PDA

View Full Version : Threading Advice…..Again!







simonl
10th May 2015, 04:23 PM
Hi all,

I'm planning on cutting a thread, M22x1.5mm. Now I have racked up quite a few thread cutting projects with great success so I think I have my technique pretty good. It's the technical (dimensional) side that I am yet to fully get a handle on, specifically the thread dimensional tolerances. For instance, according to a thread table on the net, the Major diameter of the external thread (bolt) for M22x1.5 can range from 21.968 to 21.732mm for a 6g fit. Similarly, the minor diameter has a tollerance ranging from 20.344 to 19.980mm. Same same for the corresponding nut which would have a 6H fit.

My question is, how do I work out from the tables the thread depth? Previously I had only made a bolt and choose a major diameter "midway" in the tollerance range and cut the thread depth until I achieved a nice fit on the existing nut but this time I plan on making the bolt and the nut.

Also, how (if any) does the thread depth change with the major diameter? For instance, is the thread depth the same regardless if the dimensions of the major & minor diameters are at either end of the tollerance scales?

Any help would be much appreciated.

Cheers,

Simon

mahgnia
10th May 2015, 04:40 PM
Simon,

It would depend on how much thread engagement you need, which would depend on what you are using the nut and bolt for.
The actual position, sizing and thread face shape of the thread in relation to the centreline (axis) of the bolt and nut do not change; the variances in major and minor diameters only affect the amount of thread engagement between the nut and the bolt.
For most purposes you can use major and minor values midway between the given values, but clearly the minor and major diameters on the nut will need to be large than those on the thread to prevent interference.

I would make the nut first and then make the bolt thread to give a good fit, unless you need other existing nuts or internal threads to fit the bolt.
It wouldn't matter exactly what dimensions you use if nothing else needs to fit your nut and bolt.

Andrew.

simonl
10th May 2015, 05:23 PM
Hi Andrew,

Thanks. When making stuff like threads etc. I prefer to make them compatible with the outside world. Eventhough on this case it would not need to be so. WRT the thread engagement etc. I thought that was taken care of with the class of fit, for instance a 6g 6H which is a standard fit.

I think I may have since answered my own question. The depth of thread = 5/8 x H where H = √3/2 x pitch. This gives a figure of 0.811mm (depth of thread for bolt) however I found another figure of 0.920mm for the bolt and 0.812mm for the nut which is different from what I calculated. Which is correct for metric pitch of M22 x 1.5mm in a 6g 6H class of fit?

Cheers,

Simon

welder
10th May 2015, 07:12 PM
Simon this is the way I work out thread depth not sure of the tolerance , you need to remember two constants.
For an external thread its .6134 x pitch
Example 0.6134x 1.5
= 0.9201mm
For an Internal thread use .5412x pitch
Example 0.6134x1.5
= 0.811 mm
Hope this makes sense I am on my phone. :doh:

simonl
10th May 2015, 07:52 PM
Much appreciated Andre,

Thanks mate.

Simon

cba_melbourne
10th May 2015, 08:05 PM
Simon,

you want to cut an external M22x1.5 6G thread.

This is a metric fine thread. And with metric fine threads, you need not to worry what its diameter is. You only need to worry what its pitch is and what is the tolerace you are going to use. This is how its is, a metric fine thread is completely and soley defined by its pitch and the tolerance, and nothing else.

I personally use thread tables, not formulas. This is what my table says about an ISO 6g 1.5mm external thread:

minimum outer diameter: d -0.268
maximum outer diameter: d -0.032
thread depth: 0.920
radius in thread valley: 0.217

d being the nominal thread diameter, in your particular case 22mm. Meaning min diameter 22 - 0.268mm, max diameter 22 - 0.032mm.

And this also answers your question "how does thread depth change with thread diameter". The thread depth of a metric fine thread does not change at all with thread diameter, it only changes with pitch. However, the thread depth is different for bolt and nut thred.

.RC.
10th May 2015, 09:01 PM
You need some thread wires... Then measure with the three wire system as you go.... It is the only way to really be able to cut accurate threads when it matters...

Ausee in melbourne sell a chinese set, but they are not as good as the US made Pee Dee wires which are around $20 a set..

bob ward
10th May 2015, 09:55 PM
I use the infeeds tables at the bottom of this link. http://www2.coromant.sandvik.com/coromant/pdf/Metalworking_Products_061/tech_c.pdf

Richard, I should get myself some thread wires, where does one obtain the Pee Dee variety?

