MG-Cars.net

Welcome to our resource for MG Car Information.

Recommendations

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGB Technical - On Re-Tourquing Head Bolts

Greetings of Spring or Autum:

It's been about ten thousand miles since my rebuild of an 18GA engine.

I would like to re tourque the head bolts as a normal part of maintence. I'd appreciate suggestions on these points:

Should I just go around the tightening sequence to tighten, or, should I loosen all and re-tourque to specs?

Or, in other words,should I tighten to spec or loosen then tighten to spec, and if so, should all bolts be loosened first.

I hope the easiest way, of just tightening to spec and NOT loosening is the way to go, but, this is just one of the many things I just don't know.

Any thoughts appreciated.


glg
glg

glg,
you've done it now! half the responses will be to loosen first, the other half go straight to the punch. me, I back off 1/2 turn and re torque one at a time. certainly don't loosen them all at the same time unless you want a giant coolant gush! now, here's where i do a hybrid--after retorque to five pounds under spec, i do a final torque of all, in sequence, without loosening. then i reset valves, then timing, then carbs. just my method...doesn't mean it's the best or the only.

paul
Paul Hanley

One at a time, in sequence, back off each 1/4 turn and pull up to torque in one smooth pull WITHOUT stopping. Plan your wrench position so that nothing interferes with this continuous pull. There is enough friction at the face to cause the wrench to "break" before reaching full torque. Even the 5 under may not be far enough down to avoid this. If you don't believe it, torque without the back-off and mark the nut positions, Then back them off 1/4 turn and retorque. Some or all will go further with the back-off. If you used an anti-sieze on the studs and washer faces the effect may not be so noticeable. New oxide or phosphate treated washers/nuts may have enough anti-sieze characteristics to prevent it also.
FR Millmore

I think 'way back in auto shop, they taught us exactly as FR described. Back off a quarter, then push on forward uninterrupted until correct torque reached.
Matt Kulka

glg What the others have said but dont forget to adjust the valves.Tighten the head with the engine hot,and if you want to do the valves cold as the book says, do them the next morning.If you have an alloy head do it all cold. DENIS
DENIS

Torque the head bolts after the engine has done 3-500 miles from a rebuild. After that don't touch them at all or you are asking for trouble.
Iain MacKintosh

I have to seriously disagree with Iain. I have routinely retorqued regular customers and my own heads every year or so for the last 40 years. I have never had a problem, and I find that if this is not done for a number of years that they will pull up substantially and unevenly. You may have a problem if you don't follow the procedure I detailed above, since without the 1/4T backoff, some (loose) nuts will turn and others won't, leading to uneveness.
FR Millmore

I've always been of the opinion: "if it ain't broke--don't fix it". Once the head torque is established, and retorqued after the rebuild, why monkey around with the thing? If compression is good and it's not leaking, Leave it alone. You risk all kinds of complications when you start tampering with stuff. Snap a stud? Could happen. Crack the head? Ditto. Years ago, I observed that the guys who were always "adjusting" their cars, were the ones who spent the most time on the side of the road--hood up. Don't lay wrenches upon your unsuspecting and righteously performing car just out of a sense of "needing something to do".
R. L Carleen

Yes, but it does not stay constant after the first, or second retorque. Constant temperature cycling causes continued gasket movement. If you do it right, and your torque wrench is accurate, the worst that happens is a small waste of time. If things have moved about, and something starts to leak, then you have to take it all apart. Triumph engines, for instance, have a habit of leaking oil out the back corner of the head after 4 or 5 years. This can increase to the point of dumping a couple of quarts in a hundred miles. You don't see that car along the road because it is on it's way to recycle heaven. Retorque at slightly shorter intervals will prevent this, and maybe keep you from running out of oil in the bargain.
The guys who "are always adjusting things" are mostly not doing it right to begin with, so it only gets worse. On my daily driver cars, I rarely do anything except oil changes to them more than once a year, sometimes two.
FR Millmore

This is not done in the UK, not part of a service schedule and really asking for trouble. When have you ever seen a head slacken off after it had settled down. You risk moviing the head, disturbing the waterways seal and the seal formed between the cylinders/ gasket/ head. Remember the risks with stripped nuts, broken studs and so on. Leave it alone there's no way it goes in my maintenance schedule.
Iain MacKintosh

How may of us retorque the head on our modern daily drivers? None, I warrant. What is different about newer engines? I suspect that it is the method of assembly - i.e. probably the use of hydraulic stud stretching to get the precise stud tension that we are trying to achieve indirectly by torquing. My replacement Ivor Searle rebuilt engine had a "do not retorque head" label attached and so far after 6000 miles it is still running as sweetly as ever with no interference from me.

