Home / Valve Test Benches / Clamping methods
How a valve is clamped on a test bench — and why the fixture decides whether the result is valid
The pressure gauge cannot tell a seat leak from a leak at the sealing plate. Both reduce the pressure in the same closed volume. So the fixture — the blind plates, the grip jaws, the end cylinders — is not an accessory to the test, it is part of the measurement. This page compares the three ways a valve is held on a hydrostatic test bench, maps each one to the valve bodies it suits, and lists the fixture details worth confirming before you place an order.
Quick answer
| Flanged gate, globe, ball valves | Blind plate + top clamp |
|---|---|
| Wafer or lug butterfly valves | Radial grip jaws or wafer adapters |
| DN350 – DN600, long face-to-face bodies | Horizontal end clamp |
| Butt-weld end valves | Weld-neck adapters |
| Mixed DN50 – DN600 in one shift | Dual station, one range per station |
Three jobs the fixture has to do at once
Every pressure test closes the valve body into a sealed volume, fills it, raises it to the test pressure and watches the pressure for a fixed hold time. The pressure acts on the inside of the body in every direction, including on the two faces where the fixture closes it. Those two faces carry the whole axial load of the test, and they are also where the fixture seals. A working fixture satisfies three conditions:
- Close the volume on the actual sealing face. A flange raised face, a wafer body face and a weld prep are three different surfaces. The fixture has to press an elastomer against the one the valve actually has, otherwise the joint weeps before the test starts.
- Resist the end load without distorting the body. The axial force grows with the square of the diameter, so the same tonnage holds a DN100 valve at a high pressure and a DN600 valve at a much lower one. This is the reason large-bore benches are quoted with high clamping tonnage and a lower maximum pressure — clamping force and pressure ceiling are one specification, not two.
- Let the operator see the seat. A shell test pressurises the whole body; a seat test pressurises one side of the closed disc. The fixture and the water routing have to allow both without dismantling the setup.
If any of the three is missing, the bench still produces numbers — they are just not numbers about the valve. The usual symptom is a good valve rejected twice and then passed on a different plate.
Method A — blind plates with vertical top clamping
The valve sits on a lower plate, a top cylinder closes it downward
Used on: SYJ-150, YSYJ-300, SYJ-600SQ, and the two stations of the SYJ2-300-600
The most common arrangement for flanged valves and fittings. Water enters through the centre of the lower plate, the top plate closes the upper flange, and the cylinder holds both plates against the valve. Plates are supplied in size bands and carry an elastomer seal on the face that bears on the flange.
- Seals where the customer's gasket seals. The test presses on the same raised face that will later carry the line gasket, which is the closest a bench gets to installed conditions.
- Covers fittings as well as valves. Elbows, tees and reducers clamp between the same plates with the appropriate adapters, so a valve workshop can take fitting work without a second machine.
- Handles both media without re-fixturing. On the SYJ-600SQ the valve is clamped once between the sealing plates and stays clamped through the water stage and the air stage — up to 40 MPa water and 4 MPa air on the same setup, with interchangeable sealing plates covering DN50 – DN600.
- Reaches high pressure in small bores. A compact vertical frame with a 30-ton clamp such as the SYJ-150 covers DN15 – DN150 at 0 – 40 MPa, where the end load is small and plate changes are quick.
Where it stops: a wafer or lug butterfly valve has no flange for the plate to bear on, and a butt-weld end valve has no flange at all. Both need a different fixture — a wafer adapter pair, or the weld-neck adapters that the YSYJ-300 is quoted with for welded-end valves.
Method B — horizontal end clamping
The valve lies on its axis and the cylinder closes it from the end
Used on: WBD-600
Same sealing principle as Method A, rotated 90 degrees and moved down to waist height. On the WBD-600 a 150-ton cylinder closes the valve between two end plates, and water is fed from both ends so the seat can be tested from either direction without re-clamping.
- Large bodies load without a hoist. A DN600 valve weighs well over 200 kg; horizontal loading rolls it into position on a trolley instead of lifting it over a lower plate.
- Long face-to-face bodies clear naturally. Check valves, globe valves and long-pattern ball valves clamp end to end; a vertical bench would need a very tall frame to clear the same body.
