Radiator and Kubota engine repair services need in Las Vegas, NV

[bevans6]

FWIW, which is not much, when I had a similar problem (albeit with a race engine in a 1961 Lotus 7) it was a crack in a cylinder wall.  It defied diagnosis till it was put on an engine dyno with a see-through cooling system - 100 gallon water tank.  It only had a problem when it was hot, it passed all the tests you could imagine cold.

Brian

[Sean]

Jim and Brian,

I am wondering how, if this is the problem, the engine is running fine for hours and hours once it is bled. In fact, once it has been running fine, it will re-start just fine too, as long as it is still somewhat warm.  It's only after it cools down overnight that we get the air bubble.

I would think that if combustion gas was blowing past the gasket into the coolant, that we'd get a gas bubble while running, followed by overheating and a shutdown.  No?

-Sean
http://OurOdyssey.BlogSpot.com

[bevans6]

When is it putting the fluid out onto the ground?  I thought that was when it was hot, but you say it runs fine.  So - it's running fine, then it's off, it gets cold, it gets air-locked (where is still a mystery, yes?), it starts up again, it over-heats, it dumps fluid out.  You have some tiny weeps in the rad, but not enough to bother it while it's running.  If full pressure won't spray coolant out all over and make a mess, the vacuum created by cooling down probably won't pull in a ton of air.  So I look for another source of air - combustion or cavitation.  Most cooling systems are self-bleeding, or should be.  The design is such that air naturally rises to the highest point, which is often where the radiator cap is and where coolant is added, but not always, sometimes the header tank on top of the radiator, sometimes a separate header tank.  That air is supposed to create a "pressure spring" to help keep the coolant from cavitating, hence the pressure cap.  Perhaps an internal bleed passage is blocked.  The water pump should definitely be able to self bleed itself, and the cooling system would usually have a gravity head of water pressure above the input to the water pump from water in the radiator or in the block.  It's a remote rad, I think you said, how is the top of the radiator supposed to self-bleed back to the header tank?  You have coolant recovery tank - is that a pressurized tank or atmospheric?  Is the rad cap on the header tank (the one that feeds the recovery tank) equipped with a top sealing ring?  If the tank is pressurized, is it's pressure cap at the same or lower PSI than the main one?  FWIW when I run a pressurized coolant recovery tank, I run a solid seal cap on any other fill points.

The story about the cracked cylinder is just to point out that cracks sometimes are hard to diagnose from outside the engine, they can defy logic because sometimes they close up tight when they are cold, and sometimes they close up tight when they are hot...  But I would also expect the thing to fail while running if it was head gasket or a crack.  Which is why I am now thinking a blockage somewhere.  Anyway, random ramblings, this problem has bugged me all day.  Sorry I don't know a shop in LV, I've never actually been to LV...   

Brian

[Sean]

Brian,

Thank you very much for the detailed response.  I appreciate the help.  I will try to answer your follow-up questions as best I can.  The genny is no longer in front of me; we are still at Lake Mead and Ben and Karen have returned to Las Vegas with their bus.  Ben tells me he has an appointment at a radiator shop this afternoon.

When is it putting the fluid out onto the ground?  I thought that was when it was hot, but you say it runs fine. 

Let me clarify the sequence.

If I don't first bleed the system, from a cold start, I will see the head rapidly heat up, with virtually no rise in temperature at the thermostat housing or bypass line back to the water pump.  It very quickly reaches the setpoint of the temperature safety and the engine shuts down.  Almost simultaneous with the shutdown (had to say if just before or just after), the radiator cap will let so much coolant into the recover tank that it overflows.  I'm guessing about two cups or so.

If I bleed it first, which seems to require opening at least one bleeder with the engine running until just fluid comes out, then I see good coolant flow, as evidenced by steadily rising temp, through the thermostat housing, and then on to the radiator once the 'stat opens.  Once I have good coolant flow to the radiator the unit runs normally indefinitely and no fluid comes out anywhere.  We noticed two perpetually wet spots on the rad, which is why I am guessing is has some pinhole leaks.

So - it's running fine, then it's off, it gets cold, it gets air-locked (where is still a mystery, yes?),

Yes.

... it starts up again, it over-heats, it dumps fluid out.

Yes, as described above.

  You have some tiny weeps in the rad, but not enough to bother it while it's running.

Does not seem like it.

  If full pressure won't spray coolant out all over and make a mess, the vacuum created by cooling down probably won't pull in a ton of air.

Hmm, OK, that makes sense.  But I will tell you that before I changed all my hoses and clamps on my main cooling system (on my Detroit, not the genny), the system ran fine with no leaks at all when it was warm, but after really cold nights, we'd find coolant on the ground under the coach, sometimes a lot of it.  It turned out that differential contraction of the materials, wherein the hose barbs were contracting faster than the hoses and clamps in the cold, was enough to let gravity alone force coolant out through the resulting gaps.  So it seems reasonable to me that something similar could be happening here, where everything is tight once the genny is at running temperature, but as the temperature drops into the low 40s or high 30s here at night, that same sort of differential contraction is letting air into the system.  That is, of course, just a working theory and not confirmed by any direct observation.