Regards,

Barry
73B
B.J. Quartermaine

FWIW, the maintenance summary in the Leyland MGB shop manual reprint states for R.H.D. and L.H.D except North America, "After Sales Check, check/adjust torque of cylinder head nuts". No other cylinder head nut torque check/adjust is mentioned. In the maintenance summary for North America, cylinder head nut torque isn't listed as a requirement. Clifton
Clifton Gordon

Some of the reasons new engines don't need their heads retorqued is that they use 'no retorque' head gaskets which are usually teflon coated, stretch to yield head bolts and better design.

Since the B engine is all cast iron which expands and contracts in unison I don't know how much retorquing the head is going to accomplish, except for the intitial retorquing after the car has been started up and run in.

Remember to never use your torque wrench to loosen the nuts if you are retorquing the head.
Mike MaGee

Service schedule or no, I have seen hundreds of cars with loose/uneven head bolts. This may sometimes be due to initial services not being done, but that is not likely if the car has 25,000 + on it. We used to do a booming business in headgaskets on MGB and midgets, on which the Pittsburgh area dealerships neglected to do either the pre-delivery or 500 mile retorques. I sometimes had 3 cars in one day with this problem. The paint on the headnuts, applied after factory assembly, was not broken. These cars invariably had 3-6000 miles. I have seen several hundred MGA/Bs & A-series with seepage out the right side above the distributor. These usually have 30-50,000 miles, and the nuts are always loose and uneven. Triumph 4 & 6 cyl at the same mileage generally show the left rear corner oil leak; again the nuts are always loose.
I have NEVER stripped a nut or broken a stud, cracked a head, or had a gasket fail using my procedure. I have seen plenty of each on cars I have not serviced. Broken studs always show signs of occult coolant or combustion gas leaks into the stud hole, with serious corrosion of the stud. This is a direct result of the head being marginally loose for a long time. Stripped nuts are always accompanied by other signs of overtightening. The Triumph nuts are marginal, and I replace them with MGB nuts with hardened washers (be sure there is clearance between these nuts and the rocker stands).
When I release a new engine to a customer, it has had 3 retorques and at least 10 full heat cool cycles over several days, so I see little movement at the 500 mile service, but a year later still see some. Note that mileage is not really significant, heat/cool cycles are. Overheating from any cause is justification for a torque check.
You are right on the modern cars not being retorqued, but that does not mean they shouldn't be. Very many have recurrent headgasket problems. My 3.0 Taurus is dead in the driveway as a result. It had headgaskets at 30,000 by Ford, and again at 75, and now it has 162 and is DOA. The 3.8 Fords are much worse, they take out heads and blocks as well. It's not just Ford either, lots of GM cars have similar trouble. They use not hydraulic tensioning but something called "torque to yield", which I am convinced is misguided or a scam. I have written something on the subject which I would be glad to send to anyone who is seriously interested in discussing it. Send me an email. I would like contact info for Ivor Searle, would like to know what he is doing.
In the days of 40-50,000 mile valve jobs you could get away with no retorque after the 500 mile service, but now that we have better materials and clean running engines it seems a shame to have to rebuild engines because of gasket failures which are preventable. I have offered this as advice based on years of serious consideration of real experience. If you don't believe me OK, but I hope that you don't regret it later.
Happy Motoring!
FRM
http://www.usachoice.net/gofanu
FR Millmore

So if I have a Triumph, the tractor motor in it will start doing all kinds of bad things if I don't do something to the head. Funny--my brother in law has a '57 TR3 he bought new, has never had the head off and everything is just fine.

If I loosen and retorque a bolt, how can I tell if it's out of torque? Nut position is not a good indicator. The stud could have turned with the nut. Seems like most problems were originally caused by either dealer neglect ( and some car dealers--not just import-- back in the mid century were not too good on service after the sale.) or owner indifference to prescribed checks. The cars only had a 12/12K warranty, I think, and so it was more profitable to wait to correct things after the warranty period was over.