- Seat test in both directions. Feeding both ends means a downstream seat test does not require turning the valve around and re-clamping it.
Where it stops: the working pressure ceiling follows the clamping tonnage, so a horizontal large-bore bench is quoted for the pressure band its cylinder can hold at DN600 — 0 – 16 MPa on the WBD-600 — not for the 40 MPa of a small-bore frame.
Method C — radial grip jaws with top assistance
The jaws close around the body; nothing is bolted to a flange
Used on: DFS2-600 (both stations)
The jaws move inward radially around the outside of the valve body while a top cylinder presses down to assist, so the valve is secured without bolting up a flange set. On the DFS2-600 water enters through the centre and vents through the top blind plate. The left station covers DN50 – DN300 with 80-ton grip clamping and a 60-ton top cylinder; the right station covers DN350 – DN600 with 126-ton grip clamping and an 80-ton top cylinder, both on one frame sharing one hydraulic power unit.
- No flange, no plate — the fixture adapts to the body. Within a station's range the jaws travel to the diameter, which is why a mixed batch of butterfly, gate and fitting work does not stop for tooling.
- Two ranges, one machine. A DN100 valve and a DN500 valve are tested on the same day without a second bench and without re-tooling — one is a change of station.
- Documented curve per valve. The cycle is automatic: low-pressure fill and vent, changeover to the high-pressure circuit, build to the test value, hold timer, then a pass or alarm against the acceptance factor. Each run is stored with its time-pressure curve and exported over USB.
Where it stops: 0 – 7 MPa is the ceiling on this frame, and the reason is geometric — at DN600 the clamped area is large, so the pressure ceiling is matched to the cylinder forces. Large-diameter valves are normally PN2.5 – PN10, and a 1.5 × PN strength test stays inside that ceiling. If your work is DN50 at PN100, this is the wrong machine and a small-bore vertical bench is the right one.
Side-by-side comparison
| Criterion | Blind plate + top clamp | Radial grip jaws | Horizontal end clamp |
|---|---|---|---|
| Seals against | Flange raised face | Body outside diameter, jaws + top assist | End face of the body |
| Valve bodies | Flanged gate, globe, ball, plus fittings | Butterfly, gate, welded-end bodies and fittings | Butterfly, gate, ball, check, globe |
| Changeover inside a range | Plate pair per DN band | None — jaws travel to the body | Plate or adapter per DN band |
| Loading a DN600 body | Hoist over the lower plate | Hoist onto the station | Trolley at waist height |
| Long face-to-face bodies | Needs frame height | Tight for long bodies | Native fit |
| Seat test both directions | Depends on water routing | Depends on water routing | Both ends fed |
| Typical machine | SYJ-150, YSYJ-300, SYJ-600SQ | DFS2-600 | WBD-600 |
Which valve body needs which fixture
| Valve body | Fixture | What to watch |
|---|---|---|
| Flanged gate / globe (ASME B16.5, EN 1092, GOST) | Blind plate + top clamp | Plate face condition; confirm the flange standard and rating so the plate matches |
| Wafer butterfly | Radial grip jaws or wafer adapter pair | No flange face — the plate must bear on the body face, not on air |
| Lug butterfly | Wafer adapter with through-bolts, or grip jaws | Lug ears take the load; check thread condition before clamping |
| Ball valve, long pattern | Horizontal end clamp | Face-to-face length; whether the seat test must run both directions |
| Check valve | Horizontal end clamp | Long body; the disc has to be held closed for the seat test |
| Butt-weld end valve | Weld-neck adapters | Pipe schedule and weld prep geometry |
| Fittings (elbow, tee, reducer) | Blind plate + adapters | Center-to-face dimensions, not just DN |
What ships with the machine, and what it costs you later
Fixture scope is where quotations quietly differ. Two benches with the same tonnage and the same pressure range can arrive with very different tooling, and the missing half is bought later at spare-part prices and at the cost of the days the machine stands idle waiting for it. The points worth writing into the order:
- Plate coverage of the whole range, not the first size. The DFS2-600 ships with upper and lower blind plates for DN50-150, DN200-300, DN350-450 and DN500-600 — one pair per range — plus four extension rods and two rod flanges, so the full DN50 – DN600 range is testable on day one.
