Lightening and balancing crank etc

[Stuart Wilson]

Does anyone know of a company in the north east, or not to far away, that can lighten and balance crankshaft/rods/pistons and flywheel

cheers

[99176]

Is Paul Burgess in that area ?

[JohnD]

It's in the North West, but I've had a Triumph Six crank and flywheel balanced at MiniSport.
http://www.minisport.com/

If you have any small industrial estates around you, poke about for a machine shop, and ask there.
If they don't do it, they will know who does.

Balancing pistons and rods is a different matter, one of weighing them and taking off metal until they all weigh the sam as the lightest.  But if you really want a balenced engine, the the rods need to balanced end for end, and that requires very specialist kit.  I tried to make my own - it didn't work!

Aftet you have balanced the rods, then theyshould be peened, to close up any cracks that your machining has opened.    Again, your machine shop may help.

Good luck!
John

[Sheepy]

Agree with john.

But also don't forget to give them your front pulley, and your clutch!

These also need balancing, and then once all complete, ( they are marked for the correct fitting) if you need to change something in the future, you only need to get that balanced! And it won't affect the rest!

Good luck!,  :)

[99176]

Yeah, agree with John.
Made my own rig for rod balancing, but it took a lot of thought to get it consistant.
MkIII seems to be spot on. (Pharma balances are really way too accurate, but use the toys you've got).
Just adapted it for F**d sizes.

[thescrapman]

99176 wrote:

Made my own rig for rod balancing, but it took a lot of thought to get it consistant.
MkIII seems to be spot on.

Johns one looks like it should work.

Any pictures of yours to see what differs??

Cheers

[oil_on_the_carpet]

thescrapman wrote:

Johns one looks like it should work.

Any pictures of yours to see what differs??

Cheers

Too much slop and friction/stiction in the balance points. Use a ball race for the big end and a ball race/gudgeon pin for the small end.

[piman]

Hello OOTC,

you don't need a ball race, it's better to have a knife edge strip of steel.

Alec

[oil_on_the_carpet]

piman wrote:

Hello OOTC,

you don't need a ball race, it's better to have a knife edge strip of steel.

Alec

Probably, but some angle with a ball race bolted on is simple and requires little in the way of resources to make a workmanlike solution with repeatability.

[JohnD]

If you look closely at my pic, there IS a knife edge as the fulcrum.
The Giant Crown Bottle Cap was because I didn't have a lathe to make a dummy journal.
You can just see the sticky label on the scales to mark its centre point, through the plastic bag used to mollify 'er indoors.
The level is to level the rod.  What you can't see is that the knife edge was adjustable in height to acheive level.

I recognised that my lash-up might not be accurate and hoped that repeated measurements would produce an average that could be useful.
But I never achieved consistent values.  They were all over the place, almost random.
The chart below is of the big ends, with the fulcrum at the other end, and a smaller bottle cap.
The small ends were even randomer.
Each series is a single rod.

I can see that there are two populations of rods, but I'd not modify any single rod on this data.
Be glad for advice on where I went wrong!

[69vitesse]

JohnD wrote:

If you look closely at my pic, there IS a knife edge as the fulcrum.
The Giant Crown Bottle Cap was because I didn't have a lathe to make a dummy journal.
You can just see the sticky label on the scales to mark its centre point, through the plastic bag used to mollify 'er indoors.
The level is to level the rod.  What you can't see is that the knife edge was adjustable in height to acheive level.

I recognised that my lash-up might not be accurate and hoped that repeated measurements would produce an average that could be useful.
But I never achieved consistent values.  They were all over the place, almost random.
The chart below is of the big ends, with the fulcrum at the other end, and a smaller bottle cap.
The small ends were even randomer.
Each series is a single rod.

I can see that there are two populations of rods, but I'd not modify any single rod on this data.
Be glad for advice on where I went wrong!

A very interest set of results.

If we take the mean of the two, we see that a variation between the two sets of rods of 20 grams!
No wonder the old Triumph boat anchors vibrate like a tractor engine.

FWIW, most tuning shops aim for better than 1 gram and the better ones aim for 0.1  gram.

http://www.youtube.com/watch?feature=player_embedded&v=qLkuMJYzWbY


Just out of interest I might pop down the shed later and stick some 2.5 rod/piston assemblies on my new scale and see what pops up.

[69vitesse]

Well that was a bit interesting.

A set of six rod/piston assemblies pulled from a 2.5TC engine with OE std sized pistons.

A 13 gram spread across the 6 rods.

Same result as John, the rods broke down into two distinct sets, in this case 4 within a gram and two wild cards at 11g and 13g over.

Holy whipping crank and hammered bearings!
One must presume that the factory tolerance for balance was +/- 1 oz rather than +/- 1 gram. Although that level of imbalance out of the box suggests Triumph showed little if any interest in the balance of it's engines.