.RC.
10th May 2015, 10:27 PM
I use the infeeds tables at the bottom of this link. http://www2.coromant.sandvik.com/coromant/pdf/Metalworking_Products_061/tech_c.pdf

Richard, I should get myself some thread wires, where does one obtain the Pee Dee variety?

I got mine from Little Machine Shop in the US...

Steamwhisperer
11th May 2015, 08:13 AM
Hi Simon,
I agree with the use of thread wires as it is really the only way you can get any dimensional accuracy (if required) when screwcutting. Mind you I 'still' don't have a set but the one Richard pointed us to is likely to be my next purchase.
In the meantime I normally go down to Spotlight armed with my micrometer and the ball park size of the thread wire I need to measure the thread I am cutting and buy three sewing machine needles to suit.
It is surprising how accurate they are for the job at hand. Probably worth a crack until your thread wires arrive.
I may also have a brand new tap here that is the size you are after that I could post out to you. I have several I think so it would be yours for the keeping.

Phil

simonl
11th May 2015, 08:49 AM
Hi Richard,

Yes it looks like I should look into getting some thread wires but they are designed to get you spot on go - no go, you still need to know the depth of thread to get you to ball park figures. Looks like I'm shopping for thread wires…. :D

CBA, Thanks for the info. It had never occured to me that the thread dimensions were purely derived from the pitch. It makes sense.

Hi Phil,

Thanks mate for the generous offer but I would feel bad relieving you of an (expensive) item for my one off use. Besides, if I keep relying on taps then I will never have a need to finally do some internal thread cutting. Sewing needles eh? I love spotlight, it's such an underated store for males!

Hi Bob, are you talking about the infeed rates (ap) at C7 where it says 0.94mm for 1.5mm metric external pitch and 0.90mm for internal nut? These figures are still slightly different to others quoted. Not by much though.

Simon

Steamwhisperer
11th May 2015, 08:56 AM
Hi Phil,

Thanks mate for the generous offer but I would feel bad relieving you of an (expensive) item for my one off use. Besides, if I keep relying on taps then I will never have a need to finally do some internal thread cutting. Sewing needles eh? I love spotlight, it's such an underated store for males!



Simon

No worries Simon, they were given to me in the first place though I do understand what you are saying about cutting threads yourself.
I actually consider Spotlight my pseudo engineering supply store :D

Phil

simonl
11th May 2015, 09:04 AM
For anyone else interested, heres a link to the set mentioned by Richard from Ausee on their ebay store. Out of interest, how do they compare to the more expensive brand sold at The little Machine Shop?
http://www.ebay.com.au/itm/Thread-Measuring-Wire-Set-with-Holder-/301515804203?pt=LH_DefaultDomain_15&hash=item4633be0e2b

Simon

bob ward
12th May 2015, 03:45 AM
.........................

Hi Bob, are you talking about the infeed rates (ap) at C7 where it says 0.94mm for 1.5mm metric external pitch and 0.90mm for internal nut? These figures are still slightly different to others quoted. Not by much though.

Simon


No, right down the bottom, page C26 and on.

simonl
12th May 2015, 12:00 PM
No, right down the bottom, page C26 and on.

Thanks Bob, found them. They are the same values as stated before, 0.94mm for external and 0.90mm for internal. These are slightly different to the values obtained using Andre's calculations. It's probably neither her nor there but it intetrests me why the variation.

Anyway, I cut the M22x1.5mm thread using my "usual" technique. I then took it out and bought a standard M22x1.5 nut to test fit and found it fits like a glove. Of course once you remove it from the chuck it's pretty difficult to make any changes if you find it a little tight. I see the benifit of thread wires. No guess work! Mind you, if you make your major diameter within the specified tolerance and you feed in at the correct depth, how can you get it wrong anyway? It's just a recipe.

Simon

cba_melbourne
12th May 2015, 12:48 PM
Thanks Bob, found them. They are the same values as stated before, 0.94mm for external and 0.90mm for internal. These are slightly different to the values obtained using Andre's calculations. It's probably neither her nor there but it intetrests me why the variation.

Anyway, I cut the M22x1.5mm thread using my "usual" technique. I then took it out and bought a standard M22x1.5 nut to test fit and found it fits like a glove. Of course once you remove it from the chuck it's pretty difficult to make any changes if you find it a little tight. I see the benifit of thread wires. No guess work! Mind you, if you make your major diameter within the specified tolerance and you feed in at the correct depth, how can you get it wrong anyway? It's just a recipe.

