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 Midget and Sprite Technical - TRE Quality?

Just recieved delivery of a pair of TREs for an early Spridget rack. The one with the fatter track rods.

I am not sure about the quality. The threads on the track rods are in good unworn condition. The lock nuts are close fitting with no "slack" feel to the threads.
But the TREs themselves screw on very losely. With 25mm of the threaded rod fully engaged, but without the lock nut tightened, there is discernably side to side slack in the TRE itself. i.e. it can be rocked radially on the threaded rod.

It probably feels worse than it actually is so I don't know if this is either normal or acceptable. Tightening the lock nut onto it would certainly secure it firmly with no wobble, but does this sound right?

GuyW

My rack's the Triumph type so not a direct comparison and as you know my memory's not the best but IIRC the last three sets of new, modern made TREs had a bit of slight wobble when thread on, about 13 turns each. Sorry I can't remember how much wobble exactly but more than I was expecting. They all seemed fine once the locknut was applied and I had no trouble from them.

One set I bought rusted so badly so quickly that I returned them and got my money back. The last two sets have been First Line ones, the latest set don't seem too rusty, the previous set I had to paint IIRC.

All but the rusted returned set have also been adjusted later and at least two sets removed and refitted with no problems.

My steering rack was an Argentina made one and fitted new 6 years ago.

HTH
Nigel Atkins

Hmm.. Not a spot for loose threads.

I somehow managed to buy some with grease nipples in em for the later rack. Not quite sure how I managed that but there was no slack in the threads
Rob Armstrong

I recently bought some NOS TRE's at beaulieu. I was trying them on a spare rack and they fit very snugly with no wobble. That said if the lock nut makes them secure its likely to be OK
Bob Beaumont

Yes Rob, these have grease nipples.

Although a little slack, I can't see them 'letting go' once the lock nut is nipped up. Just not sure what to be prepared to accept.

I could put a couple of turns of PTFE tape around the thread.
GuyW

Sounds a bit loose. I can't recommend a good supplier as mine fitted OK but the rubber boots perished very quickly (but found a different supplier - Leacey - to just replace the boots).

In the old days I used to think the genuine parts were just OK and pattern parts often rubbish. Now I know how spoilt I was in having the choice that included something up to the job!

Let us know if you find a source of good TREs.

Thanks
MIke
M Wood

I have tried some measuring, and then loked up some imperial tap size charts. But this is beyond me - I don't understand what the information is telling me!

The track rods are fatter than the later ones. These are 5/8 diameter. Using a digital caliper gauge:

The max diameter across the peaks of the threads on the track rod is 0.622" Is this what is called the Major diameter?
The internal diameter max reading I can get on the TRE threads (again, this is obviously across the thread peaks) is 0.574

The thread pitch is 18tpi

But what does it all mean!

I presume that the threads are "meshing" to a depth of (0.622-0.574)/2 = 0.024", but there are lots of them! So what should it be? Perhaps that's OK?
GuyW

Sorry Guy my measure of wobble was obviously a lot less than yours if you need ptfe tape, I think 'd send them back.

TREs with grease nipples are freely available AFAIK I had a set but the boots didn't last so I got a set of Rhino rubber boots and they lasted longer but still perished. Does general grease speed up the boot demise?
Nigel Atkins

Guy,

I don't know how much wobble is acceptable. But, I think if it is too loose, the fact that the lock nut stops the wobble does NOT make it okay.

If there is not a tight enough thread engagement (I think expressed as %) then the TRE could pull off of the rack, and the pressure from the lock nut could make this more likely.

By the way, is it time for you to post another photo of your progress?

Charley
C R Huff

Thanks Charley,
That is sort of my feeling, but I was hoping that someone with the right technical knowledge would know what the depth of engagenment of the threads should be, given the 5/8" diameter and the 18tpi thread pitch.

Alternatively, can someone (in the UK) give me a known reliable supplier. These came from a company that I have bought from over the years and believe to be sound. I know that they will refund or replace without quibble. And that is exactly why I am seeking advice on what to expect of the fit. I don't want to go back and complain if these are "normal". OTOH if they are deemed not acceptable, then the company needs to know that they have a bad batch on their shelves!
GuyW

I have to say, it doesn't sound good.

I may have a NOS TRE that I can measure.

