Melted piston?

[highmark00700]

Last season I melted my 03 1/2 summit rev turbo on a long pull. I made a couple of tuning runs at the hill and here is what i was seeing. afr was 12.3 but creeped to 12.5 so I opened power jets another 1/8 turn before I went for it, egts were 1200-1225, boost was just under 9#, fuel pressure was 3# over boost. I cant remember for sure on the jetting but if I recall it was 400 mains with the power jets at 1 1/2 turns out. I was running the mxz head (12:1 compression) and straight vp110 from the pump. The turbo was a mitsu evo3 (now a garrett). It was the fifth ride on fresh spi dual ring pistons. Let me know what you think was the cause. I'm still scratching my head and its almost ready for this season. I dont want to melt it first ride again if im overlooking something! It only melted one piston, the other still looks good.

 

[zebiraross]

Mitsubishi evo turbine housings are small. A lot of back pressure with 2 strokes mated with that turbine housing. Im confident that was a big part of it. What Garrett turbo did you go with?

 

[gus bohne]

zeb is dead on with the mitsu used in the cpi kits.but here are a few more things for you to stop this as i had that system on joshs 04 800.

make sure you replace the mag crank seal, your piston shows a leanout burndown NOT a direct pipe heat kill..When a 2 stroke turbo has a leaking crank seal it goes two ways. OFF BOOST it is an air leak that leans out that cyl.
ON BOOST it is a BOOST LEAK that lowers the power that cylinder is putting out.
you feel it as slightly lower boost on the gauge and softer transition on off boost.

for drag racing or snow drags we know when we loose the mag seal on our doos right away,, boost drops in my case from 28 psi to 17 psi..same settings just a leaking MAG SIDE SEAL..nothing more.
idle was unaffected showing that the seal was not so bad off boost to cause a lean leak but wont hold under boost..

replaced the seal and all is well.. doos have issues with mag side seal life. they are very small and get heat effected over just 2 seasons..

enjoy the garrett,, no more heat,, much much better response and piles of power..fwiw, the mitsu at 22 psi on same sled ran 117 mph in 660 ft.
the garrett needed 6 grams more per pin and runs 125 mph !!!

gotta love the real air moving turbos!

Gus

 

[highmark00700]

Zeb:

I ended up going with a 2860. w/ a .86 exhaust housing. I picked the 2860 for a couple reasons. I'm not running big boost, and I liked how responsive they were at 10k ft. I think its going to be a fun all around setup.

Gus:

When I rebuild the motor five rides ago, I put in a new crank and both new crank seals so they shouldn't be leaking yet. Before I did that It was leaking grease out the mag crank seal for some amount of time, but it didn't seem to hurt it.

After inspecting some things a little more last night, I noticed a couple other things. One was that I have signs of deto on the pto piston but its very minor. It could be from the trail ride unloading at 4500 ft, but I would think that 110 octane and only 9# of boost wouldn't deto at 4500 ft especially where its not really loading the motor hard on the trail? The other thing I noticed was there is a small scar in the head across the inner o-ring sealing surface. Its very small, .005" deep or less. But looking at the discoloration in the bottom of the scratch compared to the rest of the sealing surface, I'm fairly confident that it was leaking. How much, I don't know, but I'm starting to think it could have contributed.

Thanks for the input so far guys!

 

[gus bohne]

I wont do a carbed turbo without using the exh pipe to reference the fuel pressure reg. it gives you higher on boost fuel pressure so you dont starve or run lean..AND it allows you to see the ehaust back pressure !!

even if you dont like to use the pipe for your reference at least put a gauge on it so you know whaere you are..
for instance,, that little mitsu at 9 psi has 4 psi in the pipe !!!
at 22 it was 9 !!!