Simon

That is why I find it best NOT to set the compound to 29.5 degrees. If one leaves the compound parallel to the bedways, one sets the depth of cut exactly as per the cross slide dial - no math needed, no chance for errors, no thread wire needed..... Of course, on a lighter lathe one has to advance the compound sideways a bit for every pass, to ensure tool cuts on the shoulder. Then final cut is a plunge cut. And as you say, turning the major diameter to tolerance is dead easy. Advancing the cross slide to no more than the specified thread depth is very easy too, proper thread depth is almost guaranteed (provided the basics are right such as tool is sharp and at center height and lubricated and machine is not totally worn out and one does not pick up the phone:D halfway through the job....)

Definitely no need for thread wires, unless you are really after a precision thread (for something other than just a fastener thread). And even then, this 3 wire technique is cumbersome to use at best. A go-nogo thread hauge is far easier to use for repetition work.

WoodBee2
12th May 2015, 04:47 PM
I have read this thread with interest. Two questions came up. I understand the use of thread wires and gauges, but aren't thread micrometers possible too?
And if you swivel the compound, wouldn't a indicator give you the actual infeed too?
Thanks,

Peter

pipeclay
12th May 2015, 05:30 PM
Thread mics would be best in critical situations,although a little expensive,the same as GO/NOGO gauges
If you swivel the compound you then need to check depth off a chart or do the calcs associated with it..

Ropetangler
12th May 2015, 06:35 PM
Thread mics would be best in critical situations,although a little expensive,

Generally more than a little expensive, particularly if you want top name brads like Mitutoyo etc. Bearing in mind that while a 0 to 1"mic or 0 to 25MM metric equivalent, with interchangeable anvils will measure quite a few different threads, the sizes up (1-2" say) cover a much smaller range. Also the micrometers often come with only one or two anvil sets, and a few anvil sets can cost more than the micrometer, so it can get expensive pretty quickly. Easy to blow a grand or so if you have it to spare.
Having said all that, there are some interesting offerings on yank ebay right now, and some with lower than usual pricing I would have thought. If you search with the term thread micrometer, a set of PD thread wires, +1-2"Mity mike and a set of angle gauges comes up, all for about AU$65 BIN +$26 or thereabouts postage. Very tempting, I'm not sure how I'm resisting:U
Rob.

.RC.
12th May 2015, 07:31 PM
Mind you, if you make your major diameter within the specified tolerance and you feed in at the correct depth, how can you get it wrong anyway? It's just a recipe.

Simon

Easy to get it wrong if your tool profile is wrong or your lathe is a bit worn and telling fibs on how deep you have cut..

Thread micrometers are probably OK, but I have not seen a set yet that have all the anvils required, plus the micrometers only do 25mm steps..

Quite a few tips and tricks on the usual forums on how to use thread wires and hold them.... Tasmanians find it easier as they can use their tail to hold the third wire :D

Ropetangler
12th May 2015, 08:26 PM
.. Tasmanians find it easier as they can use their tail to hold the third wire :D

+, with 2 heads we're twice as smart! :D:D

snapatap
12th May 2015, 09:14 PM
One method I use when I have full profile inserts is to cut the thread until the od is a bit under the nominal size. Works great 99% of the time. The other 1% I blame the bloke who made the nut .

simonl
13th May 2015, 07:44 AM
That is why I find it best NOT to set the compound to 29.5 degrees. If one leaves the compound parallel to the bedways, one sets the depth of cut exactly as per the cross slide dial - no math needed, no chance for errors, no thread wire needed..... Of course, on a lighter lathe one has to advance the compound sideways a bit for every pass, to ensure tool cuts on the shoulder. Then final cut is a plunge cut. And as you say, turning the major diameter to tolerance is dead easy. Advancing the cross slide to no more than the specified thread depth is very easy too, proper thread depth is almost guaranteed (provided the basics are right such as tool is sharp and at center height and lubricated and machine is not totally worn out and one does not pick up the phone:D halfway through the job....)

Definitely no need for thread wires, unless you are really after a precision thread (for something other than just a fastener thread). And even then, this 3 wire technique is cumbersome to use at best. A go-nogo thread hauge is far easier to use for repetition work.