... So I look for another source of air - combustion or cavitation.  Most cooling systems are self-bleeding, or should be.

Agreed, and my generator setup, based on exactly the same Kubota engine, does work this way and I don't even have any bleeders on the system.

...  The design is such that air naturally rises to the highest point, which is often where the radiator cap is and where coolant is added, but not always, sometimes the header tank on top of the radiator, sometimes a separate header tank.

This system actually has two tanks, each with a pressure cap.  There is the header tank on the remote radiator, which is connected to an overflow tank at atmospheric pressure.  Then there is a separate tank near the engine, with another pressure cap but no overflow.  The tank near the engine appears to be plumbed with a tee into the coolant return just before it reenters the block.  It is higher than everything on the engine, but lower than the radiator tank and some of the hoses.

The tank near the engine is equipped with an 11 PSI cap and we have not seen that one release at all.  The tank on the rad was equipped with a 14 PSI cap.  However, the first time we saw this cap release pressure, not only did some go into the overflow tank, but some came spilling out around the cap, too, so I replaced it.  I could not source a 14 locally so I settled for a 13.  The 13 has a top gasket, which I did not notice on the 14

  That air is supposed to create a "pressure spring" to help keep the coolant from cavitating, hence the pressure cap.  Perhaps an internal bleed passage is blocked.  The water pump should definitely be able to self bleed itself, and the cooling system would usually have a gravity head of water pressure above the input to the water pump from water in the radiator or in the block.

Perhaps this is the purpose of the extra tank on the return side.

It's a remote rad, I think you said, how is the top of the radiator supposed to self-bleed back to the header tank?

Not sure I understand this question.  The radiator, while remote, is a conventional style with a tank soldered to the top and a pressure cap.  This would appear to be the high point of the system although there may be a very short bit of hose run, intended to let the genny slide in and out of the bus, that loops to this height or just above it -- hard to tell.

You have coolant recovery tank - is that a pressurized tank or atmospheric?

Atmospheric.

Is the rad cap on the header tank (the one that feeds the recovery tank) equipped with a top sealing ring?

Good question.  The 13 PSI cap I just installed has one; the 14 PSI cap it replaced apparently did not.  That cap was put on by some previous technician, so it's unclear what the design calls for.

If the tank is pressurized, is it's pressure cap at the same or lower PSI than the main one?  FWIW when I run a pressurized coolant recovery tank, I run a solid seal cap on any other fill points.

Hmm.  So with an unpressurized recovery tank, should I go back to the gasketless 14 PSI cap?

The story about the cracked cylinder is just to point out that cracks sometimes are hard to diagnose from outside the engine, they can defy logic because sometimes they close up tight when they are cold, and sometimes they close up tight when they are hot...  But I would also expect the thing to fail while running if it was head gasket or a crack.

Yeah, that's my thinking, too.

  Which is why I am now thinking a blockage somewhere.

OK.  That would not surprise me, since the coolant looks crappy and when I pulled the thermostat housing off I could see light debris in the coolant (although hard to say it was not just old gasket sealer from the stat housing).  My first recommendation to Ben was a backflush and refill with proper coolant, in addition to replacing the hoses and testing the rad.

I welcome any further thoughts you may have based on the answers I just provided.

-Sean
http://OurOdyssey.BlogSpot.com

[Boomer]

Sean have you considered a collapsed inner liner on a radiator hose?

[artvonne]

Sean have you considered a collapsed inner liner on a radiator hose?

  Hahahaha, thats one you dont hear very often these days. The lower hose could also be missing the spring.

  For those who dont know, the water pump wants to draw coolant from the bottom of the radiator into the pump, but cannot do so unless the thermostat is open, which then allows flow from radiator to pump, pump to engine, engine to top of radiator. When an engine is cold there is no coolant system pressure, and suction from the pump can collapse the hose. A good pump can keep that hose collapsed no matter how hot the engine gets. Once it blows and you shut it down, if you restart it it often times has enough residual pressure to keep the hose inflated and, well, it can cause a lot of head scratching.

  I once had a lower hose that got an embolism. The inner section of hose got a leak that was retained by the outer layer. Pressure forced coolant within the two, and the inner liner separated from the outer part of the hose and pulled away, until it completely choked off the lower hose. Externally the hose looked fine, it was only when I squeezed it, you could feel this odd lump inside. That one had me scratching my head for a while.

  Brian may have hit on the right idea, you dont want two pressure caps. You should only have a bleeder at the remote radiator, with no pressure cap. The Pressure cap should be highest point near the engine, and overflow bottle should be adequate volume plumbed off the pressure cap. Alternatively, many later model vehicles put the pressure cap directly on the expansion tank, making a 100% closed system. Many often run a bleeder off the T stat housing to the tank as well, making a self bleed cooling system.

[Boomer]

In the progression of easiest/cheapest to difficult/expensive with this problem,  I think installing new hoses first would be a good idea.  Cut open the old hoses lengthwise to see if by chance they have delaminated.  Sounds like the hoses are due for replacement anyway.

