Yanmar 4hj5E engine temp/overheat issue

[vasiliy]

Hello, I have yanmar 4jh5E engines on our Leopard catamaran (and SD50 sail drives). They were just given some love by having heat exchangers, sea water pumps, belts, thermostats replaced as we had some overheat issues on our starboard engine (which seem to be fixed now as the temp gauge shows a constant 80C at 2500 rpm).


We did not have an issue with the port engine overheating. However, during a motor run today, the temp was showing 90-110C at about 2500 rpm at the gauge for this engine; it kind of moves up and down a little. Note: the "idiot" light (the overheat alarm light on the engine panel) does not (and never did in the past) signal that overheat is happening - maybe the switch is bad? or maybe there is no real overheat issue at the engine?

Here is what I have done so far for the port engine to isolate and fix the problem:
- a new heat exchanger
- a new sea water pump
- a new belt
- a new thermostat
- a new temp sender

- fresh water comes out at a good rate from the exhaust
- swapped thermostat from the other engine: no difference (so the thermostat itself should not be the issue)
- swapped temp gauge from the other engine panel: no difference (so the gauge itself should not be an issue)
- removed the thermostat and ran the engine at the dock without a load (i.e., sail drive not engaged) for 20 mins at 2000 RPM: temp gauge showed a stable 40C (so it looks like coolant was flowing through the heat exchanger?)
- installed the thermostat back and did the same no-load engine run at the dock: the gauge showed 100C (a bit of up and down movement +/-10C) and the temp at the place on the water pump where the temp sender and thermostat are located was 76C (which is what the thermostat is rated at for opening) - based on my infrared temp gun readings (same as on the other engine which does not overheat)

I am puzzled now. It does not look like the thermostat is the issue. When it is removed the temp at the gauge showed 40C and when the thermostat was installed back, the gauge showed 90-110C while the temp at the water pump where the temp sender/thermostat are showed 76C. I am confused why having a thermostat in the pump creates these temp differences/overheating "signal" even though the thermostat itself is okay.

1) Is checking the temp at the temp sender/thermostat sufficient to determine if the engine really overheats? Does it matter if the engine was not under load but high rpm? How to conclusively check with an infrared gun if the engine really overheats?


2) If the engine does overheat based on #1, what should I be checking next? I can only think about replacing the fresh/coolant water pump and potentially checking if coolant is flowing through the engine block, etc. (how - do we need to remove the engine head?)


3) If the engine does not overheat based on #1, it sounds like a wiring issue between the sender and the gauge - does that sound right?


Thank you!

[pcmm]

Why do you think 90c is overheating? that's normal engine temperature! The thermostat is doing exactly what its supposed to be doing. Staying closed until the engine warms up! And keeping you in that 80-110 range. When you remove the thermostat the engine can't regulate itself so runs cold, which is not really desirable.

[vasiliy]

Originally Posted by pcmm

Why do you think 90c is overheating? that's normal engine temperature! The thermostat is doing exactly what its supposed to be doing. Staying closed until the engine warms up! And keeping you in that 80-110 range. When you remove the thermostat the engine can't regulate itself so runs cold, which is not really desirable.

I don't know if it overheats, I want to confirm that with a temp gun as one of the steps in my process and one of the issues is that I don't know if just measuring at the temp sender is enough. Another one is, should I be measuring at the dock with no load on the engine.



To answer your question about why I think this may be an issue in general is related to how the two engines compare. The starboard engine stays at 80C flat, does not move up or down on the gauge. The port engine gauge swings up and down (10-20C swings) and is 90C at the lowest temp shown, usually 100C or more. I don't believe that is the temp it should be at (it should be at 80C?)

[Tin Tin]

Have you checked the exhaust elbow, whilst the water flow may look the same as the other engine it is hard to tell for sure. Swap over the elbows and see if that makes a difference. It only takes about 15 minutes, assuming no seized nuts.

[pcmm]

Originally Posted by vasiliy

I don't know if it overheats, I want to confirm that with a temp gun as one of the steps in my process and one of the issues is that I don't know if just measuring at the temp sender is enough. Another one is, should I be measuring at the dock with no load on the engine.



To answer your question about why I think this may be an issue in general is related to how the two engines compare. The starboard engine stays at 80C flat, does not move up or down on the gauge. The port engine gauge swings up and down (10-20C swings) and is 90C at the lowest temp shown, usually 100C or more. I don't believe that is the temp it should be at (it should be at 80C?)

