giant exhaust elbow

[rslifkin]

The Yanmar design is a poor one. A better way would be to have a dry riser for the up and over, or a coolant jacketed riser. Then an injection elbow on the down-side. Depending on the design, it can either be a simple water dump (my generator is done this way and doesn't get the rubber hose right after the elbow hot) or it can be a showerhead type mixer that spreads the water around the outside of the hose better, providing more cooling as it mixes.



You don't want raw water involved anywhere before the downhill, as it's a risk of things corroding through and dumping salt water into the engine.

[markxengineerin]

@rslifkin, I agree with all you said, but prefer to get as much cooling for the exhaust as is practical before hitting the rubber hose. The hose I'm ordering is good for 500F, but I have no idea what the maximum EGT is, or how effective this cooling will be, so best to target the most cooling possible without excessive complexity or cost. Leaning towards the 2nd design from my last post, because it will be easy to add fins to the inner water pipe for more surface area, at minimal expense to exhaust flow.

Can you elaborate on the shower head type mixer? Not sure what you mean but it sounds interesting.

Any concern with more clogging deposits ending up elsewhere instead of in the elbow w/ revised design? I wonder if they will even stick to the Flourosilicon or Nomex lined tubing. It would be really nice if they don't.

[rslifkin]

Originally Posted by markxengineerin

@rslifkin, I agree with all you said, but prefer to get as much cooling for the exhaust as is practical before hitting the rubber hose. The hose I'm ordering is good for 500F, but I have no idea what the maximum EGT is, or how effective this cooling will be, so best to target the most cooling possible without excessive complexity or cost. Leaning towards the 2nd design from my last post, because it will be easy to add fins to the inner water pipe for more surface area, at minimal expense to exhaust flow.

Can you elaborate on the shower head type mixer? Not sure what you mean but it sounds interesting.

Any concern with more clogging deposits ending up elsewhere instead of in the elbow w/ revised design? I wonder if they will even stick to the Flourosilicon or Nomex lined tubing. It would be really nice if they don't.

The water jacketing doesn't do all that much to cool the exhaust (that happens mostly when the water is injected). The jacketing is mostly to keep the outside surface of the riser cooler. So if you go dry, it would need heat wrap. Or you could do 2 pieces (riser, then a mixer on the downhill) with the riser part jacketed as part of the closed cooling loop (to avoid corrosion).

An example of a showerhead type mixer is the picture below. Instead of just dumping the water into the exhaust stream through a single hole, it adds it all the way around the exhaust at the outlet.

[markxengineerin]

The jacketing is mostly to keep the outside surface of the riser cooler.

This would cool the interface to the downstream hose too, which seems important. And also makes me wonder why Yanmar has such a long pipe prior to the injection point, but, they did a few things I don't understand.

So if you go dry, it would need heat wrap.

Which would increase the interface temp while benefitting ambient temp. The original was insulated (page 1 of this thread)

Or you could do 2 pieces (riser, then a mixer on the downhill) with the riser part jacketed as part of the closed cooling loop (to avoid corrosion).

Like the idea, but it's pretty complex if something simpler will work.

An example of a showerhead type mixer

Thanks for pic- it looks like it's asking to clog up though, with the narrow gap, and based on where the deposits formed in the Yanmar. It would be easy to implement but harder to clean.

Compromise that takes all of this into account:


Total cost is probably <$150 including the rubber hose. Still easy to clean if it clogs (no fouling of bent areas), very unlikely to overheat. I want to find a thicker stainless U-bend- anyone have sources? The better the grade of stainless, the better. I want this to last 10+ years and think it's possible with this design.

[markxengineerin]

Well, went with 2 each of these, will decide which seems better and weld together 2 90's to make a 180. I couldn't find a schedule 40 with the better grade stainless, or any 180s with good wall thickness. A 5/8" OD stainless pipe will serve as the water entry barb, same as on the Yanmar. It will be after the holiday before the parts arrive, but I'll follow up with photos when done.

[rslifkin]

Your last design should work just fine, and even if something fails after enough years of service, it should fail in a safe (non engine damaging) way. For a 1.5 inch exhaust, the simple injection port should give good enough mixing (it does on my generator for a 1 5/8" hose). Fancier injection ports mostly become needed with bigger exhaust systems.

Having a smooth bend without pinch points should also greatly reduce the risk of clogging. I'd also expect the hotter dry system (until after the top U bend) will collect less soot deposits, and the general injection design and smooth walls should reduce salt buildup as well. Of course, extended idling can gum things up with many diesels due to mediocre combustion and low exhaust temps and isn't ideal.

[markxengineerin]

I'd also expect the hotter dry system (until after the top U bend) will collect less soot deposits

Yes, this can be seen in the cross section of yanmar part- there are close to zero deposits before the water entry point.

[Uncle Bob]

Originally Posted by markxengineerin

A cut into the elbow made me realize an obvious requirement I was missing- cooling the exhaust gas enough for the rest of the system downstream to be made from cheap and convenient materials (plastic, rubber). I think this is why the water enters an internal passage and takes a long U turn instead of just being shot in with a simple fitting on the downhill side.

It's not obvious how to make the perfect elbow, but I'm thinking about something that starts here:
https://www.summitracing.com/search/...-degrees%3A180

For what its worth a piece of 4mm 1x19 ss wire about 300mm long and a battery drill will clean out an exhaust elbow rather well. The wire in the drill when inserted into the elbow will splay out at the end and scrape a lot of gunge quite quickly while remaining flexible enough to get into all the nooks and crannies. You have control of speed and direction with the drill.

[markxengineerin]

Cool idea, thanks

[Compass790]

I liked yr teardown report

I made a straight pipe sched 10 316 exhaust on our 8hp Yanmar that has the same style exhaust elbow as u pictured, can see it in this thread
.https://www.cruisersforum.com/forums...-258162-2.html
It has no cooling problems but a better design might be like this if you are worried about hose deterioration. Its pictured in the thread too

[markxengineerin]

@compass, I like the concept, but it would mean changing the diameter of the mating hose to muffler.

[Compass790]

Originally Posted by markxengineerin

@compass, I like the concept, but it would mean changing the diameter of the mating hose to muffler.

Ok then you could cap the end of the larger dia. pipe or use 2 reducers welded together & drill smaller holes around the inner pipe for the cooling water before you install the outer jacket. Have the inner pipe extend long enough to get the hose on.
I should have done another drawing but maybe you understand what I'm trying to say.
Thought you wanted a giant elbow? I'd just try a bigger dia. one if you can make it yourself.
Following with interest....

[markxengineerin]

I do see what you mean, thanks. I started with the idea of "giant elbow" as a solution to the clogging problem, but after the other input received, changed strategies to something much simpler that seems like it will work. I'll take some photos when done.

[dbraymer]

What you had clogged, changed dimension and became smaller. The envisioned replacement will do the same and eventually be clogged to the point that it has the same effect as the original part. Prior to the clogging to the original dimension, I cannot see a performance down side.

So this is all a good thing, as long as you make sure that the water injection is going to put water out of the engine, not into the engine.

I did the same with a mercruiser V-8 and I did create a problem. The engine ran more RPMs and created more power, and a longish full speed run would overcome the cooling system.

As long as you are not running/racing your sailboat 18 nm (Lake Charlevoix) with the diesel going full bore, I do not see a problem.

Most basic HP improvement advice I ever got from my WWII airplane mechanic father was "let the engine breath"

[markxengineerin]

Here's what I ended up with. Wont' know how well it works until late Spring, but I am happy w results, should be really easy to clean. Not quite done welding yet.