We hear the exothermic welding vs clamps debate weekly, and the deciding factor is not day-one performance — both can measure beautifully at commissioning. It is what happens to ground connection resistance after 20 years of soil chemistry, thermal cycling and fault currents. I am Jane Yang, sales manager at Xinchang Shibang New Material, a factory that has manufactured both exothermic welding systems and mechanical grounding clamps for more than 17 years, so we have no stake in pretending one technology wins everywhere. In this guide we compare how welded and bolted joints actually age underground, what IEEE 837 and utility specifications demand, and the specific situations where we still recommend a quality clamp over a weld.
- We remind buyers that a bolted ground clamp relies on contact pressure; US NRC aging literature (NUREG-1800) lists loosening of bolted connections from thermal cycling and increased connection resistance from oxidation or loss of preload as documented aging mechanisms.
- In our molds, an exothermic weld fuses conductors into one molecular joint — there is no interface to oxidise and no bolt to re-torque, which is why utilities like Arkansas Electric Cooperative mandate exothermic connections underground and inspect them before backfilling.
- IEEE 837 qualifies permanent substation grounding connections — the welds within the grid, the ground leads joined to it, and the leads joined to equipment — and passing it is the strongest evidence a joint will hold resistance over the installation’s life.
- Our advice over a 20-year horizon follows engineering practice: re-test buried earthing connections every 5 years — every 3 years in aggressive soil — a maintenance cycle a welded joint largely escapes.
- We still specify mechanical clamps above grade, in inspectable positions and on retrofit work; our brass and bronze clamps carry a rated service life above 50 years when installed where they can be checked and re-torqued.
Why Connection Resistance Decides Whether an Earthing Joint Survives 20 Years
We describe every earthing system to new buyers as a chain: electrode, conductor, and the joints between them. The electrode can be perfect and the conductor oversized, yet the system still fails if one joint’s resistance climbs. Fault current and lightning current both take the path of least impedance, and a degraded joint forces that current through a bottleneck — heating it further and accelerating its decay.
In our training sessions, the physics of a bolted joint is where we start. Current crosses a clamped interface only through microscopic contact points; even well-machined surfaces touch at asperity peaks covering a small fraction of the apparent area. Constriction resistance at those points, plus film resistance from oxides and soil chemistry, is what a commissioning test actually measures. Anything that reduces contact pressure or thickens the film — creep, thermal cycling, corrosion — raises the joint’s resistance, and the added heat from I²R losses feeds the loop.
We find that framing changes the comparison entirely. We tell distributors a weld and a clamp are not two prices for the same function; they are two different aging profiles — one with a metal interface that physics will attack, and one without an interface at all.
What Two Decades Underground Does to a Bolted Clamp
We rely on well-documented engineering literature when we explain how bolted connections age. The US Nuclear Regulatory Commission’s aging-management reference, NUREG-1800 (the Generic Aging Lessons Learned report) — the document utilities use to justify decades of extended plant operation — lists for switchyard connections exactly what we watch for in earthing joints: loosening of bolted connections due to thermal cycling and ohmic heating, and increased resistance of connection due to oxidation or loss of preload. Loss of material to corrosion compounds both.
In our failure reviews, each load cycle heats the joint, the metals expand at different rates, and the bolt relaxes by microns. Each relaxation increases contact resistance, which raises joint temperature, which accelerates oxidation — the self-reinforcing loop maintenance engineers know as the hot-joint spiral. We point out that underground, you cannot see a degrading clamp, thermography cannot reach it, and re-torque means excavation. This is why engineering practice for buried earthing calls for periodic re-testing — industry guidance on IEEE 80 installations recommends re-testing every 5 years, and every 3 years in aggressive soil. We calculate that over 20 years this amounts to four to six test-and-excavate cycles per suspect joint.
We want to be clear that none of this makes clamps bad products — it makes them products with a maintenance requirement that must be honestly priced into the specification. We sell thousands of clamps every month; the mistake is not buying a clamp, it is burying one and forgetting it.