One last item for ya..
DO NOT take coolant flow from your head and put it to your turbo then back into the bottle !!!! ON a DOO the bottle is on the cold feed to the engine. taking 130 degree coolant from the head thru a 200 degree mitsu journal bearing chra then dumping this 200 degree coolant into the cold side is just completely wrong and was a short cut method. every person on this site who has listened to me and corrected this has seen engine coolant temps drop and NOT climb on longer pulls..I first sound this when running my cpi 800 down the lake at 7 psi.. coolant temps just climb and climb..typically on the 800 it shows up as mag piston damage that can look lean. the lean is there but was made worse by the 200 plus degree coolant overheating that cylinder..

I see some of the smarter kit builders have fixed this, the others ignore it rather than admit the error..

the 2860 will be a much friendlier unit. just remeber to air to the rich side of the tune.. and keep the coolant temp IN the HEAD below 140..
i dont think the scratch was leaking. had it been you would have seen a coolant bottle pressure issue soon as you hit boost..the light det you saw on the pto could be from exhaust cross contamination, when one syl is going down slowly like yours there is allot of aluminum particals on the return pulse that will effect your piston appearance.



be safe.

 

[highmark00700]

Gus,

Its funny you mention the coolant hose, with my evo turbo I was using the head coolant and returning it into the bottle. Since the garrett upgrade I removed it. Then with the scratch in the head I went to my old 01 head and machined the temp sensor hole for the newer larger threads on the rev. It doesn't even have the port for coolant anymore. I haven't decided if I'm going to make some fittings to tap into the coolant where it goes into the tunnel cooler. Ive noticed that some kit builders cool the chra and some don't. Does it need it on a ball bearing turbo? I felt it was necessary on a journal bearing turbo but I'm not sure on the garrett?

 

[gus bohne]

I dont cool them but I do vent the fron and rear of that side panel to allow airflow past and thru that side.

the garretts dont run near as hot as the mitsu or aero.

the plan you have above is a solid route for coolant flow. plumb it and run it.!

 

[simbatheking]

gus, im a bit confused by this "Ive noticed that some kit builders cool the chra and some don't. Does it need it on a ball bearing turbo? I felt it was necessary on a journal bearing turbo but I'm not sure on the garrett?

what's the "chra"? i am installing an r&d turbo kit and the instructions say to rout coolant from head to turbo to bottle which u say is wrong. what would be the best way to route it?
also, the instructions i have for my kit dont mention how to route the lines for the boost controller/bov. there is a line coming off the mag side carb and im guessing that'd go to the bov. there's 2 ports to connect to on the bov. can u help me out?

 

[Spaarky]

I am prolly going to take a Azz beating on this but here is goes.

The Mits that was on the CPI kits was the small tdo516g if I remember correctly. I understand the small housing. That small 16g is only 5 cfm short of the 2860 from my research. So why the big difference in air flow?

Gus what Garr where you running to see that big of a difference?

Also, maybe I am not understanding your comment on pipe pressure. If I am running the fuel pressure off the pipe pressure, wouldn't I see the difference in my fuel pressure?

My buddy that helped me build my carb turbo is a HUGE supporter of the EVO's. I have considered switching to a Garret just for ease of oiling it. I am not mechanically inclined, but typically catch on fairly fast. My electrical system to run my pumps is a nightmare for a novice.

My experience is much different. I ride like a average person. I am not making huge pulls. Ride trees, climb some hills, play in meadows. I have worked it pretty hard, and checked temps. I can lay my hand on the turbo, and my charge temps are not out of line.

I have been on some Gar sleds. I don't notice any quicker spool up. Matter of fact my sled engages around 4000 and I have boost right there. It builds right along with me hitting the flipper. I have lined my 700 up with 800s and it's a dead heat or I out pull them a little. Maybe they were poor 8's, not sure.

I will admit, mine is far from a stock out of the box EVO. I just don't get the big difference in experiences. Not that I am overly experienced in the turbo world. I keep reading over and over, Evo is junk, Evo is junk. It really makes me want to see if the pasture is greener on the other side. It's just hard with the experiences I have had.