I've never tried plunge cutting a thread. I have just assumed the results would be disappointing since my lathe is no monster. I don't set my compound to 29.5 degrees, I did when I first started learning and wondered why my thread profiles looked like a saw tooth. Then I realised that my particular lathe manufacturer saw fit to make 0 degrees when the compound is parallel to the bed ways! :(( So it's 60.5 degrees for me, well actually I have no idea what I set it to with any real accuracy but it's a smige more than 60, which equates to a little less than the required 30 degrees ( for a 60 degree V thread)

From the required thread depth (minus 0.05mm) I work out the compound depth required from the hypotenuse of that triangle and multiply in that correction factor, feed in that amount. Then I plunge the last 0.05mm with the cross slide to achieve the full form from the tool.

It's a system that seems to work well for me but I have always had the corresponding threaded part to test fit and adjust the depth if needed. This will not always be the case and so thread wires seem to be a good investment.

Simon

cba_melbourne
13th May 2015, 10:18 AM
I've never tried plunge cutting a thread. I have just assumed the results would be disappointing since my lathe is no monster. I don't set my compound to 29.5 degrees, I did when I first started learning and wondered why my thread profiles looked like a saw tooth. Then I realised that my particular lathe manufacturer saw fit to make 0 degrees when the compound is parallel to the bed ways! :(( So it's 60.5 degrees for me, well actually I have no idea what I set it to with any real accuracy but it's a smige more than 60, which equates to a little less than the required 30 degrees ( for a 60 degree V thread)

From the required thread depth (minus 0.05mm) I work out the compound depth required from the hypotenuse of that triangle and multiply in that correction factor, feed in that amount. Then I plunge the last 0.05mm with the cross slide to achieve the full form from the tool.

It's a system that seems to work well for me but I have always had the corresponding threaded part to test fit and adjust the depth if needed. This will not always be the case and so thread wires seem to be a good investment.

Simon



Simon, I am not sure if you did understand the technique for threading with the compound set parallel to the ways. This is not a single plunge cut. The thread is cut in several passes. And you actually cut on one edge of the cutting tool, just like with the other 29.5 degree technique. It is just the way how you achieve this that is different. You cut one pass with the left cutting edge, next with the right, and so forth. Only the very last cut is a very light plunge cut to get the profile right. Works well on light hobbylathes.

Maybe it gets clearer if I list in sequence for your M22x1.5g6 thread

- OD is finished to tolerance, like 20.4 to 20.6mm for g6
- advance threading tool until tip just touches the OD (use cigarette paper if needed)
- zero cross slide dial (of course without slackening the backlash)
- zero compound slide
- go to thread start and engage threading gear
- advance cross slide for first pass say 0.25mm and cut first pass
- slacken cross slide and return to thread start
- offset the compound slide by 0.1mm to the left (that is 0.1mm past the backlash to the left)
- advance cross slide to 0.5mm and cut second pass
- offset the compound slide by 0.1mm to the right (that is 0.2mm past the backlash to the right)

and so forth, until you come to the last pass, where you put the compound to 0 (always accounting for the backlash of your slide) and do a last plunge cut (plunge cut = cutting equally on both sides of the tool).

The amount of passes that you do depend on the pitch, as well as how rigid the lathe, material etc.
I choose by feel and experience as I go. But there are tables that one can use as a guide for the first time. This is what such a thread cutting table says for M1.5

pass 1 in-feed advance 0.26mm
pass 2 advance 0.23mm
pass 3 0.17mm
pass 4 0.14mm
pass 5 0.12mm
pass 6 0.06mm (final plunge cut)

For other pitches, these are recommended figures for number of passes and progressive in-feed for each pass

M1.25 - 6 passes 0.22, 0.19, 0.14, 0.12, 0.10, 0.06
M1.5 - 6 passes 0.21, 0.17, 0.13, 0.11, 0.06
M1 - 5 passes 0.21, 0.17, 0.13, 0.11, 0.06
M0.75 - 4 passes 0.19, 0.15, 0.11, 0.05
M0.5 - 4 passes 0.12, 0.1, 0.07, 0.05

Of course one can calculate the exact amount of compound side offset required for each pass, but it really is not that critical and after a short time you do that just by feel and experience.

simonl
14th May 2015, 08:46 AM
Sorrt mate. I know what you mean now. I have heard of that technique. Its interesting how everyone gas their own technique they like and get godd results with. Ive used my method which I picked up from reading the workshop series on threading (they list 3 methods from memory) and ive got to a stage where its now routine, even if I havent threaded for 12 months.

Cheers.

Simon

PDW
14th May 2015, 10:38 AM
+, with 2 heads we're twice as smart! :D:D

Well, perhaps...

How do you arrive at the IQ of a crowd?

Take the average IQ of a single person then divide by the number of people in the crowd.

Apply this logic to double-headed Tasmanians and what is the result?

