When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I'd like to flush out the heater matrix on my 2007 Hawk STi.
I can disconnect the rubber hoses at the firewall which will expose the two aluminium matrix ports. The ports sit next to each other left and right, rather than one on top of the other.
I'm likely to add Holts Speedflush neat to the matrix via a PVC hose / funnel. My idea is to add the Speedflush neat and leave to soak for maybe 30 minutes, But which port from looking at the bulkhead in front of me is the inlet to the matrix and which is the outlet? I ask so I can backflush with water etc.
I tried soaking the matrix with distilled malt vinegar for an hour last week, a few bits came out but not much from a water flush, I think the matrix is scaled up. When refilling with coolant from the left port, I started seeing coolant come out of the right port - but I had only added around 350ml of coolant which would suggest the matrix is heavily restricted. Not forgetting the volume in the aluminium pipes would have been using up some of that 350ml.
This shows my matrix set up 72130FE000 but I'm not sure if after the flush it would best be filled from the bottom up (so add coolant to the left hand port)....or....fill with coolant from top to bottom using the right hand port, I want to minimise air locks.
I've added above what I think is inlet / outlet, but does anyone know if it's bottom fill or top fill ideally? Is it better to bottom fill it with coolant from the left port so that any air will bleed out the right hand port as it fills?
Last edited by Hawkeye D; Mar 27, 2026 at 06:54 AM.
Really just because I found the overflow pipe nipple on the radiator hard line was totally blocked with junk, I know the passages in the matrix are narrow.
It's not been great for a while, and now made worse by the RCM Cyl4 mod recently fitted.
I think maybe my water pump has done it's time, it's the original factory pump, 2007 and 120k miles on it...
I've had a couple of cambelt / tensioner services over the years, but looking at the invoices, both invoices only mention water pump gaskets, not the actual water pump.
Is 120k miles about right for a water pump to become lazy?
if your issues started after fitting then there is your "head gasket" issue..............
it ***** up the natural flow of water around the system - throw it in the BIN
if your issues started after fitting then there is your "head gasket" issue..............
it ***** up the natural flow of water around the system - throw it in the BIN
SNAKE OIL
No the heater matrix hasn't been getting hot on and off for a while, before the RCM mod, I hear a ticking / rattle which I wonder is a cambelt tensioner or the water pump bearings are shot...
It's not been great for a while, and now made worse by the RCM Cyl4 mod recently fitted.
I think maybe my water pump has done it's time, it's the original factory pump, 2007 and 120k miles on it...
I've had a couple of cambelt / tensioner services over the years, but looking at the invoices, both invoices only mention water pump gaskets, not the actual water pump.
Is 120k miles about right for a water pump to become lazy?
The knock sensor is fitted above number 4 for a reason. If you hadn't fitted if you hadn't fitted that silly cooling mod your gaskets would chances are be ok.
Can an original 120k mile water pump have shot bearings and suck air into the coolant system whilst running?
My water pump has never been changed, factory original.
Can an original 120k mile water pump have shot bearings and suck air into the coolant system whilst running?
My water pump has never been changed, factory original.
if it could suck air in then itd also be leaking coolant out
Now tonight an EML / CEL has appeared, little to no boost driving, but my PSI3 data monitor is giving me no codes, current or historic.
I've been hearing a ticking noise which goes after about 3.5k rpm, I wonder if it's an idler or tensioner that is noisy, and perhaps not driving the water pump properly.
I'm not sure if the ticking continues after 3.5k rpm, as it could be being covered up by road / wind noise.
Last edited by Hawkeye D; Apr 9, 2026 at 06:07 PM.
Neither would cause a CEL. Nor would i expect an idler not to drive the water pump properly....given its all part of the timing belt youd have far bigger issues if youd a knackered idler, also it wont tick as its a bearing
If the CEL is still on then psi3 should be able to read it, or else get a wifi/bluetooth dongle and read it via TorqueApp or similar
I use to get random cels which was caused by a cracked casing on the knock sensor that allowed moisture in, then after a while engine heat would dry it out and cel disappeared but could read it using psi3 as/when it was on
Ok, so borrowed the Snapon diag machine from work tonight.
I've got P0140 - Oxygen Sensor Circuit Range / Performance, Bank 1 Sensor 2.
Interestingly, I have the PSI3 user manual and it doesn't even list that code. Anyway, P0140 is the code stored.
I assume it's the rear sensor near the cat, underneath the car about half way, I can borrow a tool from work which I used for my front O2 sensor last year. I guess the rear sensor plugs in somewhere near the gearbox, straightforward or a blimmin nightmare?