The Exothermic Weld: A Molecular Joint With Nothing Left to Re-Tighten
We value exothermic welding because it removes the interface entirely. A measured charge of copper oxide and aluminium powder reacts inside a graphite mold crucible — our AF-0401 and AF-0314 molds are quality graphite rated for 50–100 service cycles, and we see 80–100 joints per mold in field use — and the resulting molten copper alloys the conductors into one continuous piece of metal. There is no contact pressure to lose, no film to thicken, no bolt to relax. The joint’s resistance is simply the resistance of solid copper.
From our export desk, we watch the specifying world converge on welds for buried work. IEEE 837 exists specifically to qualify permanent connections used in substation grounding — the connections within the grid, the ground leads joined to it, and the leads joined to equipment and structures. We see utilities write it into contracts: Arkansas Electric Cooperative’s interconnection requirements state plainly that underground connections shall be exothermic type, and the cooperative reserves the right to inspect every connection before backfilling. When a utility can only look at a joint once — before it disappears underground — it chooses the joint that does not change.
Our factory position is simple: for any buried, inaccessible, or safety-critical connection, specify the weld. We like to say the 60 seconds a mold cycle takes at installation is the last attention that joint will ever need.
Head-to-Head: Welded vs Bolted Ground Connection Over a 20-Year Horizon
We walk distributors through this comparison when they specify a project. Note that neither column is winner-takes-all — the right answer depends on where the joint lives.
| Factor Over 20 Years | Exothermic Weld | Mechanical Clamp |
|---|---|---|
| Joint structure | Molecular alloy — no interface | Pressure contact at bolted interface |
| Resistance trend | Stable; equals conductor resistance | Creeps upward with oxidation and preload loss |
| Documented aging mechanisms | Corrosion at same rate as conductor itself | Thermal-cycle loosening, oxidation, preload loss, corrosion |
| Maintenance over 20 years | None — inspect once before backfill | Re-test every 3–5 years; excavation and re-torque if buried |
| Fault and lightning current | Carries like the conductor; no hot spot | Degraded joint becomes an I²R hot spot under fault |
| Installation skill and tools | 60-second mold cycle, no special skill, no power | Wrench and torque discipline; needs re-access |
| Best position | Buried, inaccessible, safety-critical joints | Above grade, inspectable, retrofit and temporary joints |
| First-cost materials per joint (from our export quotations) | Runs roughly 30–50% above a clamped joint, primarily because each weld consumes a single-use powder cartridge plus a share of mold life (50–100 joints) | Lower first cost — hardware only, no consumable charge |
| Scheduled maintenance interventions over 20 years | 0 — inspect once before backfill | 4–6 excavation, re-test and re-torque cycles (3–5-year cadence) |
| Rated service life | Matches the conductor itself — no interface to age | Above 50 years in inspectable positions (our XJ-J1047 series rating) |
We built this selector as a first-pass decision aid — then send us the project detail for a configured answer:
Weld or Clamp? 20-Year Decision Aid
Over 20 years, the welded joint wins on resistance stability wherever maintenance access is poor. Because a clamp’s resistance depends on preserved preload and clean contact films — both of which soil chemistry erodes — while a weld’s resistance is just copper, the maintenance-free option dominates every buried or safety-critical position.
When Mechanical Clamps Are Still the Right Answer
We believe an honest comparison has to say this plainly: clamps remain the correct engineering choice in real situations. Above grade — at transformer neutrals, test links, earth bars and equipment frames — a clamped joint is inspectable by definition, and our XJ-J1047 series in brass, bronze and pure copper carries a rated service life above 50 years with ISO 9001:2015-certified production behind it. We recommend a clamp on retrofit work where conductors cannot be cleaned and moulded around — it gets the circuit safe today. We consider a weld the wrong tool by design anywhere a joint must be deliberately disconnectable for testing — a test link is the classic case.
Our discipline with distributors is position-based: weld what you bury, clamp what you can see. We see mixed BOMs in most export orders — a lightning protection job might weld every below-grade conductor crossing and clamp the air-termination test links, from one shipment.
