Proper torque for cylinder head nuts

[Member 10617]

A friend and I replaced a cylinder head on his racing car yesterday (a Sprint engine). The specs called for tightening the cylinder head nuts to 40 lb. ft. (iron head, not an aluminum head). Just out of curiosity, I checked for proper torque for the TR3. There was no torque given in the Practical Hints (only the sequence for proper tightening), but Haynes and the Service Instruction Manual both indicate 100 to 105 lb. ft. (!) Can that be correct? It seems high. I wonder why there is such a difference between the TR3 engine and the Sprint engine (.... just curious).

 

[CJD]

Because the Triumph is areal engine. The sprint is a toy!

 

[CJD]

Kidding aside, Ed...105ftlbs is a bit on the high side for heads I have worked on, but 40ftlbs is WAY on the low side. (I think my model airplane engines were torqued more than that). It may be correct, but I'd double check to be sure.

 

[Member 10617]

As for the 40 lb. ft., the instruction manual for the engine clearly indicated that. That's all I know. And the owner/driver has been racing this car for many years, so I guess it is correct. (I may be wrong about if it is a Sprite engine, but I'll double check).

105 lb. ft. sounds high to me also, but both Haynes and the Triumph Instruction Manual call for 100 to 105 lb. ft. If I ever have to work on my cylinder head, I'll keep it down to 100 lb. ft., unless I learn otherwise. Guess I just have to trust the manuals.

 

[Member 10617]

Yep... it was a Sprite engine... but the torque was 42 lb. ft., not 40. Still much lower than TR3's 100-105 lb. ft. (?)

On a related issue, I have read that about every 1000 miles of street use of one of these LBCs, the cylinder head should be retorqued. If the car is raced, it should be retorqued before every race. I wonder how many people retorque after the first 1000 mile check-up after an engine rebuild, and whether is it is reallly necessary.

 

[TR3driver]

Ok, I'm just guessing here ... the Sprint has much smaller diameter head studs, right? I suspect the trick is not so much that a high clamping force is required, but that the studs have to stretch enough to maintain force over a small variation in length, due to differential expansion as the engine warms up. I know that ARP supplies "necked" studs for some engines, just for that reason. Anyway, just a WAG. 105 ftlb is also the "standard" torque for a 1/2-20 Grade 5 bolt. Somewhere, I've got a nice article on bolted joint design that talks about bolt stretch and why it is critical for a reliable joint. Let me know if you'd like a copy.

At any rate, yes, the 100-105 spec is from the Triumph factory and I've been using it for a lot of years. There was even a service bulletin that restated that value (and specified to do it cold, which ISTR contradicts the info in PH).

IMO the reason they give a range is because torque wrenches (and the humans reading them) are never perfectly accurate. In other words, the true spec is 102.5 ftlb, plus or minus 2.5 ftlb. So that is what I shoot for.

 

[Member 10617]

Randall. I don't know if the Sprite has much smaller diameter head studs... they looked the same diameter as the ones on my Triumph. But you may be right.

The only difference I noticed is that the Sprite engine has nine studs, and the Triumph engine has ten studs.

You speak of studs stretching, but Moss seems to warn about stretching in its catalog... as a reason to replace studs after some use (or perhaps they just mean "distortion" of the studs). Seems that they are saying that when the studs stretch (from over-torquing), you can no longer torque them properly. I don't know anything about this...

You are right that the Triumph factory recommends torquing the cylinder nuts when the engine is cold. And you are right that this contradicts the info in PH. But the PH seems to referring to "retorquing" from time to time, and not setting the initial torque. I don't know why this would make a big difference, but perhaps it does.

 

[Geo Hahn]

...105 lb. ft. sounds high to me also, but both Haynes and the Triumph Instruction Manual call for 100 to 105 lb. ft...

Except in my Haynes manual at least -- page 48, paragraph 59.7 they state the correct torque as 45 ft/lbs. A rather significant error.

As for that 105 -- worth noting that compared to many of its contemporaries, the head nuts on the TRactor engine look about the size used for the deck guns on the Bismarck.

 

[Don Elliott]

I have re-done the head and/or head gasket on the 1991 cc engine in my 1958 TR3A at least 15 times over the past 54 years and I always torqued the nuts on the head to 100 - 105 ft. lbs torque. These studs are still the poriginal ones installed in Coventry and they have been in the block for all 54 years and 191,000 miles.