 

[gus bohne]

you should cool the chra on a trail sled. e mail me
gus.bohne@comcast.net ON YOUR OWN email nit thru this site !!!

i will be happy to help you .

addict. I know you two love your mitsu. what size cpi used is not something I cared to even investigate. our pipe pressure was 8 psi over boost at 22 psi.

it was a soggy boggy dog, swapping NOTHING else but the turbo to the gt 2876 .84 ar is boosting at 3800 rpm.

mitsu ran 119 mph at 22 psi on 660 ft
garrett runs 125 at 22 psi. 127 at 27 psi

We both love our turbos bro, i know there are other options and I certainly believe your sleds both have great tunes..

 

[Spaarky]

I was hoping you would give me one of those Gus headaches that would make me think for days and jump ship.

 

[turbo800]

I am prolly going to take a Azz beating on this but here is goes.

The Mits that was on the CPI kits was the small tdo516g if I remember correctly. I understand the small housing. That small 16g is only 5 cfm short of the 2860 from my research. So why the big difference in air flow?

Gus what Garr where you running to see that big of a difference?

Also, maybe I am not understanding your comment on pipe pressure. If I am running the fuel pressure off the pipe pressure, wouldn't I see the difference in my fuel pressure?

My buddy that helped me build my carb turbo is a HUGE supporter of the EVO's. I have considered switching to a Garret just for ease of oiling it. I am not mechanically inclined, but typically catch on fairly fast. My electrical system to run my pumps is a nightmare for a novice.

My experience is much different. I ride like a average person. I am not making huge pulls. Ride trees, climb some hills, play in meadows. I have worked it pretty hard, and checked temps. I can lay my hand on the turbo, and my charge temps are not out of line.

I have been on some Gar sleds. I don't notice any quicker spool up. Matter of fact my sled engages around 4000 and I have boost right there. It builds right along with me hitting the flipper. I have lined my 700 up with 800s and it's a dead heat or I out pull them a little. Maybe they were poor 8's, not sure.

I will admit, mine is far from a stock out of the box EVO. I just don't get the big difference in experiences. Not that I am overly experienced in the turbo world. I keep reading over and over, Evo is junk, Evo is junk. It really makes me want to see if the pasture is greener on the other side. It's just hard with the experiences I have had.

OK, with the Mitsu's there is small 16g, big 16g, and EVO III also considered a 16g. The small 16g has a 46/60 compressor wheel. The big and EVO use a 48/68 compressor wheel. The EVO flows the most.

A GT2860 uses a 47/60 comp wheel. Most closely matched to the small 16g. The inducer is within 1mm of each other, but the exducer is 8mm bigger on the Mitsu EVO. The EVO flows about 575cfm but I don't know how it stacks up to the 2860 as they are rated in lb/min.

The CPI's used the EVO which is a TD05H turbine. They are rated in cm2 instead of A/R. It has a 7cm2 housing which would convert close to .49 A/R. I think that it is their only big downfall. When you start working them hard, that is the bottle neck. As a result you start to see high EGT's on long pulls. They are easy to change out with the V band fittings. I ran a TD06 w/ 8cm2 on my hybrid I built, but its was still only a .57 A/R. An 11 or 12cm2 housing would get you close to a .86 A/R garrett housing for size. They are available but usually have a different bolt flange pattern.

 

[highmark00700]

Im finally getting close to finishing it up. I havent had much time to work on it lately. I was a little off on what my jetting actually was. I was running 360's in it and I checked to see where my power jets were at when it went down. They were at 2 1/4 turns out. I jetted up to 400's for my maiden voyage with the garrett.

What mains is every one running with their 800 ho turbos? It seems im a bit richer already than what others have said they are running?

On a side note I just remembered one change that was made prior to the burn down but im not sure how much it affected it. I had my dealer retard the timing 2 degrees from stock to try to help lower egts on long pulls but I decided I didn't like how it seemed to build boost slower. I had him put it back to stock and this was the first long pull it made after that. Is anyone else messing with the timing in the ecm?

Is an intercooler going to be worth while on the 2860? I noticed a big drop in my charge tube temps with the mitsu turbo but it will be a bit easier to build my new charge tube without the intercooler.