I see that Moss list a GSJ169 at £9.40 and an 'alternative' GSJ169Z at £8.30

It would be interesting to see the difference.
Dave O'Neill 2

Yes, this is a GSJ 169 at a miserly £4.93 each. Same number doesn't mean same supplier.
GuyW

I believe that Moss use the 'Z' suffix for cheaper, pattern parts, whereas the non-suffixed parts #should# be OE spec.
Dave O'Neill 2

Dave, that matches their pricing, 'though still 2x the price of my supplier. The 'wobble' on mine may be normal for these early ones. But it doesn't feel right. I was hoping someone could give me a judgement on those dimensions.

Claimed OEM spec doesn't mean there cannot be a manufacturing fault. These look to me as if the drill and tapping into the end has been done on a lathe with a loose or worn headstock, so has cut oversize.
GuyW

Anyone tried these?

£16.63 each:
https://www.scparts.co.uk/sc_en/british-cars/mg/austin-healey-sprite-1958-1971-and-mg-midget-1961-1979/steering/steering/track-rod-end-74513.html

Or at £7.20:
http://www.leacyclassics.com/parts/austin-healey/healey-sprite/steering/gsj169z.html

Or from Morris Minor suppliers such as:
http://www.morrisminorspares.com/steering-c92/steering-all-c93/track-rod-tie-rod-end-1958-onwards-with-grease-nipple-5-8-unf-thread-88g414-gsj169-p831419

http://www.morrisminorspares.net/shop_item.php?ID=3024

Cheers
Mike
M Wood

Mike, thanks for those links. My plan is now to buy another pair, probably from one of those suppliers, and then use the best fit ones and probably return the other 2.

I today did find the old TREs stripped off this rack by the PO. They have been butchered during removal so aren't useable but I could still measure the internal thread diameter as a comparison. They are a tighter fit, though they still have some wobble effect.

The internal diameter on the new (loose) ones was 0.574
The old pair now found measure 0.562
The mesh of the threads works out as 0.030 as against 0.024 for my new loose ones. I don't know if that 6 thou" difference is enough to be significant
GuyW

" I don't know if that 6 thou" difference is enough to be significant"

25% is quite a significant increase, although it could be more significant than it sounds.

If the thread engagement is too loose, it is only the tips of the threads that will be in contact and the cross-sectional area is considerably less than a full engagement.


Dave O'Neill 2

Yes Dave, that is what I was thinking. Obvious practical geometry really, but without the knowledge of what is expected it is hard to be sure if common sense geometry is enough!

I am just surprised that there isn't a technical standard that would confirm what the depth of engagement of a male and female thread should be. Clearly if it were 100% then the friction over the length of the engaged thread (which is quite a lot more on a TRE than on, for example, a half nut) would be far too great and it would just bind up solid. At the other end of the spectrum, at 0% it simply wouldn't fasten up at all!
GuyW

Does this help, Guy? It turns out that there is a technical standard.


Greybeard

Well Greybeard, l am neither an engineering or a mathematician, but l can follow a set of instructions even if l don't know what they mean!

Your diagram has p = 1/tpi,
so for my TREs, p = 1/18 = 0.055555.

It then says the depth of thread engagement is 0.54127p
0.54127 X 0.055555 = 0.03007

So that is the target depth of thread engagement. Interestingly it is independent of the overall diameter, relating only to the thread pitch.

My knackered old TREs match this 0.030 to 3 decimal places, whereas my new ones fall short by 0.006.

It's pleasing in that a) the figures do add up and show me that the theoretical figure does match the old test ones, and that b) my new ones are indeed machined oversized, so l have a legitimate case for complaint. The only missing perameter is that no +/- tolerance is given.

Many thanks GB !
GuyW

The fact that your TREs are machined oversize, presumably means that thousands of others are, also.

I wonder what the suppliers/importers would do about potentially dangerous parts being fitted to cars, many of which don't even have an annual safety check.
Dave O'Neill 2

Woke up to realise it is worse than l thought, which is unusual. Things usually seem worse in the early hours!

0.006 oversize doesn't sound like much, but it's 20% of the specified 0.030 thread engagement. And because of the radial wobble this 0.006 can be doubled on one side, giving only 0.018 of thread engagement when it should be 0.0300. And also it's only the much weaker peaks of the thread that are meshing and would be very close to stripping or jumping a thread when under tension from a hard cornering wheel.
GuyW

Guy.
When you talked about thread engagement in percentage terms I wasn't entirely clear about what you meant, but I had assumed (or speculated) you were thinking about a percentage of the area of the thread flank.