How We Support Both Choices From One Factory
We manufacture both sides of this comparison, which is why our advice stays honest. Our factory, Xinchang Shibang New Material, has produced exothermic welding molds and powders alongside grounding clamps for more than 17 years from Xinchang, Zhejiang, two hours from Ningbo and Shanghai ports. Our official company profile records the quantified track record: products sold to more than 60 countries, a 3,000 m² factory with more than 40 production workers and 5 R&D personnel, monthly output above 200,000 pieces, and exports exceeding USD 15 million in 2022. Our exothermic line supplies 10,000 pieces per month, production is assessed by CEPRI (China Electric Power Research Institute), and our QC team passes every mold and clamp batch a six-stage quality chain from incoming inspection to final quality check.
In our export work, the projects that go smoothest share one habit: the buyer sends the conductor schedule and joint positions, and our team returns a position-by-position connection plan — welds specified with mold model and cartridge size, clamps specified with material and torque guidance — in one mixed container with one document set. If you are weighing exothermic welding vs clamps for a live project, that plan is what our export desk will send you.
Explore both ranges: our exothermic welding systems, the thermit welding mold family, our copper grounding clamps, and the full exothermic welding category.
Frequently Asked Questions
Does exothermic welding really last 20 years underground?
Yes — the weld alloys the conductors into one metal body, so the joint corrodes at the same rate as the conductor itself and has no interface to oxidise or bolt to loosen. Utilities specify exothermic connections for buried grids precisely because nothing at the joint changes with time; Arkansas Electric Cooperative, for example, mandates exothermic underground connections and inspects them before backfill.
Why does clamp connection resistance increase over time?
We cite three documented mechanisms: thermal cycling relaxes bolt preload, oxidation thickens the contact film at the interface, and soil corrosion removes material. US NRC aging literature (NUREG-1800) lists loosening of bolted connections from thermal cycling and increased connection resistance from oxidation or loss of preload as known aging effects — each cycle raises resistance, which raises joint temperature, which accelerates the cycle.
What does IEEE 837 mean for my connection choice?
IEEE 837 is the standard for qualifying permanent connections used in substation grounding — the joints within the grid and the leads to equipment and structures. A connection type that passes IEEE 837 has demonstrated it will carry fault currents and resist the environment for the installation’s design life; exothermic welds are the classic connection this standard qualifies.
How often should buried ground connections be re-tested?
We follow industry guidance for IEEE 80 installations: re-test every 5 years, tightening to every 3 years in aggressive soil. Over a 20-year design life that means four to six test cycles — practical for accessible clamps, costly excavation for buried clamped joints, and unnecessary for welded joints.
Are mechanical clamps ever better than exothermic welds?
Our answer is yes — above grade, at test links and earth bars, on retrofit work, and anywhere the joint must be disconnectable. In inspectable positions a quality brass or bronze clamp delivers its full rated service life (ours exceed 50 years) because the maintenance it needs is actually possible there.
Is exothermic welding difficult to perform in the field?
We say no. The process needs no external power and no special welding skill: clean the conductors, clamp the mold, load the matched powder cartridge, ignite with the flint gun, allow to cool. Our molds are rated for 50–100 joints each, and we mix the powder to your conductor material and size so the joint forms correctly every cycle.
Can one supplier provide both welds and clamps for a project?
We answer yes, and we recommend it. We manufacture both, so a single order can cover welded buried joints and clamped above-grade positions in one container with one certificate set — and the connection plan tells your crew which technology goes at which position.
Specifying connections for a 20-year earthing system?
Send us your conductor schedule and joint positions — we will return a position-by-position plan: welds with mold and cartridge numbers, clamps with materials and torque guidance, one mixed container.
Jane Yang — Sales Manager, Xinchang Shibang New Material Co., Ltd.
Jane has spent 12 years on the export desk of our earthing and lightning protection factory, helping overseas clients choose and source exothermic welding systems, grounding clamps, ground rods and conductors with full inspection and documentation support.
Post time: Aug-17-2026