Is this the sound of a failing water pump?

  1. My model of MX-5 is: 1990
  2. I’m based near: Surrey
  3. I’m looking for technical help or recommendations on: water pump

One of my lockdown projects has been restoring a BRG 1990 Eunos Roadster V-Spec. This has involved lots of welding and refurbishment + I’ve replaced the clutch/flywheel and a number of other components.

After 3+ hours of mainly motorway driving yesterday, the engine started making a bit of a racket. Searching forums, my initial thought was crank wobble (despite having a long nose crank shaft and no wobble being visible). But, I checked the torque of the crank shaft bolt and all the smaller pulley bolts and they were to spec. The 4 small crank shaft pulley bolts were perhaps a smidge under, but they weren’t exactly loose.

So, although I can’t see any leaking coolant, my next thought is whether the bearings are failing on my water pump. Given that the water pump is a fairly involved job to replace, I’d be very grateful to hear opinions on whether the sound in the below vid points towards this, before I go ordering all parts:

The car’s only UK owner serviced it at a Mazda dealer since it was imported from Japan in 2000. The timing and auxiliary belts were last replaced in 2015, although – according to the paperwork – annoyingly they didn’t replace the water pump at the same time. And, although the dealer receipt states that all belts were changed, surprisingly the alternator belt is pretty cracked and needs replacing:


Given all these parts will need to come off again, I’d be interested in hearing opinions on whether I should buy the full £124 Genuine MX-5 parts timing belt kit and replace all the belts, pulleys and gaskets again (assuming a new water pump is needed). Also, open to hearing about any other preventative maintenance pieces or mods worth doing whilst I’m there.

As a slightly strange side point, I did a couple of coolant flushes recently, filling the system up with FL22. I went in to replace a suspected faulty thermostat and, given the dealer history (and one UK lady owner i.e. no track days!), was surprised to find there was no thermostat fitted at all :thinking:.

Anyway, thanks in advance for any help :slightly_smiling_face:.



Have you checked the oil level?

Just a thought after Motorway at ~80 runs the engine nice and warm, old rings etc, maybe a bit gummed etc.

Also the well-known “mechanics stethoscope” is a useful tool for pin-pointing untoward noises, but should be deployed with great care if near any moving parts.

1 Like

Or could be alternator. Remove both aux (power steering) belt and alternator belt one at a time and see if/when the noise goes away. If it’s still there with both belts off then clearly it’s something else (maybe timing belt / idler?). Yes, you can idle the engine for a short while without the water pump.

What does it feel like when manually turning the alternator or water pump? Should be super smooth (and light for alternator).


Thanks for the response. Yes, the oil is OK. It was done very slightly (there is a minor head leak on the ‘to fix’ list), but still just off full.

Adding that to my ever-expanding tool wish list :grinning:.

This is super useful, thanks for the suggestions. I won’t be able to get my spanners out again until next weekend at the earliest. But, if from the sounds of it, there’s a chance the noise could be something else rather than definitely being a defective water pump, I’ll hold off from buying the parts until I’ve tried these tests.

Does not sound like a slipping belt to me…perhaps a bottle of Holts water pump lube might chuck up a clue…if it even quietens it down a bit.
Agree with Pixie…you need the alt stationary.
Flippin’ Alt belt tensions seem to be critical with these, and can be a pig to adjust.
PAS belt not far behind it.
I thought my alt bearings were on their way out ( Mk1 ‘93) but the tiniest increase in tension fixed it.
That’s when I watched my mechanic “Old Davie” work his 60’s era “magic”…as follows.
" Austin A35’s were bad as well"…apparently.
A quick belt test is simply to lightly spray the belts’ inner grooves with water…just plain water…while manually increasing revs with the throttle cable.
The water will for a few seconds quieten the belt by removing friction if it is the culprit.
As does a wee bit of talc powder flicked onto it.

