Reverse engineering

Is there such a thing as reverse engineering when it comes to thread cutting. I have a dividing head with 3 different dividing plates, and I am able to cut a great variety of threads using the calculation tables supplied by the manufacturer.

However I have a 1919 carburettor with the cap missing. I have used my BSF,Whit and other thread gauges but cannot find what thread it is. Is it possible to reverse engineer the dividing head to cut the cap threads from the existing body threads, and if so how do I do it ?

Thankyou.

I'm not sure I understand the problem as I don't understand the link between dividing heads and screwcutting. However, if I may be so bold.......

Screw a suitably-sized wooden dowel into the hole. When the dowel is withdrawn it will hopefully have the mysterious thread cut into it. This will give you an approximation of the "male" thread required. Use this to take your TPI and OD measurements, perhaps subtracting a few thou from the OD to allow for clearance. If the thread still doesn't conform to a known standard thread, then make a new screw on the lathe to match the dowel.

I've fallen at the first hurdle Roger - like David T I don't know how to cut a thread with a dividing head. Does yours have some sort of attachment? How does it work?

If it were me cutting the threads on a lathe, I'd measure the number of turns per inch on the carb hoping to find a nice round number. If that failed I'd try turns per centimetre and calculate the pitch in the hope of it being metric. Once the pitch is established, I'd look for a banjo gear combination that matched perhaps settling for an approximation.

The thread form is probably 55 or 60 degrees although there are other possibilities like BA which is 48.5 To test that I'd get a 55 degree HSS cutter and offer it to the thread. Over tight means a sharper tool is needed, sloppy suggests 60 degrees and a metric / UN cutter.

My main concern would be ensuring the fit was leak proof - a carburettor thread might be a special form?

Dave

If we knew what make of carburettor was involved, it should be easier to offer suggestions about the thread. And if we knew whether the thread was internal or external we could make more logical offerings, and whether the cap or its screws were threaded ...

Over to you, Roger

Cheers, Tim

Another one mystified by, and ignorant of, cutting threads with a Dividing Head.

As for finding out what the thread was:

1) measure the I D. This will give an idea of the Core diameter.

2) Use the already mentioned soft wooden dowel method to find the t p i, (Being a carburettor of unknown provenance, to us, it may be metric, so the pitch will probably be less than a millimetre).

Armed with Core Diameter, hopefully O D, Pitch, and Thread Form, You should be able to determine if the thread is a standard one, or unique to the maker of the carburettor. (Don't be surprised, if you finally find that it is a "One Off" unique to that particular item).

It is not new for manufacturers deliberately using non standard sizes, so that are the only source of spares or tools.

Bear in mind that Whitworth form threads will be be 55 degree, whilst Metric should be 60 degrees. But at that time, it is possible that folk did not have the rules and standards to which we adhere today.

Again, it is possible that some of the figures that you find will not be exact, so you may have search for several possibilities, before you find the right one. So when you think that you know the thread, cut one, in Aluminium, or Brass, or even better in Nylon, Delrin, or Acetal. The soft material should do damage to the original thread, if you have guessed wrong! Even so don't go forcing things down the thread, that will damage it Otherwise you will end up having to retap (if there is enough material around the hole) to alloow the use of modern non prototype hardware.

Howard

A lot of early carburettors where to the manufactures own specification. A bit like Douglas motorcycles using 25 TPI instead of 26 TPI cycle thread. Take your best guess from measurement and cut a bit of aluminium or EN1A and try it on the carburettor.

.

Sorry to be a boring swot ... BA thread form is 47.5

MichaelG.

.

https://britishfasteners.com/threads/

Edited By Michael Gilligan on 17/08/2018 17:32:12

I'm also mystified......but there's nothing new there

One thing i can say with some certainty is the carb body will most likely have an external thread. So sticking a dowel in is probably not an option.

It's usually a fine pitch, which i suppose narrows it down at least a little bit.

Edit: Just been and measured a 1930's Amal.....there is only about a 1/4" of thread on top to measure, but it looks like 55 degree 24tpi

A quick google brought up this old thread https://www.model-engineer.co.uk/forums/postings.asp?th=127000

Edited By Alan Waddington 2 on 17/08/2018 18:01:30

If you can mount both the carburettor body and the blank cap concentrically to each other on the lathe you might be able to make up an assembly that works in the same way as a Unimat threading attachment (in the video below). With a follower and cutter linked together the thread on the body would guide the cutter to produce the same pitch on the cap. You might be able to use the compound slide with its screw removed for this. Personally I think it might be easier to try the nearest standard pitch with conventional tooling though.

Brian

Doh! You're right of course.

Löwenherz threads are 53.8° if that retrieves my shattered reputation.

Dave

.

Quite right!" had forgotten BA, ( a sort of Metric) which is a possibility in a UK product.

Although if a Solex or Stromberg, my money would be on metric. But given the age, could be anything, such as a "My special" .

Will be interesting to know what it actually turns out to be!

Howard

There are times when it's better to just slink off and have a beer.

The angle for a Lowenherz thread is 53 degrees 8 minutes.

And of course the thread angle for BA is 47 degress 30 minutes.

Andrew

Good advice. And when I've finished the beer I shall take up knitting.

I think beer is the best idea offered so far,good enough for me anyway !

.

How much does that differ from 47.5° please, Sir ?

MichaelG.

Mathematically it doesn't; but it is expressed in a way that is appropriate for the time when BA threads were in commercial use. In the same vein decimals would not have been used for linear measurements - fractions rule. For evidence I've just scanned a couple of copies of ME from 1959 and 1961; not a single decimal dimension in them.

Andrew

Roger please see your personal messages PM

Roger, Your unknown thread is very likely to be metric.

AMAC carbs in particular (not Amal which came later) often used metric threads.As others have said it might well be a non-standard metric thread.

What make/size carb is it? A photo would be good.

Chas