Nov 13,2026
How Many Stages Does a Multistage Pump Need?
Calculate stage count from duty head and head per stage, then check the operating point shift, motor margin and casing pressure before you order.
How Many Stages Does a Multistage Pump Need? (With Worked Examples)
Quick Answer
Divide the duty head by the head one stage produces at your duty flow, then round up. Check the resulting pump against the system curve, because rounding up moves the operating point to the right — and that shift, not the stage count itself, is what overloads motors and cracks casings.
stages = duty head (m) ÷ head per stage at duty flow (m)
Two numbers decide the answer, and most buyers only have one of them.
Why "head per stage" is not a constant
Head per stage is set by impeller diameter and rotational speed: head rises with the square of tip speed, so a 2-pole motor at 2,900 rpm produces roughly four times the head per stage of a 4-pole motor at 1,450 rpm. Impeller diameter grows with flow, so bigger models also produce more head per stage.
Typical values for stainless in-line multistage ranges:
| Duty flow | 2-pole (2,900 rpm) | 4-pole (1,450 rpm) |
|---|---|---|
| 1–3 m³/h | 6–10 m | 1.5–2.5 m |
| 3–10 m³/h | 8–13 m | 2–3.5 m |
| 10–30 m³/h | 10–16 m | 2.5–4 m |
| 30–80 m³/h | 13–19 m | 3–5 m |
| 80–250 m³/h | 16–24 m | 4–6 m |
Ranges vary by family and impeller trim. Always read the value off the specific curve at your duty flow — not at shutoff, and not at best efficiency point unless they coincide.
That last warning matters more than it sounds. Head per stage at shutoff can be 12–25% higher than at duty flow. Divide by the shutoff figure and you will under-count stages by one or two, then wonder why the pump cannot fill the tank.
Worked example 1: clean water boosting
Duty: 25 m³/h at 96 m, ambient water, 50 Hz, 3 bar suction pressure.
From the table, this flow band gives roughly 13 m per stage at 25 m³/h. So:
96 ÷ 13 = 7.4 → 8 stages
Eight stages produce 104 m at 25 m³/h. That is 8 m more than the system needs at that flow, which means the pump will not sit at 25 m³/h — it will run out along its curve until pump head equals system head.
Assume a system with 60 m static lift and 36 m of friction at 25 m³/h:
| Stages | Head at 25 m³/h | Shutoff head | Actual duty after shift | Shaft power there | Motor |
|---|---|---|---|---|---|
| 6 | 78 m | 87 m | 19.8 m³/h at 83 m | 7.2 kW | 11 kW |
| 7 | 91 m | 102 m | 22.5 m³/h at 88 m | 8.7 kW | 11 kW |
| 8 | 104 m | 116 m | 27.8 m³/h at 105 m | 12.9 kW | 15 kW |
| 9 | 117 m | 131 m | 29.6 m³/h at 110 m | 14.4 kW | 18.5 kW |
Read the table left to right and three things become visible:
Six and seven stages cannot do the job at all. They do not reach 96 m, so flow collapses to 20–22 m³/h. A buyer who rounds down to save money gets a pump that runs continuously, never satisfies the system, and burns out its motor in the first hot spell.
Eight stages overshoots by 11% in flow. That is usually acceptable — but only because the motor was sized at 15 kW for a 12.9 kW duty. Had the specification been written from the 25 m³/h point (11.4 kW), an 11 kW motor would have been chosen and would now be running at 117% of rating.
Nine stages breaks the casing, not the hydraulics. Shutoff head reaches 131 m, or 12.8 bar. Add the 3 bar suction pressure and the case sees 15.8 bar — within 1% of the PN16 limit, with no margin for a pressure surge or a cold-start density increase. This is how a hydraulic decision quietly becomes a mechanical one.
The real cost of one extra stage
Adding a stage is not a linear price increment. It moves four other things:
| What changes | Effect per extra stage |
|---|---|
| Head at duty | +8 to +24 m, depending on model |
| Shutoff / case pressure | +9 to +27 m, plus suction pressure |
| Shaft power | +4 to +6% at the same flow |
| Motor frame | Jumps a standard size every 2–4 stages |
| Efficiency | Falls 0.1–0.3 points as stack length grows |
| Price | +4 to +7% |
The efficiency point is worth stating plainly: more stages do not make a pump more efficient. Leakage through wear rings, disc friction on impeller shrouds and hydraulic losses in the diffusers all accumulate along the stack. A 15-stage build typically runs 1–3 points below a 5-stage build of the same flow. The reason to add stages is head, never efficiency.
Worked example 2: when the casing decides
Duty: 45 m³/h at 340 m, 30 °C water.
Head per stage at this flow is roughly 16 m:
340 ÷ 16 = 21.25 → 22 stages
Twenty-two stages is at the mechanical ceiling of most in-line ranges — shaft critical speed, tie-rod stiffness and stack length all become marginal. But the real problem is pressure: 352 m at duty and 394 m at shutoff, which is 38.6 bar.
| Casing class | Pressure limit | Verdict at 340 m duty |
|---|---|---|
| PN16 | 16 bar (163 m) | Not possible |
| PN25 | 25 bar (255 m) | Not possible |
| PN40 | 40 bar (408 m) | Possible, but only just |
| Barrel design | 40–100 bar | Standard answer |
So the selection question is no longer "how many stages" but "which pressure class". Three routes:
- A PN40 in-line build, if the range offers it. Confirm the flange drillings and the test pressure, not just the casing casting.
- Two pumps in series, each about 170 m. Cheaper per unit, but the second pump's casing sees 17 bar of suction plus its own shutoff head — so it still needs a PN40 body, and now you own two seals, two motors and a control interlock.