[Sean]

So just a quick update here:

I am still at Lake Mead, although we will be rolling down the hill to B&B Coach to fix a problem of our own later today (http://ourodyssey.blogspot.com/2012/02/not-what-we-meant-by-pop-rivets.html), and Ben brought his bus to a radiator shop in town yesterday afternoon.  I spoke to the tech on the phone to brief him on what we had already done and the symptoms.

They removed, inspected, and pressure tested the radiator, which turned out to be in good shape.  They also removed and inspected all hoses, which also turned out to be fine.  What had looked to us like age cracking in the hoses was actually in some clear coat that Liberty had applied to the whole system, hoses and all.  We also learned that, while it is marked as an 11, the cap on the lower accumulator tank in the return line has been blocked off and so is actually solid, so the 14 cap on the radiator header at the high point in the system is the only actual pressure relief cap, connected to a proper (atmospheric) expansion tank.

The shop drilled out the bleed hole in the thermostat to a larger size, and pressure tested the entire system, which passed.  They backflushed the cooling system and took a sample of the old coolant for analysis.  As of the last I heard last night, it was all back together and working.

Of course, only an overnight cool down will tell us if a backflush, fill, proper bleed, and larger bleed hole in the stat will have cured the problem.  As of this writing I have not heard back from Ben as to whether he's tried it yet today.  I will update here when I learn more.

Thanks for all the help.

-Sean
http://OurOdyssey.BlogSpot.com

[luvrbus]

Get the air out those little guys they are not much of a problem for 20 to 25 thousand hrs I never had to replace a head or gasket on one but they are hell to bleed that is why I use the vacuum pump, that is the only way you can bleed the model that does not use a water pump and the 70/30 mix fwiw

good luck

[Busted Knuckle]

  Brian may have hit on the right idea, you dont want two pressure caps. You should only have a bleeder at the remote radiator, with no pressure cap. The Pressure cap should be highest point near the engine, and overflow bottle should be adequate volume plumbed off the pressure cap. Alternatively, many later model vehicles put the pressure cap directly on the expansion tank, making a 100% closed system. Many often run a bleeder off the T stat housing to the tank as well, making a self bleed cooling system.

I don't why you say this. Every Setra I've ever had has had 2 pressure caps. I know they've been doing it since 1986 (the oldest I have worked on) and probably longer than that and it works for them.
  BK 

[bevans6]

BK, the reason I suggested that was that, in this case, the cap marked with the lower pressure (which turned out to not be a pressure cap after all) was not the one that the overflow recovery was hooked up to.  All things being equal, the lower pressure cap would normally blow off first, and in this case, the higher pressure cap would never let coolant out to the overflow tank.  When things are old, maintenance is spotty, things are changed from the OEM install, parts get replaced with what's at hand, and most importantly - the thing ain't workin' right - you tend to try to go back to first principles and look for root causes of problems.

I confess that bleeding cooling systems, and designing them so they self-bleed, is a bit of a bug with me.  I had a race car that you had to "burp" the cooling system on.  That consisted of picking it up by the tail and hanging the entire car at a 45 degree angle, nose down, so that the air bubble would come out of the front rad.  That was indeed a PITA...    and I fixed it in a hurry!

Brian

[artvonne]

  I had a race car that you had to "burp" the cooling system on.  That consisted of picking it up by the tail and hanging the entire car at a 45 degree angle, nose down, so that the air bubble would come out of the front rad.  That was indeed a PITA...    and I fixed it in a hurry!

Brian

  I dont know if the Italians are any brighter than the Brits, but the 308 never had much problem bleeding. And the early ones like mine didnt even have a bleeder at the T-stat.

  Dont know anything about a Setra either. All I can say is the simpler the system, the less  problems, the easier the diagnosis. Why hard to bleed engine designs dont come with self bleeding coolant systems is beyond me. Its not like its rocket science.

[Oonrahnjay]

I dont know if the Italians are any brighter than the Brits, but the 308 never had much problem bleeding. (snip)

      Something about six of one and half a dozen of the other strikes a famiilyah note.  On 1987 and 88 Range Rovers sold in the US, you had to drive the front wheels up onto a 4-post lift and raise it at least 30 degrees in the air.  Otherwise air trapped around the intake manifold and top of engine would never burp out.

[artvonne]

      Something about six of one and half a dozen of the other strikes a famiilyah note. 

  I dont know except everything I ever owned that was British seemed to need to be flipped upside down to get the air out. And lets face it, the Japanese copied a great deal of British technology, so its not a big surprise their engines air lock and need some goofy procedure to get the air bubble to cooperate.

[edroelle]

I saw a very similar problem on a Country Coach.   These systems are susceptible to ingesting air because the radiator and header tank are marginal for height.   I did not understand if you changed both the radiator cap and header cap.   In the example that I diagnosed, the problem was the header tank cap.   That was only after the owner changed the thermostat, waterpump, and was ready to pull the head.

Ed Roelle
Flint, MI