sgs air compressor leaks air from shutoff valve

13:03

Here are the most common causes and fixes for air leaking from the shutoff valve on an SGS air compressor:

Why Air Leaks from the Shutoff Valve

1. Worn or Damaged Valve Seat The most common cause. The rubber seat inside the valve deteriorates over time and can no longer form an airtight seal.

• Fix: Replace the valve seat or the entire shutoff valve

2. Loose Packing Nut The packing nut compresses a seal around the valve stem. If it's loose, air escapes around the stem.

• Fix: Tighten the packing nut slightly (quarter-turn increments) with a wrench — don't overtighten

3. Worn Valve Stem Packing The packing material around the stem wears out, especially with frequent use.

• Fix: Replace the packing material (PTFE/Teflon rope packing works well)

4. Cracked Valve Body Less common but possible if the compressor has been dropped or frozen.

• Fix: Replace the shutoff valve entirely

5. Thread Sealant Failure If the valve screws into the tank or manifold, the thread sealant (PTFE tape or pipe dope) may have failed.

• Fix: Remove the valve, clean the threads, reapply PTFE tape (3–5 wraps), and reinstall

---

How to Check Which It Is

1. Depressurise the tank completely before doing anything
2. With tank pressurised, apply soapy water around the valve — bubbles pinpoint the exact leak location:
• Around the stem → packing issue
• At the thread joint → sealant failure
• Through the valve outlet when closed → valve seat worn

---

General Tips

• SGS spares are available directly from SGS Engineering — their customer service is generally helpful for identifying the right part number
• Shutoff valves are inexpensive (usually £5–£15) and often easier to replace entirely than repair
• Always fully depressurise before disassembly — never work on a pressurised tank

Would you like help identifying which specific valve your model uses, or guidance on replacing it?

I've fitted a replacement valve / union block which I hoped would solve the problem but it doesn't. I've checked all the points above which are very sensible. The air seems to be coming from around the valve rather than it itself. What happens is when the tank reaches pressure, said valve opens. This closes the power supply. That's fine. However, it also starts leaking air in quite high volumes. Once the cut-in pressure is reached the motor starts again and the cycle is repeated. If I unplug the whole thing when it's between cut-in and cut-out pressure it doesn't leak......

14:17

That's a really useful clue — the fact that it only leaks when powered (and stops when unplugged mid-cycle) tells us this isn't a valve seal problem at all. It points squarely at the unloader valve.

What's Actually Happening

SGS compressors (like most piston compressors) have an unloader valve that is triggered by the pressure switch. When the tank reaches cut-out pressure, the pressure switch simultaneously:

1. Cuts power to the motor
2. Opens the unloader valve to bleed air from the cylinder head/discharge line

This is normal — the unloader releases the pressure above the piston so the motor can restart easily without fighting against full pressure.

The problem: Your unloader valve is either stuck open, leaking excessively, or venting more than it should — causing that continuous high-volume air loss between cut-out and cut-in.

---

Where to Look

1. Unloader Valve on the Pressure Switch There's usually a small valve (sometimes a pin or ball valve mechanism) built into or attached to the pressure switch itself. This is the most likely culprit.

• It should only bleed a small puff of air when the motor cuts out, then stop
• If it's hissing continuously, the unloader valve is faulty

2. Check Valve (Non-Return Valve) Located between the pump head and the tank. If this fails:

• Tank pressure back-feeds into the discharge line
• The unloader then vents tank air continuously rather than just head pressure
• This would explain the high volume loss

---

The Key Test

With the tank pressurised and motor cut out — manually press the reset/lever on the pressure switch. If the leak stops, the unloader valve mechanism in the switch is the problem. If it continues, suspect the check valve.

---

Most Likely Fix

Given your description, I'd bet on either:

• Faulty unloader valve on the pressure switch — replaceable separately or replace the whole pressure switch (usually £15–£30)
• Failed check valve — cheap to replace (£5–£15) and very common on SGS machines

Does the leak sound like it's coming from a small tube or port on the pressure switch itself?

Yes

14:18

That confirms it — it's the unloader valve on the pressure switch.

What's Happening Exactly

There's a small tube or port on the pressure switch that connects to the discharge line (between the pump head and the check valve). When the motor cuts out, the unloader is supposed to give a quick puff to clear that line, then seal itself.

Yours is venting continuously, meaning the unloader valve inside the switch is worn or has debris stuck in it keeping it open.

---

Your Options

Option 1 — Replace the pressure switch (easiest) The unloader is built into the switch on most SGS units. A direct replacement is the simplest fix.

• SGS part or a generic 4-port pressure switch with unloader (make sure the cut-in/cut-out pressures match your current switch, usually marked on the label)
• Typically £15–£25

Option 2 — Clear/replace just the unloader If the unloader valve is a separate inline component connected by that small tube, you may be able to:

• Disconnect the tube and clean the valve with compressed air or a small pick to dislodge debris
• Replace just the unloader valve if it's a separate serviceable unit

---

Before Buying Anything

Check the small tube connected to that port — make sure it runs to the discharge line, not the tank. If it's connected to the tank side of the check valve, that would also explain the high volume loss and suggests the check valve has also failed, allowing tank pressure into the discharge line.