- A barrel or can-type multistage pump, designed from the start for this pressure. Usually the correct engineering answer above roughly 300 m.
Buyers who count stages and stop there order a pump that is hydraulically perfect and mechanically impossible.
2-pole or 4-pole?
| 2-pole (2,900 rpm) | 4-pole (1,450 rpm) | |
|---|---|---|
| Head per stage | 4× higher | Baseline |
| Stages for the same head | About 1/4 as many | 4× as many |
| NPSHr | Higher | Lower, often by 1–2 m |
| Noise | 70–78 dB(A) | 62–70 dB(A) |
| Seal and bearing life | Shorter | Longer |
| Purchase price | Lower | Higher |
| Best for | Clean cold water, cost-sensitive, space-tight | Hot water, low NPSH available, night-time noise limits, long duty |
The rule of thumb: if NPSHa leaves you less than 1.5 m of margin over NPSHr on a 2-pole selection, switch to 4-pole before you switch to a bigger pump. The slower machine solves the suction problem, the noise problem and the seal life problem at once — and costs less than the downtime it prevents.
Trimming versus adding a stage
When the duty lands between two builds, there are two ways to fine-tune:
Trim the impellers. Reducing diameter cuts head roughly with the square of the diameter. Trimming every impeller in the stack by 5% removes about 10% of the head, at the cost of 1–3 efficiency points and a poorer match between impeller and diffuser. Beyond about 10% trim, the losses become ugly.
Change the stage count. One stage up or down moves head by a clean increment and leaves the hydraulics matched. It is almost always the better answer when the increment is smaller than the head you would have to trim away.
The exception is a pump already at maximum stage count — then trimming is the only lever left, short of changing model.
When the stage count runs out
Three arrangements extend the range:
- Series. Two pumps in series double the head at the same flow, minus a small loss. Watch the downstream casing pressure and the shaft seal on the second unit.
- Parallel. Two pumps in parallel double the flow at the same head. Better for redundancy and for staged operation with VFDs.
- Higher speed. A 2-pole machine already runs at 2,900 rpm at 50 Hz; a VFD above base speed is possible on some designs but usually voids the warranty and eats seal life.
For most duties between 250 m and 400 m, two pumps in series with a pressure-rated second casing beats one very tall stack on both price and serviceability.
Five errors in stage counting
- Dividing by shutoff head per stage. Under-counts by 1–2 stages.
- Ignoring the operating point shift. On a flat system curve (mostly static head), one extra stage can add 15–20% flow and overload the motor.
- Sizing the motor at the duty point only. IEC motors carry no service factor. Size for the end of the curve, at the coldest water density you will see.
- Forgetting case pressure. Shutoff head plus suction pressure must sit below the casing rating with margin.
- Rounding down to save money. A pump that cannot reach the head runs at high flow, low head, and elevated current until something fails.
The calculation, in code
If you are building a selector for your website, this is the whole logic — the curve lookup is the only part that needs your data:
function selectStages(Q, H, headPerStage, efficiency) {
const stages = Math.ceil(H / headPerStage);
const headAtDuty = stages * headPerStage;
const shutoff = headAtDuty * 1.12; // typical, confirm per model
const shaftPower = (Q * headAtDuty) / (367 * efficiency);
const motor = nextIecFrame(shaftPower * 1.12); // 12% margin, no service factor
return { stages, headAtDuty, shutoff, shaftPower, motor };
}Two lines deserve attention: the 12% margin and the nextIecFrame step. Those two habits prevent most of the overload failures we see in the field.
FAQ
1 Does adding a stage increase flow?
No. Stages are in series, so each one adds head at the same flow. Flow is set by the impeller width and the operating point on the system curve — not by stage count.
2 Can I add a stage to an existing pump later?
Sometimes. If the model family uses the same shaft and the tie rods have spare length, a stack can be extended — but the motor, seal and sometimes the casing class must change too. In practice it is rarely economic against buying the correct build.
3 Why not just buy the highest stage count available?
Because the operating point moves right, the motor steps up, the casing pressure climbs and efficiency falls. An oversized stack is not a safer choice than a correct one — it is a different failure mode.
4 How accurate is the head-per-stage table?
Within about 10–15% between product families, which is enough to shortlist and not enough to order. Confirm against the published curve at your duty flow before you commit.
5 Is a pump with more stages less efficient?
Slightly. Leakage, disc friction and diffuser losses accumulate, so a 15-stage build typically runs 1–3 points below a 5-stage build of the same flow. The difference is small next to the effect of running off the best efficiency point.
6 What is the minimum flow for a multistage pump?
Typically 10–25% of best efficiency point flow, depending on the model. Below that, water in the stack recirculates and heats up. If your demand falls below the minimum, fit a bypass or use a VFD.
Getting the stage count right
The arithmetic takes two minutes. The judgement is in the three checks that follow it: where the operating point lands after rounding up, what pressure the casing sees at shutoff, and whether the motor has margin at the end of the curve. Get those right and the stage count stops mattering — because any reasonable count will work.
Send us the duty point and the system curve, and we will return two or three stage-count options with the operating point, motor size and casing pressure for each, so you can compare them on more than price.
Related News
LATEST
INFORMATION
Get the latest product information of the company
NAVIGATION
PRODUCTS
CONTACT US
Telephone: +86 13305761511
Email: info@cntecho.com
Add: RM. 602, BUILDING B, NO.1551 SHUANGSHUI RD., LUQIAO DIST., TAIZHOU CITY, ZHEJIANG, CHINA
Copyright © 2026 TECHO ELECTRICAL & MECHANICAL (TAIZHOU) CO., LTD. All Rights Reserved.