[JohnD]

As you say, 69, rods should be less than a gram between them.
But my measurements varied excessively.
Eyeballing shows the two groups, but the values are so all over the place, vene that has to be in doubt.

Remembering that each 'Series' is one rod, the range of weights for each is between 26 and 52 grams.
The standard deviations (StDev) for each rod are between 8 and 16 (lets not bother with decimals)
For a reasonable distribution, 99% of values should lie within 6 StDev of the mean.
In these populations, 6 times the StDev is about TWICE the range!  200% of the values!
In other words, its CRAP DATA!
There is too much variability in the measuring method.

I wouldn't sharpen a pencil on that basis!  Let alone adjust the wieght of some rods.  How could I know if I had got it right or not?

In fact I found someone who could do it, for a fee, and I have lovely balanced rods.
But not unreasonably, he won't show me how!  Why should he?
So, I'd love to know where I am going wrong, and how others do it!

John

PS I am talkng the big end weight, as above.
NOT the total weight, which is easy to measure accurately on any set of scales that will weigh down to 1 gram.
Were you measuring total, or big end weight, 69?
Thanks for the data.  Interesting that you found the same.
Just a thought.  Was your engine as it left the factory, or could it have been rebuilt?   I could see different batches of rods getting mixed up at the rebuilders.

PPS When I was doing this, I eBayed for a set of scales to this accuracy.
There are hundreds for sale!
Often with a handy mirror attached.
For, ahem, powdering your nose, as it were.
God's Teeth! Is everyone on the game?

[69vitesse]

Your being too critical. Although the data is all over the place, it is trending, and the trending points to a significant variability in the weight of Triumph con rods.
Pistons tend to be very consistent, as do rings and pins, that leaves the rods. Just a quick look at the rods shows quite a variability in the beams and the flash thereon.
Just look at the variations in the dressing of the flash in this photo of two NOS rods.



What's going wrong? My moneys on the knife edge balance. They are fickle things and like apothecaries balances, very accurate, but very reliant on operator 'feel'. By using a rotational balance point, the part will find it's own balance point.

[69vitesse]

JohnD wrote:

PS I am talkng the big end weight, as above.
NOT the total weight, which is easy to measure accurately on any set of scales that will weigh down to 1 gram.
Were you measuring total, or big end weight, 69?
Thanks for the data.  Interesting that you found the same.
Just a thought.  Was your engine as it left the factory, or could it have been rebuilt?   I could see different batches of rods getting mixed up at the rebuilders.

I went for the simple overall weight. Pistons etc are well known for consitency being die castings, same with rings etc,

Factory was as it left the factory, that's the scary thing!
Having said that, not at all surprised.

Having owned a Sprint back in the day, the variability in parts and tolerances was simply shocking. The story of why the Dolomite Sprint came to be the Sprint rather than Dolomite 135 is well know, test engines were churning out a spread between 120-150 BHP, the high end with carefully assembled ones. The average of production engines came in at 127.
When I rebuilt the engine, the whole lot went off to Oselli in Oxford. The amount of material removed from the crank in particular was rather surprising.

When I get the NOS 2.5 crank balanced, I'll make a point of asking them to note down the out of balance. Looking at the crank, it seems at best 'balancing' at the factory was just chopping off a chunk of the web as near enoughish'..

[andy thompson]

2729 wrote:

Well that was a bit interesting.

A set of six rod/piston assemblies pulled from a 2.5TC engine with OE std sized pistons.

A 13 gram spread across the 6 rods.

Same result as John, the rods broke down into two distinct sets, in this case 4 within a gram and two wild cards at 11g and 13g over.

Holy whipping crank and hammered bearings!
One must presume that the factory tolerance for balance was +/- 1 oz rather than +/- 1 gram. Although that level of imbalance out of the box suggests Triumph showed little if any interest in the balance of it's engines.

By the time Triumph / BL were building the 2500TC I think any care factor was long gone - some of those engines were rough as guts - others (by random chance?) were smooth as honey.

I helped strip a 1968 TR5 motor that was factory original. The guy stripping it had sold it new from the Perth main dealer and remembered it.  (It badly bent in 1978 and stored until 2011).

This factory engine showed all the signs of complete engine balance by the factory with individual rods shaved to get everything spot on - the crank was spot on and the flywheel / Laycock clutch were drilled for balance. Not suprisingly when sent to the balancer for checking - it was spot on. :)

[JohnD]

Also Interesting, Andy!
I don't suppose we'll ever know for sure, but it's helpful warning, i hope to anyone who wants a really well balanced motor.  For a road engine, I suspect that you would have to be unlucky and on the fringes of the bell-curve to get a significantly rough engine.