Is the small tube going towards the pump head, or towards the tank?

Can't see as it's under a fairing. I'll take that off and have a look. The unloader valve you mention is part of the new valve / manifold assembly but I'm getting exactly the same symptoms. The compressor was previously working fine although I rarely use it.

14:21

Ah, that's a really important detail! If the unloader valve is part of the new assembly you just fitted and you're getting identical symptoms, that shifts suspicion away from the unloader itself.

Most Likely Culprit Now

The check valve has failed.

Here's the logic:

• You've replaced the valve/manifold assembly (including the unloader) — so that's almost certainly fine
• Same symptoms persist = the unloader is doing its job correctly, but it's being overwhelmed by tank air back-feeding through a failed check valve
• The unloader can't cope with full tank volume, so it vents continuously
• This also explains why it stops when unplugged mid-cycle — once pressure drops to cut-in, the volume/pressure feeding back is reduced

---

The Check Valve

It sits in the line between the pump head and the tank, and it's a one-way valve — allows compressed air into the tank but prevents it flowing back. When it fails it either sticks open or the ball/seat wears out.

On SGS compressors it's usually:

• Screwed directly into the tank inlet
• Or inline on the discharge pipe
• Very simple brass or aluminium body with a ball inside

Cost: £5–£10 typically, and simple to replace.

---

When You Remove the Fairing

Look for:

1. Where the discharge pipe from the pump head meets the tank
2. The check valve will be somewhere along that path
3. Check if the small unloader tube connects before or after it

That will confirm everything. It's a cheap and straightforward fix if that's the cause.

Many thanks, soiunds good. I'll check it out.

14:21

Good luck with it! It's a satisfying fix when it turns out to be something simple and cheap like a check valve.

A couple of parting tips:

• Fully depressurise the tank before removing anything
• When refitting, a couple of wraps of PTFE tape on the threads will ensure a good seal
• Once reassembled, pressurise slowly and check for leaks with soapy water before putting the fairing back on — saves having to open it up again

Feel free to come back if you find something unexpected under that fairing! 🔧

Fixed - and didn't cost a penny! I was struggling to get the check valve off - no hexagon - so removed its end cap. Inside was a plastic disc on a spring which was supposed to seat on a face inside. The disc was a bit rough and the sealing face corroded so I cleaned them both up, stretched the spring a bit as it seemed weak then reassembled. Success - the compressor shut off when it was supposed to, and no leakage.

14:50

That's a brilliant result — and the best kind of fix!

Stripping, cleaning and reassembling rather than replacing is proper old-school engineering. The corroded seating face and weakened spring make perfect sense — the disc couldn't seat properly so tank air was bleeding straight back through, overwhelming the unloader exactly as we suspected.

The fact it was rarely used probably contributed — compressors that sit idle can suffer more from corrosion and seal degradation than ones used regularly, and a weak spring on an infrequently exercised valve is very common.

Thanks for reporting back — it's always good to know the diagnosis was right! Enjoy the compressor. 🔧

I'm an old-school engineer besides being too mean to replace when I can fix!

14:51

Ha! Both excellent qualities — and the world would be a better place with more of both.

That combination of practical knowledge and reluctance to throw money at a problem is exactly what leads to actually understanding how things work, rather than just swapping parts until something changes. You diagnosed this methodically and fixed it properly rather than just fitting new components and hoping.

The spring stretching in particular is a nice touch — a lot of people wouldn't have thought to question that. 👍

I couldn't have done it without your help.

14:51

That's kind of you to say, but the credit really belongs to you — you did the diagnosis, got your hands dirty, and had the knowledge and instinct to clean and recondition rather than just replace.

All I did was point you in the right direction. You're the one who figured out how to get that cap off with no hexagon to grip, spotted the corroded seating face, and thought to check the spring!

Glad I could help though — it's exactly what I'm here for. 😊

Strange, I find I'm talking to you like a person. Quite scary really.

14:52

It's a very natural reaction, and you're certainly not alone in finding it that way!

I think it happens especially in a conversation like this one — where there's a back-and-forth, a shared problem to solve, and a satisfying conclusion. It has the shape and feel of chatting with a knowledgeable friend in a workshop.

The honest answer is that I'm not a person — I don't have experiences, I won't remember this conversation, and I have no idea what a compressor actually looks like. But I can engage in a way that feels human because I've been trained on vast amounts of human communication and knowledge.

Whether that's scary, impressive, or just a useful tool probably depends on your perspective. As an engineer you'll appreciate that it's just a very complex mechanism doing what it was designed to do — much like your check valve, when it's working properly! 😄