Switch the thermostats between the engines and see if the temp behaviour follows.

[vasiliy]

Originally Posted by pcmm

Switch the thermostats between the engines and see if the temp behaviour follows.

I did that (see my original post) and the behavior on the port engine stayed the same.

[pcmm]

Originally Posted by vasiliy

I did that (see my original post) and the behavior on the port engine stayed the same.

You swapped the gauge, but did you try swapping the temp sender? Its the device that sends the resistance value to the gauge. Swapping the gauge should be low on the list of comparatives, but at least you know now that the gauges are both calibrated the same

[vasiliy]

Originally Posted by Tin Tin

Have you checked the exhaust elbow, whilst the water flow may look the same as the other engine it is hard to tell for sure. Swap over the elbows and see if that makes a difference. It only takes about 15 minutes, assuming no seized nuts.

I will need to see if I can do that easily. The hoses are somewhat old and stale so it is not too easy to remove them from the exhaust manifolds.



However, we did do a bucket test where we measured the amount of water coming out of each exhaust for 15 secs and it was about the same between the two sides. We did this several times to make sure we didn't have a bad measurement.

[GordMay]

According to the manual* [page 53] the normal coolant operating temperature is 76° ➛ 96°C [169°➛194°F].

* Operation Manual
3JH5E, 4JH5E, 4JH4-TE, 4JH4-HTE1 4JH4-HTE ➥ https://www.yanmar.com/media/global/...JHM-M00201.pdf

[vasiliy]

Originally Posted by pcmm

You swapped the gauge, but did you try swapping the temp sender? Its the device that sends the resistance value to the gauge. Swapping the gauge should be low on the list of comparatives, but at least you know now that the gauges are both calibrated the same

I will give it a try. The port engine sender is new (a few days old) but it never hurts to try.

[vasiliy]

Originally Posted by GordMay

According to the manual* [page 53] the normal operating temperature is 76°➛96°C [169°➛194°F].

* Operation Manual
3JH5E, 4JH5E, 4JH4-TE, 4JH4-HTE1 4JH4-HTE ➥ https://www.yanmar.com/media/global/...JHM-M00201.pdf

That is helpful. Does it matter if I measure under load or not? It did go up to 110C (sometimes 120C) when we were motoring the other day.

[Madehn]

+ 1 on the temp sender being the issue. Great omitting other possibilities. Strongly suspect that the sender is the issue. Good luck.

[Tin Tin]

If you have old hoses, one of the hoses may have collapsed internally and slightly restricting coolant flow, it is why they are in the service manual as a replacement part every 4 years / 1000 hrs. I imagine plenty of folks go way past this time line.

[GordMay]

Basic Engine Gauge Theory and Testing
➥ https://www.cruisersforum.com/forums/f54/basic-engine-gauge-theory-and-testing-645.html

Temperature Gauges & Senders:
1. Disconnect (Tan?) Sender Wire from Gauge (Terminal “S”) and turn Instrument Power ON.
The Temperature Gauge should indicate ZERO Scale (< Low Temperature).

2. Short the Gauge Sender Terminal “S” to Ground Terminal “G”.
The Gauge should deflect to FULL Scale (> High Temperature).

3. Measure Resistance from Tan Sender Wire (disconnected from Gauge) to Ground.
American Temp. Senders will read: 450 Ohms (Engine Cold @ 100°F)
or 29.6 Ohms (Engine Hot @ 250°F )
European Temp. Senders will read: 281 Ohms (Cold @40 °C) and 22 Ohms (Hot @ 120°C)

4. Disconnect Tan Wire from Sender on Engine Block.

5. Test continuity of Tan Wire.
Zero Ohms from Block End to Gauge End - If not, repair or replace Wire.

6. Measure Resistance from Sender Terminal (on Block) to Ground on Block.
Should read as per (3) above - If not, replace Sender.

[skipperpete]

This advice comes a bit late but the first and second things you should have done were 1. Ascertain the actual engine tempereature with a reliable thermometer and 2. Open the harness plugs and check for corrosion, first the one on the engine and then the one at the panel. Not a bad idea to ground out the temp switch while the engine’s running to see if the overheat alarm actually does work.