69,
I would like to think that our data do show two populations of rods (maybe there are more?) but I don't think that the flash on them can make that much difference.    1-2 grams, maybe, but 10-20gms???
They have to come from mismatched moulds, of different thicknesses;  different batches.  As I suggested, possible at a rebuilders, but not excusable at the manufacturer.  Don't think we can sue Triumph now.

My original data above was on a set from one engine.  My other solution was to rootle around in my boxes of parts and make up a matched set, but others can't do that.  You have to have been mucking about with these a while before you have that sort of back catalogue!

John

[Nick Jones]

When I was building the PI engine I had the rods from 2 x 2.5 engines scattered about plus the notes from my Vitesse engine build.  All from the 4 sets (18 in total, two sets being the early type and one the later type with the hole at the end of the BE bolt), were within a 3 gram spread top to bottom, with those from the engine sets being within 2 grams.  I was mostly surprised that the two types weighed the same.  All Donor engines pre '74 I think.

My experience with pistons has been that if you buy them as sets then they are pretty much spot on.  If not, they may vary by several grams.  Weight difference is usually in the gudgeon pins.

Nick

[piman]

Hello John,

regarding your test results, was each reading taken by removing and replacing the rod or by lifting the big end off the scales and replacing it? My first thought, whichever method you used, was that the scales are the culprit? I would expect a more consistent result just by lifting and replacing the rod end. I also feel that the plastic bag won't help but appreciate the reason for it.
An alternative method of suspension of the free end is by very thin fishing line, which, if you do not feel the scales are the problem, you may like to try?

Alec

[69vitesse]

lagerzok wrote:

By the time Triumph / BL were building the 2500TC I think any care factor was long gone - some of those engines were rough as guts - others (by random chance?) were smooth as honey.

I helped strip a 1968 TR5 motor that was factory original. The guy stripping it had sold it new from the Perth main dealer and remembered it.  (It badly bent in 1978 and stored until 2011).

This factory engine showed all the signs of complete engine balance by the factory with individual rods shaved to get everything spot on - the crank was spot on and the flywheel / Laycock clutch were drilled for balance. Not suprisingly when sent to the balancer for checking - it was spot on. :)

Now that s an interest bit of info.

The TR5 was always regarded as a bit stellar and was pure Triumph before BL and Triumph being made to value engineer their engines in the early 70's.
Exhibit? BL requiring Triumph to drop the Dolomite  Sprints nitrided crankshaft and switching to an  inferior non hardened crankshaft.


John.

Thinking it over last night I came to the same thought as you, a missmatched set of  dies for the rods.
I'll have another close look at the rods and see if I can note any features or variability in the measurements.


I think we can all agree on one thing, things got rather cheap and nasty when BL stuck its oar in.

[JohnD]

Alec,
Thanks for those ideas to improve my technique.   I would go through the set one by one, arranging each on the rig and taking a weight, moving on the next.   This seemed to me a better way of sampling them, as the errors would be randomised each time.     Taking repeated measurements of one rod with the same set-up,  would insert the same rrors each time, but randomise them between rods.

I had seen the suspension method used, online, but it looked even more complex. This site shows a useful loking compromise
http://www.performanceengines.com/performance_images/RodBalance.jpg
and now I have a small lathe, next time I will try to make journal dummies for both ends and use them both, one to suspned and one to act on the scales.  Maybe I shall have to join the nose-powderers and buy one of those dinky mirrored scales!

John

[piman]

Hello John,

a further thought; you use the level to set the rod horizontal, then remove it? Could that introduce an error , i.e. back to blaming the scales, stiction is what I'm wondering? An alternative method to set them level is to use one one these cheap laser levels?

Did you weight each rod and then compare the total end weights of each weighing test? If one of the tests does match the rod weight, is that the most accurate?

Note, on your picture the small end doesn't have a dummy journal.

Alec

[69vitesse]

Had another look at the rods last night.

Nothing obvious dimenjsionally, although some of the rods seem to have been deflashed by a monkey rather than a nominally skilled worker.
One has actually been notched into the beam face during the grinding off. Ok for a cooking engine, not something that would  last if you tuned it.

Basic Blueprinting is obviously a must with these engines.

The big end caps, two have a different lot number on them, and in enginebuildland, thats generally not a good thing. Only a couple of digits in it, but seems to point towards a lack of consistency.
Mixed up during an overhaul? Possibly, but I no reason to suspect but can't prove.
I do however have an old engine I'm going to weigh in with known from new unrebuilt provinence. Might drop the sump and see if mixing lot numbers was happening at the factory.

[Sheepy]

Just a thought, but why do you need to have a dummy journal at all?

All you need is a mounting point for both ends, one on scale, one off, as long as they suspend the rod it will be the same as using a dummy journal?
As long as it hold them in the same place every time, it should be the same. (if you understand what I'm blabbing on about without a picture)  :)