Where is the best thread making video featuring correct method?

First off, I recommend you make your blue thread profile curve run from mid-valley to mid-valley and recreate the thread. That way both ends will coincide with your solid core. At the moment, with it starting a little way up the S-curve any attempt to combine thread with core is going to leave you with gaps at both ends.

Will try to follow up with further tips as time allows.


1 Like

So these bolts look like they were formed by thread-rolling. Most mass-produced bolts are done by some form of thread-rolling, and are not machined.

In thread-rolling, a cylindrical blank (possibly with conical reliefs and/or chamfers) squeezed against a die, and the threads are created by cold-forming. In thread-rolling, sometimes the dies are round with helical threads. Other times the dies are flat and “unrolled” versions of the round dies. They’re flat, with straight (but slightly diagonal) grooves cut into them. The blank rolls across the die, and the straight groves (cold) form the threads.

I think you can find a video of this on YouTube. Search for something like,“How It’s Made Bolts.”

Not sure how to model this in CAD, because the thread-rolling process actually pushes metal away from the valleys and towards the peaks. The circled red area is where the die starts to slowly engage the blank (the peaks of the die here are more shallow). I think the bolt blank has a conical taper in this region that expands from the pitch-diameter to the full-diameter of the unthreaded part.

But if you ignore that, you could union the thread with a cone? Maybe good enough?

–Anthony K. Yan

P.S. I admit to having watched too many episodes of How It’s Made (Discovery Channel). Most, if not all the episodes are on YouTube these days (unofficially).

1 Like

Buy a machinist handbook?

There’s already lots of threads and videos now on this matter, not to mention data all over the internet (that’s already been mentioned referenced and linked to your threads), and you’re still not content?

If I or someone else puts a day or a week or a month of time explicating this to you further in a professionally edited video on a silver platter and on youtube what would they gain? and would you be content then? have you studied threads for 10,000 hrs yet? if not what would you gain from some miraculous 10 min video of your dreams? Maybe watch videos on how to make threads IRL? Then study and read the machinist handbook? Then study how to use the helix tools and sweep tools in Rhino? Then make me a 10 min video for me to watch and learn what threads are?

It doesn’t need to be solid. Rhino is “freeform” surface modeler, not a “solid” modeler. You don’t have to use ‘booleans’. I demonstrated booleans in my video that wasn’t to your liking. There’s many ways to make geometric entities. Use a boolean if you want, but you don’t have to.

Doesn’t look like your cutter extends far enough…

this might work…


That’s understandable. But the machinist just needs the specs only. (GD&T’s).

Interesting :sunglasses:

I recommend ‘tap magic’ :smiley:

Not to forget the small irregularities that should be present in a 100% perfect model:

1 Like

Ouch, …wrong side of bed ?..I have been asking for this, and found none, and been sent none to this thread…

well if there was a video on how to model a real thread in Rhino using a chart with diagram for the profile, as I am wanting to do, and achieving at last with folks help, (thank you all) , then someone would have, as I requested, provided a link to it. and they have not found one yet…but you say this quoted at me.

So you say there is not one but many…well please then provide a link to such., as I have asked for at start of this. note the title of the thread.

As such Tom_P has kindly shown what to do after I created the real thread profile.

I have not seen that before anywhere.

and Jeremy kindly tells me how to make the solid, and suggests the profile I originally was going with, valley floor to valley floor, easier to join with the shaft to get a solid. I have not seen that anywhere in these supposed many videos, thank you Jeremy.

I was to make a solid then boolean difference it with the red end shape.

I now await a link to an existing video on Genuine thread making with Rhino, showing all this that you say exists not once but many times out there, perhaps they are all on vimeo ?..rhino modelling a real thread !

Why this tirade ?
All I was looking for was a vid of how to create a genuine thread, is that too much to ask, I had a genuine thread and there was nothing about making one in Rhino.
What was so wrong to seek out such…
Others might also find it useful…but clearly you decide what is of interest to folk.

I was not expecting to be flamed for wanting to see how its done. This has been a very nice and interesting thread, but then whammo.

…and this thread asking for video links…has gone into areas of machining and so on, all very interesting, and I proceed with the Rhino thread making meanwhile matching the item I need to recreate visually, even if I hand over the item with pitch dia and tpi to a machinist. He or she will have something to follow showing how it should end up.

yes it can be all ignored and data sent to a machinist, but it should be possible to recreate a thread in Rhino for visuals as I need.

I chose to opt for Boolean Difference, I like them, Rhino does them so where is the harm in using them ?..Now I am told avoid booleans, as if they are anti Rhino, bad … if I had opted for trimming then I would have been told Booleaan Difference is easier for this,
To have this as a solid then trim away with a cutter should work, or would that approach fail ?
Must remember Rhino is NOT A SOLID MODELLER.

Upset now… :roll_eyes:


Indeed. I prefer using ‘form taps’ when I run mass production CNC cycles on ‘female threads’. The taps last longer, the threads are stronger, etc.