Risdon Vale.

PDW

simonl
14th May 2015, 11:47 AM
Well, perhaps...

How do you arrive at the IQ of a crowd?

Take the average IQ of a single person then divide by the number of people in the crowd.

PDW

That maths does not seem right….. :rolleyes:

Simon

PDW
14th May 2015, 04:56 PM
That maths does not seem right….. :rolleyes:

Simon

I'm only an immigrant Tasmanian, I don't have to be able to count to 2....

PDW

.RC.
14th May 2015, 05:49 PM
I'm only an immigrant Tasmanian, I don't have to be able to count to 2....

PDW

Computer programmers can only count 0 and 1 can they not... I cannot program computers so never got past 0... :) In any case this is Qld here, still the 1950's so no computers..

PDW
14th May 2015, 08:33 PM
Computer programmers can only count 0 and 1 can they not...

Depends on what number base you work in...... a lot of Tasmanians can work in base 12 by counting on their fingers.....

PDW

.RC.
18th May 2015, 06:47 PM
For anyone else interested, heres a link to the set mentioned by Richard from Ausee on their ebay store. Out of interest, how do they compare to the more expensive brand sold at The little Machine Shop?
http://www.ebay.com.au/itm/Thread-Measuring-Wire-Set-with-Holder-/301515804203?pt=LH_DefaultDomain_15&hash=item4633be0e2b

Simon

I just compared measurements from Pee Dee to Ausee.. The Ausee ones were a bit undersized on all the ones I measured... But only by say 5 microns if my micrometer can be trusted... The Pee Dee ones seem to be very close to on size..

I just used them the other day, had to make a puller to pull a diesel injector.. The injector had a 22 X 1.5 thread, so cut the thread on the lathe and used the thread wires to get to size... They are a bit of a pain to try to hold though and when using use the fixed anvil of the micrometer on the two wires and the moving anvil on the single wire...

Ropetangler
19th May 2015, 12:30 AM
I have never used thread wires Richard, but seem to remember grease being suggested to hold the wires while you rounded them all up between the micrometer anvils. Did you try grease or is there a good reason to avoid its use, perhaps it interferes with the accuracy of the measurement.
Rob

BaronJ
19th May 2015, 12:50 AM
I seem to remember someone suggested sticking the wires into a lump of Blutac to hold them whilst measuring took place.

Michael G
19th May 2015, 07:35 AM
I used mine the other day and found that a strip of masking tape helped keep track of them all. (two close in on one end and the third on the other end of a loop)

Michael

simonl
20th May 2015, 06:25 PM
I just compared measurements from Pee Dee to Ausee.. The Ausee ones were a bit undersized on all the ones I measured... But only by say 5 microns if my micrometer can be trusted... The Pee Dee ones seem to be very close to on size..

I just used them the other day, had to make a puller to pull a diesel injector.. The injector had a 22 X 1.5 thread, so cut the thread on the lathe and used the thread wires to get to size... They are a bit of a pain to try to hold though and when using use the fixed anvil of the micrometer on the two wires and the moving anvil on the single wire...

I was going to buy the Ausee ones, me being a tight and all but was not sure if they came with tables. I emailed the seller asking such but never got an answer. Under such circumstances I'm inclined not to buy if they can't even be bothered answering a simple question. I think Pee Dee deserve the sale more since they are upfront with what you get.

Simon

WoodBee2
26th May 2015, 06:17 PM
Re thread micrometers and use of indicator:
Thanks for the replies, sorry for the late response, I have been elsewhere for some time.
I understand the cost issue, but as it happens I have been very lucky to buy a chest full of all kinds of micrometers and indicators for what is essentially peanuts. Among them two thread micrometers (0-25 and 25-50mm) using the same (almost full) set of tips.
Have not used any before, that's why I am curious if they can do what thread wires an gauges can do. (any special considerations when using these?) I understand gauges are faster to use in a production environment, but you would need quite a set, especially if using odd combinations of diameter and thread distance. (there were some in same box too btw). As I understand, there is no need for wires now I have these micrometers?
Guess I should really start single pointing threads to find out how to really measure threads.

As for the use of an indicator:
My understanding from watching YouTube video's is that when putting a indicator (dashboard gauge?) Against your compound slide square to the ways, it will show the real infeed/thread depth when the compound is angled 29,5° (or any other angle for that matter)

I understand the dial on the handwheel gives no direct reading in this application, but according to my little experiments and understanding of matters this should work fine. Any experiences and/or opinions?

Cheers,
Peter