I still say-- "if it aint broke, don't fix it" and if it isn't in the maintenance schedule--there's a reason why.
R. L Carleen

Don't be so quick to dimiss the advice of those who have years of experience in favor of the guy next door's sage advice.
Cheers,
David
David

"So if I have a Triumph, the tractor motor in it will start doing all kinds of bad things if I don't do something to the head. Funny--my brother in law has a '57 TR3 he bought new, has never had the head off and everything is just fine."
Tractor motor TRs are better than average - low compression and heavy castings, with enough cast iron to not really care, they still blow head gaskets, though less than some.

"If I loosen and retorque a bolt, how can I tell if it's out of torque?"
If you can't tell the difference in 12 or so bolts, you are well advised to keep your paws out of the machinery. When you have done as many as I have, you should be able to hit within 5 lbft by feel.
" Nut position is not a good indicator. The stud could have turned with the nut."
So mark the stud, nut and head; back it off, look at it. Most of the time the stud does not move; many brit cars have the same thread at both ends so it doesn't matter anyhow. After you tighten it, check the nut to head marks.
"Seems like most problems were originally caused by either dealer neglect ( and some car dealers--not just import-- back in the mid century were not too good on service after the sale.) or owner indifference to prescribed checks. The cars only had a 12/12K warranty, I think, and so it was more profitable to wait to correct things after the warranty period was over."
How true, and some of it by those who don't study the problem and also don't listen to those who do. You really think dealers have changed? The factories are trying to eliminate poorly trained and sloppy workers by designing things better, but poorly engineered things are worse and cost a whole lot more to fix, if they can be fixed at all.

I still say-- "if it aint broke, don't fix it" -the rest of the line is "So keep it from being broke" which is called preventive maintenance - and if it isn't in the maintenance schedule--there's a reason why. - One of which is a marketing decision to meet the competition's claims of "No maintenance".
FR Millmore

Well, I do thank all you chaps.

How well I know how the "Well, I'll just do this little somthing-or-other" can turn into a situation that's far beyond a minor tweeking.

With only 12,000 miles since the rebuild/restoration, it still runs great, the plugs burn normal, oil stays clean -- perhaps I will just thank my lucky stars and leave well enough alone. I've never asked for trouble and there is none now.

Thanks again -- always an interesting scope of opinions here, but still, this board is one of the best head-ache remidies around!


glg
glg

The one issue that has not been addressed is the material of the head gasket. (Mike MaGee touched on it) Perhaps the old school thought of retorquing every 10k miles or so is based on the copper gaskets. With today's modern composite materials (Payen) maybe it doesn't need to be done so often?
Paul Hanley

I think the copper cased gaskets were prone to greater settling, but also more forgiving with greater resilience. The modern ones tend to be beyond hope once they start to leak. I have saved quite a few coppers by tightening as soon as they show signs, much less success on the later type, as the liquid or gas leak gets into the gasket material and forces bits out.
FRM
FR Millmore

Just a few observations -

Some reconditioners use Payen gaskets and advise NOT to retorque at 500 miles or thereafter.

ARP recommend that new bolts / nuts are taken through 5 cycles of torquing up and undoing in sequence before the bolt settles down to give a constant amount of stretch at the correct torque.

Not all torque wrenches are created equal.

Measuring the stud / bolt stretch is the only way to ensure that the stud / bolt is loaded to 75% of yield point.

Once correctly loaded the stud / bolt will not lose tension - it is the gasket that may compress and reduce the tension in the fixing.

My advice is to use the Payen gasket, to ensure the nut is a smooth turn on the stud, to use a graphite lubricant on the thread AND under the washer, to torque up in sequence and undo 5 times on initial assembly and to retorque after the first run of the engine. Retorquing at 500 miles should be the last time you need to touch the head.
Chris at Octarine Services

Chris, that's the first time I've heard that studs should be stressed five times before settling down. I'm not arguing about this as there may be very good reasons for it.

However, consider the mechanical properties of the materials. Steel can be stressed up to the Modulus of elasticity at which it will still return to its original shape or length. If it it taken to the Yield Point which is just beyond this that is too late and it will quickly fracture. However a factor of safety is built in and this may be a factor of say four to six meaning that materials are stretched nowhere near to the Modulus of elasticity and will certainly revert to original size after stretching/compression.

