This is a continuation of my excessive coolant expansion question regarding my 98 Mazda 626 2L. When I drained the coolant I noticed this black sediment which settled to the bottom of the container I was draining into:

enter image description here

It seems to be fine grains of iron. I also got some almost BB sized chunks when I backflushed the heater core.

Basically, I drained the coolant and back flushed both the radiator and heater core, then let everything completely drain. I hooked everything back up, except the hose to the IAC since it's the highest point in the system, and filled as much as I could from the filler neck. I then went through the WSM procedure of idling till the fan comes on, running at 3k rpm for a minute, shutting off, filling at the neck, repeat.

However, I'm still getting excessive expansion into the overflow bottle, in fact overflowing it, and when I feel the two hoses to the heater core they are both ambient temperature. The thermostat works fine and the bottom radiator hose is hot ( not as hot as the top one of course ), so I know coolant is flowing through the radiator ( it wasn't before ).

So I'm wondering what is going on? Could I have air trapped in the heater core that's expanding to force fluid out and not letting fluid in? Do I need to disconnect one of the heater core hoses and fill it manually? I know it's not clogged because water flowed through it with no problem when I back flushed it. The hoses to the heater core both curve down before going back up to where the connect to the engine, so maybe that's not letting the air get out / the fluid get in?


So I parked on a hill, front of the car uphill, and was finally able to get fluid flowing into the heater core. Now the heater core intake hose gets as hot as the others, but the out hose is only getting warmish and I'm only getting a little hot air from the heater, so I guess the coolant flow through the core isn't too hot ( pun intended ). I'm still getting excessive coolant expansion, although not as bad as before.

So I figure there must still be air and / or a clog in the system, I'm going to monitor it for a few days and then update this question.

EDIT Oct. 4th 2016

I was doing some reading in Tim Gilles' "Automotive Service" and ran across this ( pg. 293 ):

Aluminum Oxide Contamination

Aluminum Oxide is very abrasive; it is what sandpaper is made of. It does not form unless the coolant has become too diluted with water. If there is aluminum oxide in the cooling system, it looks like black sand beads.

So I guess I now know what that black sediment stuff was.

EDIT Oct. 5th 2016

Well, it's still overflowing the expansion bottle so much that I can't drive the car. I back flushed it again today with the hose, and got a bit more grit out of the heater core. The thing is, that it seems like I've got pretty decent flow through the radiator; if I jam the water hose in the lower radiator hose, a little shoots back out, but most of it comes out the upper radiator hose. Of course I guess the radiator could be partially clogged, but maybe I've got rust / aluminum oxide that's clogging something in one of the blocks' water passages and that's causing a hot spot which is boiling and forcing out the coolant?

I took a picture of part of the water pumps' impeller when I swapped out the thermostat a few months ago ( using an endoscope ):

enter image description here

I don't think it could have fallen apart in the few thousand kilometers I've driven since then.

Don't know what to think really. I'll probably do a Mercedes Citrus Flush tomorrow and hope that'll remove any scale that's built up in the system and hopefully solve my problem. Otherwise, I'm at a loss.

POSTSCRIPT Oct. 6th 2016

So @vini_i was right:

enter image description here

There be not a single fin left on the impeller! I bet allot of those little pieces of metal I was flushing out of the system where the remains of the impeller blades! I even cranked the engine a few times to get a visual of the impeller spinning, and all that's left is one tiny little stub of one fin.

Reviewing some video I took after I vinegar flushed the system a few thousand kilometers ago it actually looks like the flush took out some of the fins and I've been running with only one or two fins, which have finally given up the ghost.


POSTSCRIPT Nov. 10th 2016

So I pulled off the old water pump and look what I found:

enter image description here

The fin in my hand was what I saw last year with my endoscope. It was sitting in the bottom of the water pump and just fell out when I opened it up. I have a strong feeling that this is the original OEM water pump that came with the car.

But guess what else I found:

enter image description here

Another fin jammed so hard in the water passage that I had to extract it with needle nose pliers. Now I'm wondering where the heck the rest of those fins are!

  • Wish I could +2 this. Commented Nov 10, 2016 at 19:07
  • @DucatiKiller Thanks! I have to say there is not enough upvoting on this site. Commented Nov 10, 2016 at 19:34
  • @RobertS.Barnes The hope is that the rest of the impeller rusted away to nothingness.
    – vini_i
    Commented Jan 12, 2017 at 9:55
  • You should post your borescope diagnostic as an answer here
    – Zaid
    Commented Jan 13, 2017 at 3:56

1 Answer 1


It's possible that the impeller blades on the water pump are either compromised or gone completely.

I ran into a somewhat similar situation with a Chrysler. The engine would not overheat driving. Idling it would overheat eventually but it took a long tome to do so. The heater core would not get hot no matter what I did. The thermostat was already replaced.

Eventually I disassembled the water pump and found that the impeller blades looked like an apple core, all chewed up. With a new water pump the heat came back instantly.

What might be happening is some parts of the engine are not seeing enough coolant flow and are boiling. This is while the rest of the engine sees just enough flow to keep from overheating.

  • I actually inspected the impeller blades a few months ago with an endoscope when I replaced the thermostat and the blades that I could see were all in one piece. Commented Oct 5, 2016 at 17:39
  • @RobertS.Barnes were the blades metal or plastic? Plastic blades are known to spontaneously crack, especially oh high mileage vehicles.
    – vini_i
    Commented Oct 5, 2016 at 18:28
  • they're metal and were covered with rust, like everything in the cooling system. Commented Oct 5, 2016 at 18:54
  • You can see a pic of the impeller in the post. Commented Oct 5, 2016 at 19:35
  • @RobertS.Barnes To me the impeller looks compromised. The blade that is directly in the middle of the picture looks like it has a hole in it. (green blotch with black in the middle)
    – vini_i
    Commented Oct 5, 2016 at 19:40

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .