Savington

iTrader: (31)

The turbine bolts do NOT see exhaust, at least not on my housing. When I trashed the first CHRA, one of the bolts broke off in the housing. I passed it off to a machine shop for extraction because they weren't through-drilled - if they had been I would have just drilled straight through the broken bolt and retapped.

Nimonic is a nickel superalloy, just like Inconel - this assumption that it's going to be easier to machine is likely a pipe dream.

edit: Scratch that, it is a pipe dream. I don't have time to go through every alloy, but I just checked Nimonic 80A on the manufacturer's website and it all but says that "this **** is a bitch to machine."

Savington

iTrader: (31)

We went over this in the stud thread too, it's a possibility, but it's not going to solve the problem. You can't put enough pre-stretch into an M6x1.0 bolt to cause this issue independently.

Physically impossible. You'll hit the water lines and/or the oil feed fittings and/or the oil drain fitting. Clocking to prevent these interferences is not an option, since the turbos must be specifically clocked to fit into the car at all.

Yes, it's further away, but a) I don't think this is the primary cause, and b) there's no reasonable correction to be made

I don't follow. Vibration makes the bolts back out, so we safety wire them, and if safety wire fixes it (it does), then vibration isn't the problem?

False. If you're a faster driver, you are at higher RPM for a longer period of time. It's not just heat.

Probably true, but I don't think anyone has (or will) try this. You're basically talking about pulsing the car from 5000rpm to 7000rpm, at a 3 second interval, for 30 minutes.

I am going with the combo. It's the same exact problem with the same experimental results we went through during the stud development. Steel studs with resbond = creeped/galled/otherwise destroyed. Inconel studs, no resbond, copper locking nuts = spun straight out. Inconel studs, resbond, Stage 8 lockers = total success.

Both times the problem has cropped up for me have been at Calspeed. At Calspeed, I spend approximately 30 seconds at full boost in 5th and 6th gear, always above 5800rpm, and for ~15-18 of those seconds I am at 7300-7400rpm in 6th gear. The first time, it was very abrupt - suddenly no boost, like it had popped an IC pipe. Second time, it was VERY gradual - there was a slight metallic squeaking noise that I chalked up at the time to a loose fender bolt. I didn't realize the extent of the issue for 2-3 events after that.

Inconel for the creep resistance, safety wire for the vibration.

JasonC SBB

Typo.

If there is a strong resonance, the vibration will be in a very narrow RPM range.

When I was working at a co-generation plant with a steam turbine, the instructions for power-up were to never dwell the RPM within certain "critical" ranges. You were supposed to ramp quickly through them.

A fast vs. a slower driver would be equally likely to dwell through resonant RPM ranges. A fast driver *will* burn through more fuel per sessions and thus will bring the hardware to higher peak temperatures.

The type of track would have a much bigger effect than the skill of the driver. A very high speed track would make the motor take much longer to climb through the rev range due to the higher gears involved, and would thus dwell at critical RPM's for much longer.

I still think the simple slowly rev it up in neutral test will show any strong resonances.

Then maybe either inconel or nimonic bolts, + resbond, + those bend up tabs will work for Hustler. (Where the bend-u tabs take the place of Safety wire, obviating the need to drill holes in difficult-to-machine material)


BTW did the Resbond help by preventing the studs from backing out?

fooger03

iTrader: (2)

I would like to postulate that, although the vibration is what causes the inconel to back out, it requires that the part and bolt in question heat up differently and expand at rates that are ideal to "loosen" the bolt before the actual "backing out"/counter-clockwise rotation of the bolt occurs. This would suggest that a street driver, driving at the same RPM range as the track driver, may never heat the turbine housing enough that the threads of the bolt loosen by thermal expansion, and therefore the bolt will not be able to actually rotate to "back out" by vibration. The street driver's bolts simply stay tight enough that they never become loose enough to actually begin turning by vibration/resonance.

Doppelgänger

iTrader: (12)

I suspect that on some housings, the CHRA fastening bolts don't go all the way though. If you want, I can take pictures of the housing I have in the garage showing that all 6 holes go through and expost the bolts to direct contact with exhaust gases. Are you and Hustler running identical turbine housings? He did admit that his see direct contact as well.

hustler

iTrader: (6)

I sure thought they did, but I should probably check and confirm. Savingaids takes his turbo out more than I do.