As you say this is readily deducible from the values given in ratio to the pitch. Up to a point anyway - presumably deformation (stretch) will change it in response to the applied torque.

So from your own measurements it seems to me that you are correct and I would not want to use the new parts either. But I think you already knew that from the "wobble" you described. You're experienced enough to know if it doesn't feel right.

I wonder if the lack of +/- tolerance is because the drawing takes no account of manufacturing quality or grade. There will be a significant difference between machined and rolled threads for example and perhaps this is addressed in the QA standard which I haven't seen.

Good luck with getting better ones!
Greybeard

Missed the window...

"Interestingly it is independent of the overall diameter, relating only to the thread pitch." Fair enough, but the pitch varies with the diameter after all so in theory it could be expressed in relation to diameter, except in cases where a pitch value is shared by threads of different sizes. Without looking it up ISTR that some (smaller) sizes do share pitch measurements.

BTW - not my drawing. Far too neat! I pinched it off the Internet like a gentleman.

I was interested to see that the male and female threads have markedly different peak cutaways and different shapes to the troughs. I didn't know that.
Greybeard

GB, l have since discovered that l have a book with that same drawing for all the different thread forms. Interestingly several have that same figure for the parameter p of 0.54127 with UNF, UNC, ISO Metric fine and ISO Metric course all using the same value. Whitworth and BSP threads use a different figure.

I expected there would be different standards of fit. Although l have never done any engineering training l am aware that when cutting threads with taps and dies one can cut them to a tight a loose fit, as needed.
GuyW

Yup - that's because UNF, UNC and ISO share a common flank angle of 60 degrees, but BSW/BSF have to make do with 55 degrees (IIRC) which obviously changes the ratio of height to pitch.

By the way we should thank you for flagging this issue. I dread to think of the consequences of one of these things coming unhitched in a fast corner.
Greybeard

Dave makes a very valid point in that there must be a load of these loose ones in circulation, although I wouldnt necessarily say thousands. Obviously whoever sharpened the drill to cut the hole did it lopsided as it is cutting 12 thou large, 0.562 being the correct size for 9/16. Therefore every one made using that drill will be wrong. Of course if the person sharpening it can get it wrong once they can do it all the time so there could be thousands the same or worse.

Trev
T Mason

Update !

I ordered another pair from a different supplier. The problem is that when ordering there is no way of knowing exactly what you are getting as the part numbers are the correct ones. The new pair arrived, and worryingly were in identical packaging with the same label number - PMEZ1002 - as the previous, loose ones.

However, when fitted onto the track rods these are a much closer fit with only a very slight lateral 'wobble'. And when measured with my digital caliper the max internal diameter I get is 0.564. This compares favourably to the old (original?) ones at 0.562 and are a good deal closer fitting than the new loose fitting ones from the other supplier at 0.574

So my conclusion is that these parts came from the same source supplier but from different batches, and had been machined either by a different operator, or on a different machine. Maybe the loose fitting ones were of a one-off poor machining, but there is no way of knowing how large the batch was. And there seems to be no way of telling other than by trial and error whether what is supplied is going to fit correctly.

GuyW

This seems to point at quality control. If the firm that made them were supplying a major manufacturer such as Ford or VAG then there would be a much closer level of scrutiny as the manufacturer could not afford the reputational risk of poor quality together with the risk of returns etc. Of course the classic car market is far less important on the basis that your stuck if no one makes it. This problem has been highlighted many times on this forum and elsewhere. I don't have an answer but it seems to me we just have to keep complaining and be prepared for an increase in cost. As mentioned I purchased a pair of NOS TRE's for the Frog as spares (BMC packets). They came complete with the grease nipple and had a castellated nut rather than a locknut. The quality was excellent, no wobble and a perfect fit. The cost was £15.00 each!
Bob Beaumont

I finally dug out the Frog rack to check some QH TREs I got from (I think) Moss last year. Good snug fit and measure .558/ .560 internally which seems correct.

But it is worrying as I've been gradually buying components in advance and not always been able to test fit them. I've assumed (possibly wrongly) that they'd be fine. Moss will exchange up to about a year but get huffy after that I've found.
Bill Bretherton

This thread was discussed between 19/09/2017 and 03/10/2017

MG Midget and Sprite Technical index

This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.