A rubber band is easy to stretch, and snaps back to near its original length when released, but it is not as elastic as a piece of piano wire or a cylinder head stud for that matter. The piano wire is harder to stretch than a rubber band but would be said to be more elastic than the rubber band because of the precision of its return to its original length. A real piano string can be struck hundreds of times without stretching enough to go noticeably out of tune. A spring is another example of an elastic object - when stretched, it exerts a restoring force which tends to bring it back to its original length.

Bearing the above in mind I find it hard to get my head round the need to torque and release five times. Perhaps there is another explanation ?
Iain MacKintosh

I think it is more to do with the fact that you are stretching the stud by measuring torque applied to the nut.

It is the friction between the parts that varies so initially the bolt will be stretched insufficiently because the friction in the system is high. This reduces over the 5 cycles and the stretch becomes more consistent for the same torque.

You could measure this by marking the nut and observing whether it ended up in the same place on each cycle.

I just accept that ARP know what they are talking about and accept their advice - it is up to you whether you do likewise or not.
Chris at Octarine Services

" It is the friction between the parts that varies so initially the bolt will be stretched insufficiently because the friction in the system is high. This reduces over the 5 cycles and the stretch becomes more consistent for the same torque. "

I concur. I don't know about ARP specs, but certain products that my company manufactures require torqueing and loosening several times until the threads and nut/washer surfaces get to know each other more intimately. We also have different specs for lubed vs dry parts.
Edd Weninger

Get the point Chris and it makes a lot of sense. That's exactly why modern cars torque to a low figure first of all and then go in angled stages. These stretch bolts frighten the hell out of me.
Iain MacKintosh

No need to be frightened - they are stretched in the same way as our B studs - it is just a way of getting a more consistent stretch by measuring a lowish torque where friction hasn't really started to affect things and then going by angle - since the thread pitch is known the angle determines the stretch length.

For bolts where you can get to both ends, like conrod bolts, a micrometer is the best way of measuring the stretch.
Chris at Octarine Services

ARP does recommend torquing and retorquing the studs and nuts 5x to mate the threads similar to mating differential gears. A good torque wrench is essential and using a correctly sized wrench. I use a 10 - 75 lb 3/8 wrench to do heads.
Mike MaGee

I agree Chris but I am of the old school and the final stage of angle tightening just does not "feel" right at all. I suppose the manufacturers know better than us.
Iain MacKintosh

I know what you mean - as the bolt stretches it feels as though it is made of cheese and is about to let go!

One day when I have got some time I am going to do some objective tests on MGB head studs and torquing them to destruction!

I have some scrap heads and blocks I can experiment on.
Chris at Octarine Services

Many of the gaskets I have used were Payen; in general they behaved much as any other decent gasket. (there are a few decidely crappy brands around) It is quite likely that Payen has changed their construction, one hopes in the direction of improved stability, but I'll believe it when I experience it.
"Copper" gaskets as used above refers to the ones with a copper wrapping on asbestos or the like. Solid copper are sometimes used on race cars and a number of motorcycles; I have seen soft solid aluminum as well. Given a good flat surface, they can work well, they are reusable by annealing, and they are the only feasible (if tedious)way to make your own. They can be made quite thin as a means of increasing compression without machine work.They will settle less but be less forgiving of looseness, having little resilience.
There are three feasible methods of service tightening: torque, torque plus angle in the elastic range, and torque plus angle to the plastic or "TTY". The last two are NOT the same. Either of the angle methods require precise optimization for the exact application in order to be accurate.
This is the subject of my paper on TTY bolts, which I again offer to anyone interested. I have added a section on the factors involved in the friction which screws up the torque method.It is about a 6 page Word 6.0 doc, which most word processor programs should handle. If neeeded, I can probably convert it to most other common programs. I am seriously interested in knowledgeable discussion of the subject.
As necessary background, here is a link to the best description of the various bolt tightening methods:

http://www.engineproblem.com.au/main.htm > procedures > TTY head bolts Most other descriptions I found are either partial or repeats of industry advertising claptrap.
FR Millmore