Savington

iTrader: (31)

Pretty sure Timmay has one we can look at too, at least a .64.

fooger03

iTrader: (2)

My garret 2560 turbine/chra bolt holes go all the way though. I took note of this fact when I reassembled my turbo 4 days ago.

hustler

iTrader: (6)

lol @ your old .64 paperweight still floating around on a delivery truck or a shelf, somewhere in America. I remember that fabled day, working in Little Rock, Johnwag on his way up to pull my broken car back to Dallas, calling you and finally Tim in desperation to get something reliable in my car...that's back when we were young and dumb. Look at all the wisdom we've gained since then.

curly

iTrader: (25)

Yup, thank god you've gotten rid of your bolt problem.

Sorry, couldn't resist.

hustler

iTrader: (6)

Considering the service interval went from 3-laps to 10-hours+, I'd say it's fixed.

dgmorr

iTrader: (4)

Are there any special markings on it? I just got one and the turbine bolts look like normal black steel.

TurboTim

iTrader: (9)

There are no markings on the bolts. They are not "black". They are shiny. Look like custom machined parts from bar stock, hence no markings. All of my tial bolts are at work getting drilled or else i'd take a pic.

I wouldn't be surprised if tial changed the bolts again. They've done it a few times already. Would black oxide bolts be an improvement over stainless? They certainly don't have the CTE of stainless but still don't have the temp rating of inconel.

I checked the 3 tial turbine housings I have here, two .64a/r GT30 and one .63a/r GT28, none of the holes go thru into the gas stream.

TurboTim

iTrader: (9)

I found some at home, see attached. These are the older 8mm hex sized tial bolts. Newer ones are back to a 10mm hex and don't have the round washer-like feature under the hex.

dgmorr

iTrader: (4)

Actually mine are silver, I haven't looked at them since I first got the turbo. Are you going to be selling drilled inconel bolts, or is this something you have in limited supply only?

TurboTim

iTrader: (9)

Those look like the newest tial inconel bolts. You're good.

I may offer a service where you send me your bolts and I'll attempt to drill them. The price is not set yet but it will probably be $40 if you bought a setup from me and $80 if you haven't. Plus shipping, flat rate box is $5.

dgmorr

iTrader: (4)

Let me know if you are successful. I'd like to safety wire mine before everything gets put together.

What are these supposed to be torqued to, and are we supposed to be using thread locker on these?

hustler

iTrader: (6)

Those are Inco. Take them to an EDM shop and get them drilled straight through at .060" for ~$90, buy .032" Inco wire, wire with tension, drive hard as ****.

My safety wire advice you can disregard if you already know this:
This is bad:

Notice how mine is tight?

I started wire across the little flanges, another fail safe considering that if one flange loses both pieces of hardware your housing with either warp, cause turbine contact, or more likely you'll have both.
I twist more than most recommend, to snug the wire up to the outer edge of the bolt head, count the # of twists for future reference, run through the next bolt, then twist to 8-10 twists per inch to increase tension.

Wire tight, break a few to find the limit. You're working with inconel and everything you read on the interweb is about wiring soft SS in light duty applications.

TurboTim

iTrader: (9)

I am pretty sure I will be successful. I did 4.5 the other night but they were the shorter 8mm hex versions like the one's i pictured above. So for the 10mm hex versions if I use one drill per 2 bolts I should be able to do it quickly without drilling from both sides. should...I still have to order more of that particular drill and try it.

I don't know the torque but I can guestimate if I do math. you'd need a trick torque wrench to measure it though. I'd make it as tight as any other bolt with a 10mm head on it. Inconel is pretty strong. I love that icon.

dgmorr

iTrader: (4)

Thanks for the link to the wire.

What is an EDM shop (Electrical Discharge Machining)? I'm looking around locally for a machine shop that can drill them. I'm not in this type of industry at all, so I'm not exactly sure which shop can or cannot handle them...even if they say they can.

Were those regular SS bolts that you had come loose, hustler?