 

[TR3driver]

You speak of studs stretching, but Moss seems to warn about stretching in its catalog... as a reason to replace studs after some use (or perhaps they just mean "distortion" of the studs).

Right. They are talking about the studs being permanently distorted, from being stretched too far. Like most metals, steel acts like a very stiff spring up to a point, and then permanently deforms if stretched beyond that point (known as the elastic limit or yield point). The stretch below the elastic limit is very small and hard to measure, but it is real and is an important part of any bolted joint design.

 

[trrdster2000]

All, I agree with the torque numbers on the TRactor heads, been doing it for years and I do need a bit of help with the pulling now, metal extension bar . I always start the engine and let it get warm, no water yet, and then re torque, drive 500 or so miles and do it again. I have done so on all LBC's and never had one come back for a leaking head gasket.

Wayne

 

[This14u]

Just received an ARP stud kit for my 1500 Spit. They recommend 46 ft/lbs using their assembly lube, manual says 46 ft/lbs. Studs are 3/8" with head thichness ~3".

 

[Lukens]

Wow, that's sayin' something. <--- Regardning Mr. Elliots TR

 

[Member 10617]

Let me ask this question again: Is it advisible/recommended/necessary to routinely re-torque at some interval? I don't use the car as a daily driver, am pretty gentle on the engine, and don't drive over 4000 miles per year. I read conflicting views on this.

BTW: My friend, whose Sprite engine we took apart yesterday to install a new head gasket, runs that engine in a Jabro racing car. Having once owned a 1960 Sprite myself, I am amazed my friend gets such power and speed from that little (modified) engine! I believe it is a Crosley engine.

 

[TR3driver]

Except in my Haynes manual at least -- page 48, paragraph 59.7 they state the correct torque as 45 ft/lbs. A rather significant error.

So it does. But they corrected it for the 1980 reprint. It's also listed as 100-105 in the chart on page 17.

The chart I have gives 40 ftlb for a standard Grade 5 3/8-20; so I guess the Spitfire studs are made of sterner stuff than SAE Grade 5. Same chart gives 60 ftlb for Grade 8.

 

[TR3driver]

Let me ask this question again: Is it advisible/recommended/necessary to routinely re-torque at some interval?

IMO no, once you're past the 1000 mile mark on a freshly assembled engine. It probably won't hurt if you want to do it, but I don't see any need. The only exceptions might be if you have an engine with a history of blowing head gaskets, or if the nuts went significantly farther at the 1000 mile retorque.

 

[Member 10617]

Randall,

Thanks for the good advice. At the 1000 mile mark on my rebuilt engine, the shop guys retorqued the cylinder head bolts. Since then, I have had no apparent trouble with the head gasket. So, I will leave the bolts alone.

My friend, whose Jabro's head gasket had to be replaced, told me that the cause of it was overheating because of a loss of coolant during a race. I don't have any overheating problems, so I'm probably safe for the time being on that score.

 

[PKPoole]

For the TR3, 100-105 ft-lbf is correct. You should always lube the stud to ensure the torque applied is going to pressure on the head not thread friction. Also, I alway retorque after I can get the engine running and warmed up (Hot-torque). I don't wait for 500 miles.

 

[CJD]

I would think the need to retorque...occasionally...lies with the copper gasket. Unlike steel or composite gaskets, copper tends to creep with stress and heat. So slowly, over time, the copper will compress under the pressure of the head and the heat from the motor. On very high compression engines I have had the copper slowly squeeze outward from the combustion chambers. It looks perfect, just deformed.

Randall is right about the bolts. They have to stretch slightly (within the elastic limit) to retain their clamping force and not come loose. If over torqued, they exceed their elastic limit and becoem permanently deformed. If that happens they must be replaced. Of an aside, many modern engines torque their heads past the elastic limit on purpose. But, they tell you to use new bolts during a rebuild.

 

[TR3driver]

I would think the need to retorque...occasionally...lies with the copper gasket. Unlike steel or composite gaskets, copper tends to creep with stress and heat. So slowly, over time, the copper will compress under the pressure of the head and the heat from the motor. On very high compression engines I have had the copper slowly squeeze outward from the combustion chambers. It looks perfect, just deformed.

I agree. However, I believe Ed is running stock compression (or nearly so) and a stock composition head gasket.

I wouldn't use a solid copper head gasket on a street engine. Even the TR racers seem to be moving towards custom made composites, because the solid copper ones are just too hard to keep sealed.