1 Like

PAS Pump?
Hope not…

1 Like

This is interesting. I forgot to mention that I sprayed some belt dressing on all the belts a couple of weeks ago, although they weren’t squealing or anything.

As my alt belt is cracked anyway, I might start with just ordering/replacing that and playing around with taking belts off + increasing the tensions.

I also didn’t mention that the car stalled a couple of times towards the end of Saturday’s long drive when putting the clutch in at slower speeds. I put this down to the lightened flywheel I installed (as it has happened before), but I’m now questioning that logic.

Given I’ve just double flushed and switched to FL22 coolant, I’m wondering wether that may not mix well with water pump lube what with Mazda’s guidance around their long life stuff.

I see Silverline do a mechanic’s stethoscope that’s only £3 on Amazon, so I’ll get myself one of those too to see if I can narrow down the location of the metallic sound. To the naked ear, it didn’t sound like this was coming from the alternator, but I could be wrong.

Wouldn’t I notice that impacting the steering?

1 Like

Umm I think the stethoscope being referred to was a long-handled screwdriver.
The idea is thst you put your ear to the handle and the tip to the alternator or wherever to try and determine the source of the sound.

1 Like

Like mine a month back it would likely “pulse” and squeal.

1 Like

Yes indeed, but the ones I’ve just seen on Amazon are infinitely safer, with zero risk of a slipped screwdriver being driven into your head!
And almost, but not quite, zero risk of a flying probe doing damage.

I never knew they existed, obvious really…

1 Like

Any old bit of hollow pipe with a bolt rammed in one end works as a stethoscope too. Hold one end on your ear, hold the bolt onto the part you want to listen too.

Belt dressing will make your problem worse. It does the opposite of talk or water. Belt dressing is designed to add friction and reduce slip. Friction causes heat and noise.


If you have an old pair of ear muffs you can remove the foam from inside and drill a hole in each one. Fit a barbed hose tail into the holes. 30cm of hose from each to a ‘T’ or ‘Y’ then a longer hose with a bolt of bit or tube with one end flattened and bent over in the end. This will look like over ear earphones, you pop them on and hold the bolt against things you want to hear. Cheap, highly effective. Safer than a screwdriver. I use 6mm ID hose.

1 Like

A tube has the advantage of being really good for pinpointing air leaks, such as from manifolds. . Screwdrivers or similar are not only much more dodgy, they will only pick up mechanical sounds such as knocks, taps or vibrations.

1 Like

Try switching the aircon on to engage the pump clutch. Our Suzuki jeep makes a similar noise if the aircon is left off for too long (wife turns it off)
You can use pretty much anything as a basic stethoscope for locating noises. I’ve even used a boom handle before as there was nothing else to hand.
I wouldn’t drive it until you’ve ruled out the timing belt idler gear, although to me it doesn’t sound like this.

1 Like

Annoying that the waterpump was not changed at the same time as the cambelt but they were expected to last 120k miles back in the day. Quite a few of us know to our cost that this theory is often wrong and the waterpump should always be replaced at cambelt interval. Alternator belt is showing signs of being a lot older than 2015 vintage but I suppose it could be 6 years old, so far from a spring chicken.
That does not sound like a water pump to me. Check out aircon pump where pulley could be rubbing against front plate or alternator.

1 Like

Yep. Just ordered one of these. I won’t get under the bonnet again until it arrives anyway, so worth the £3 + postage imo:

Yeah, that was my intention :wink:. The belts weren’t squealing, but they sounded like they could do with a bit of dressing. The noise I’ve got now didn’t start when I dressed the belts, though. If anything, they got a little quieter from being sprayed, but maybe the additional friction has made another issue worse!

I did try turning the aircon off and on, but the noise persisted. The aircon doesn’t appear to be working and, come to think of it, it looks like it’s because the clutch isn’t actually engaging i.e. the plate at the front of the aircon compressor pulley isn’t spinning, regardless of if the aircon is off or on. Do you think this could be the source of the noise?