- Spare elastomer. Sealing elements are consumables. A spare set on the shelf is cheaper than a rejected batch traced to a hardened seal.
- Two stations: parallel or sequential? The two stations of the SYJ2-300-600 share one hydraulic power unit and are used one at a time — the gain is range and floor space. The ZZC2-200 is the opposite case: two stations alternate, one is reloaded while the other is under test, and the PLC runs the strength, lower-seal and upper-seal sequence to a printed record. If you need throughput, that is the architecture; if you need range, it is the other one.
- Adapters for the body types you actually sell. Wafer adapters and weld-neck adapters are machined to your standard and schedule; ordering them with the machine avoids a second shipment.
What to send us for a fixture-matched quotation
- Valve types and body styles (flanged / wafer / lug / butt-weld), with the flange standard and rating
- DN range you actually sell, and the sizes that make up most of the volume
- Face-to-face length for long-pattern bodies
- Test pressure per size, or the PN rating the test is derived from — the shell test is 1.5 × PN and the seat test 1.1 × PN, see the hydrostatic test pressure chart
- Test medium: water, air, or both on the same valve
- Valves per shift and the hold time you must record
- The leakage acceptance criterion your customer's inspector applies — see API 598 and ISO 5208 leakage rate classes
With those seven points we can name the clamping method, the tonnage and the pressure ceiling for your size band, and put them in writing with the quotation.
Frequently asked questions
Can I test a wafer butterfly valve on a bench with blind plates?
A wafer or lug butterfly valve has no flange face for the plate to seal against, so a plain blind plate will not close the body volume. It needs a wafer adapter pair that seals on the body faces, or a bench with radial grip jaws that close on the outside diameter, such as the DFS2-600. Tell us the body type when you ask for a quotation and the fixture is matched to it.
Can a worn sealing plate cause a false leak reading?
Yes, and it is the most common reason a good valve is rejected. The pressure drop is measured in one closed volume, so a leak at the plate seal and a leak through the seat look identical on the gauge. Inspect the plate faces in the daily check and keep spare sealing elements — re-test a failed valve on a fresh plate before scrapping it.
Do I need a separate sealing plate for every valve size?
No. Plates are supplied in size bands. The DFS2-600 ships with upper and lower blind plates for DN50-150, DN200-300, DN350-450 and DN500-600, so the full range is testable on delivery. A bench with radial grip jaws needs no plate change at all inside a station's range, because the jaws travel to the body diameter.
How do I test a butt-weld end valve?
With weld-neck adapters instead of flange plates. The adapter is machined to the pipe schedule and seals on the weld prep, so the valve is tested in the condition it will be installed. It is a standard option on the YSYJ-300 and can be quoted for the other vertical benches.
Which clamping method is fastest for mixed batches?
Radial grip jaws, because nothing is exchanged between sizes. On the DFS2-600 the two stations are permanently fitted with their own jaws and cylinders, so moving from a DN100 valve to a DN500 valve is a change of station, not a change of tooling.
How much clamping force does my valve need?
Enough to hold the end load with a margin, and the end load is the test pressure multiplied by the sealed area — it grows with the square of the diameter. That is why a large-bore machine is specified with high tonnage and a lower maximum pressure than a small-bore machine. Send the DN, the valve type and the test pressure and we will confirm the tonnage and the pressure ceiling for your size in writing.
Benches and the fixture each one uses
DFS2-600 · Grip jaws
- DN50 – DN300 (80 t grip) / DN350 – DN600 (126 t grip)
- 0 – 7 MPa water
- Two stations, one power unit
SYJ-600SQ · Blind plates
- DN50 – DN600
- Water 40 MPa / Air 4 MPa
- 90 t clamp, one clamping for both media
YSYJ-300 · Blind plates / weld-neck
- DN50 – DN400
- Water 40 MPa / Air 0.8 MPa
- 80 t top clamp
For the selection logic from pressure to tonnage to station layout, read the valve pressure test bench buyer guide. For how the machine is loaded and where it stands, read horizontal vs vertical valve test bench.
Flange standard, DN range, face-to-face length and test pressure are enough for a fixture-matched proposal: clamping method, model, datasheet and price within 12 working hours.