solderable or welded afterwards. If a drawing says only “electroless nickel”, we come back with these questions before parts run rather than picking for you.
Gleco is AS9100D and ISO 9001:2015 certified, ITAR Registered, DFARS compliant and RoHS compliant; the AS9100D certificate covers three sites. Detail on certifications and approvals.
Start with mid phosphorus — Type IV, 5 to 9 % — unless the part gives you a reason not to. Go to high phosphorus, Type V, if the service is acidic, chloride-bearing, sour, or if the part must be non-magnetic. Go to low phosphorus, Type II or III, if the part sees caustic or high-pH cleaning, or if it needs maximum hardness and cannot be baked. Do not order Type I: it places no phosphorus requirement on the deposit at all, so any of the three could arrive and still be compliant.
Yes, and that is the main reason to use it. There is no current, so there is no current distribution — the reaction is autocatalytic and runs at the same rate on every wetted, catalytic surface. Internal diameters, blind features, cross-drillings and internal threads build at the same rate as the outside. The two practical limits are solution exchange in deep blind holes and long small-bore tubes, and trapped air in a feature that never wets. Both are fixturing questions, so mark the internal significant surfaces on the drawing.
It depends on the phosphorus. Low-phosphorus deposits are microcrystalline and ferromagnetic. As phosphorus rises the structure goes amorphous and the magnetic response falls away; high-phosphorus Type V deposits are treated as non-magnetic. Published sources put the transition anywhere between about 7 % and 11 % phosphorus, so if non-magnetic is a requirement rather than a preference, specify Type V and add a maximum relative permeability to the drawing. B733 itself has no magnetic requirement.
As plated, roughly 500 to 600 HV, with the harder end at low phosphorus. After a Class 2 hardening bake — 260 to 400 °C, with the peak near 400 °C — nickel phosphide precipitates and hardness reaches about 1000 HV, comparable with hard chrome. B733 Class 2 requires a minimum of 850 HK100. The costs are real: the bake reduces corrosion resistance in high-phosphorus deposits, it has to stay below the substrate’s tempering temperature, and it makes the fatigue penalty larger.
Yes, over a zincate pretreatment — aluminum re-oxidises too fast to be plated directly. A double zincate (zincate, strip, re-zincate) is the more reliable route across the widest range of alloys. High-silicon castings and high-copper 2000-series alloys are the awkward ones and should be identified at RFQ. The adhesion bake is part of the process: B733 Class 4 at 120–130 °C for age-hardened alloys such as 6061-T6 or 7075-T6, Class 5 at 140–150 °C where the temper allows it.
B733 classifies the deposit — type for phosphorus, service condition for thickness, class for heat treatment — and points at test methods. AMS 2404 is the aerospace process specification and puts its weight on solution control, lot acceptance, periodic testing and pre-plate condition, but gives you no way to specify a phosphorus band; it allows a broad range and leaves it at that. The class numbers are not interchangeable — AMS 2404 Class 2 requires 800 HK100, B733 Class 2 requires 850. If a drawing cites both, state the two designations separately on the purchase order.
Under B733, steel above 1000 MPa ultimate tensile — about 31 HRC — gets a stress relief before plating per B849 and an embrittlement relief bake after plating per B850, and the bake is to start preferably within 1 hour and not more than 3 hours of plating. The important part is that B733 makes both of these apply when specified by the purchaser. They are not automatic. Put substrate and hardness on the drawing and call the bake out by name.
Solder, yes — all three phosphorus ranges take solder with a mildly activated rosin flux, and low phosphorus is the pick when contact resistance matters too. Weld, no. Phosphorus in the deposit contaminates the weld pool and produces porosity and hot cracking, so weld the assembly first and plate afterwards. If the sequence cannot change, mask the weld prep and show the masked area on the drawing.
Type, service condition, class, substrate and hardness — and tell us which surfaces are internal. If the callout is incomplete or the phosphorus range fights the service environment, we will raise it before the parts run.
Precision metal finishing since 1979. Plating, anodizing, coatings and passivation from three Texas facilities.
Our paint, powder & Cerakote division: Gleco Paint & Powder Coating
Gleco Rowlett
2220 Grisham Drive
Rowlett, TX 75088
972-475-4300
Gleco McAllen
3800 W. Ursula Ave
McAllen, TX 78503
956-800-4069