Had some problems continuing my last post.
The link to ARP tech is:
http://www.arp-bolts.com/pages/tech/fastener.html
A careful read will reveal that the 5-cycle procedure is what they have found works best for their fasteners, may not apply to others. It will also confirm much of what I say here and in my paper, though in less detail.
Theory is what you use to build something, experience tells if you were right, theory tells you how to fix what experience shows, experience tells ...etc. Other folks' xperience can be used as well, but all experience and theory must be complete, accurate and understood to work. If these rules are followed, observed failures are usually the result of not adequately carrying out the theory/experience cycle, not a failure of T/E itself. The engine remembers ALL of the factors which affect it, our goal is to catch up.
Chris, you don't need to test to destruction; it might be fun, but I and others have already done it. That said, I will later post a new thread with a real life challenge that will get us all some more real data, and maybe save a few engines in the bargain.
Cheerio,
FRM
http://www.usachoice.net/gofanu
FR Millmore

FR,

But has anyone done it with the currently available sources of MGB studs/nuts?

There are recurrent problems with studs snapping - probably through poor heat treatment - but in what circumstances?

By measuring stretch on different manufacture of studs over a range of torques and to the point of failure, hoprfully we can find a source of reliable studs that can be used with confidence.

Also - and with no disrespect - what applies in the US doesn't necessarily apply in the UK - while ARP are undoubtedly superb quality and made of 85 ton steel, how do they compare with a current supply from the UK of rolled steel studs made from 75 ton steel? In cost terms we are talking over £100 for the ARP stud set but £25 for the UK set. Some studs are made in India and are cheaper still but I have had several of these fail at low torque figures.

If you have data on all these variables, specific to the MGB then I would be interested to see. But I will still conduct my own exploration!
Chris at Octarine Services

Chris,I was totally unaware of the problems you mention. I can't recall ever seeing a broken head stud on either an A or B-series engine; certainly I have never broken one. This includes some which I would rather have replaced due to corrosion, and some which showed clear evidence of over tightening - washers extruded out from under the nut like rubber. Also includes the early non-centerdrilled ones, which take a bit less torque. I have seen some stripped nuts, with no significant stud damage.The last set of B nuts I got from Moss appeared to be standard SAE "Hi-nuts" which I believe are grade 8. They are a bit taller than the factory ones.
I wonder what happened to the studs the cars came with. I read a lot of stuff about replacing all the studs etc. This may be wise IF you know that you have good ones available and are entering LeMans with high compression, but for most cases, just keep the factory ones. The gaskets do not blow from bad studs, but from bad assembly and maintenance; the factory studs do not break from being poor quality, but from abuse, if at all. I think it's a scam on those who think they are building race cars, encouraged by parts sellers. In the US, the VB catalog is clearly pointed this way. I have never bought an ARP stud, or used anyhing but the factory ones; unlike rod bolts, head stud failure is not a major disaster, merely an annoyance. This does make me rethink my practice of replacing corroded studs with good used ones from my stock, and not charging the customer for it!
Regarding building race cars, I once questioned Tom Boffo, builder/driver of a pretty well known SCCA championship Mini, what he did special when bolting it together - he said "just do what the factory manual tells you, with factory parts." I did that.
I fully agree that given the situation, you should conduct tests on the stuff available today. Do you know if the factory stud quality changed after about '75? Have had no trouble with late ones but the majority of my experience is with earlier cars. I have always wondered what "tons" the British steel specs refer to - 2000lb, 2240 lb, or other; and how close these terms are to real measurements - they always seem to be in 5 ton steps.
Cheers,
FRM
FR Millmore

The question was raised as to the specs on the ARP studs. According to the spec sheet that I have, they are 190,000 psi. The installation procedure that is given on the sheet that I got with the studs from ARP, the nuts are snuged up and loosened three times and on the third time tightened to the required torque. This is to insure that the lubricant is completely dispursed so that there is not metal to metal contact of the threads. As regards the lubricated threads vs dry threads giving a different tension, different lubricants will give different tension for the same amount of torque. ARP gives two torque specs on their head studs. If the threads are lubricated with engine oil, they say to torque the nuts to 60 ftlbs, but if using their lubricant, they are only torqued to 40 ftlbs. In actuality, torque is a poor indication of the pre-tension being applied to bolts and studs due to the many variables involved. Unfortunately, it is difficult to measure the streach on many applications (if it is even specified) and short of using crush washers as are used in the aircraft industry for critical applications, torque is the only choice we have. Cheers - Dave

David DuBois

This thread was discussed between 24/04/2004 and 01/05/2004

MG MGB Technical index

This thread is from the archive. The Live MG MGB Technical BBS is active now.