Exactly :sweat_smile:

I was tryna find some work I did on similar textures, but this might be close:

Oh that’s right I think the key word is ‘displacement’ …

Yes I took a jab, cause you have about 3 or 4 permutations of threads asking about threads, and there’s now a major brain storm on this matter, which is fine. But it’s extremely silly to repeatedly say things like “been sent none” or “there’s none” etc. when there’s plenty.

The best way to learn anything is to ‘research it for 10,000 hrs’. It’s a well known testament for the tuition required to learn something. If your strategy is to tap into it here with multiple threads asking about threads and saying there’s no videos explicating threads using the data and charts that’s already at everyone’s fingertips, then sure let’s relentlessly reflect on it all and see where it goes. Maybe we’re only at 1,000 hrs so far in the last month.

If that specific thing pays their bills and puts bread and butter on their table, then you might luck out. I recommend making anew thread with all the specifics. I could probably compile them for you, since I’m finally seeing some signs of said specs.

Where’s the compilation of all your specs? “there are none” :joy: :crazy_face:

Which thread? There’s many permutations now.

There’s literally a link of ‘how it’s made nuts n bolts’ a few posts upwards, in this thread and probably the others too. There’s lots of vids on how to make threads on youtube. All these things help learn how to make threads in 3D CAD’s.

Maybe even a physical science video that explains the mechanical advantage of an ‘inclined plane wrapped around a cylinder’? Hence, ‘helix’.

Yes he did an eloquent job reiterated what others have already mention time n’ time again.

lol 10,000 hrs of trying first, and then let me know if you still find nothing? If I find more time for compiling it all for you or doing a revisionary video etc., I’ll let you know.

Cause you’re clearly not absorbing the data that’s already out there and that members here have been reiterating on this matter for about a month now. And it’s hardly a “tirade”. But let’s just exaggerate that’s okay too.

For someone who clearly doesn’t study the (however sparse) data that exists, let’s act like there’s no data on how to make “genuine thread” and expect members to spoon feed your specific scenario to you on a silver platter in a 10 minute professionally edited video on youtube or now vimeo is it?

“formed”, “cut” or “rolled” thread? I suppose you want them texture displacement manufacturing anomalies too right? That would be sweet :sunglasses:

Just calling you out for apparent lazyness and lack of absorbing the data that’s already been disseminated to you over and over, and your lack of appreciation for what’s already out there.

I am curious where this will lead though. Maybe you’re the ultimate critic who is ok to exaggerate and have discontent for the time and energy that’s been repeated for you on this matter.

Maybe it’s all going somewhere, where you’ll finally be content with how to use the helix and sweep and chamfer tools in Rhino to make threads.

I mean how many ways can a thread be made? “whammo” :crazy_face:

Yes, GD&T’s are essential to manufacture things. Most people don’t comprehend this. They think things are supposed to be flawlessly perfect, and they don’t realize the expense, and disregard for reality of such impossibilities and expectations.

Sending data to a machinist, shouldn’t mean ignore the GD&T’s. Forcing the machinist to figure out your design intent, is probably not a good idea.

Yeah, usually the threads are just explicated via the GD&T specifications, and followed up with some terrible sketch or layout of thread imagery. They’re used to bad drawings and bad thread depiction.

That’s why all they care about is the GD&T’s of your thread. They reall really really don’t care how well you draw the imagery and texture of said “threads”. They just want the parameters and tolerances etc.

Yes and this has been iterated many times in multiple threads in the last several weeks. It is possible, because we’ve seen it time and time again.

That’s fine. There’s no harm. But also not necessary to say:

Unless using ‘boolean’? I suspect it can be done without ‘solid’ but I could be wrong.

It’s all a matter of workflow, and technique, etc…

If a member here recommends a solution to your inquiry and you don’t like it, that’s fine. But I don’t think that means it’s “anti Rhino” or “bad”.

I guess that depends deterministically on the quantum of the 4-dimensional waveline of spacetime.

Booleans are nice, when they’re setup correctly and used with projection into the future – with perspective of what will happen down the line, later in the workflow. Otherwise they can make a mess just like anything else can. Booleans merely have a way of ‘compounding’ a mess into a bigger mess.

It can work, and it’s been demonstrated several times already.

Everything is relative. And for any given phenomena there are infinitely many interpretations.

Don’t be upset. Just try harder to absorb the information that is already there.

I could be imagining it. Relentless reflection and iteration is fine. Except when the information is there. Why say it’s not?

Jim’s mutual trim and join is a better approach than a cutter when working with the thread surfaces rather than solids. You can do something similar to complete the other end, making the purple solid redundant. I think you need surfaces shaped more like the left example. (proportions may not be right).

If you don’t taper this end as well you are left with a little cliff at the end of the thread which doesn’t fit with your photos.

You may need to run the thread further to get a full intersection.


1 Like

and with no replies , I tried the cutter route, just to see how my idea of make a solid then boolean difference it would look,…, before your post appeared.
here is the result.
I turned off the purple solid (screen grab not show this) , instead I sliced planar the thread with two surfaces, capped the now planar ends result and got a solid, then boolean differenced my cutter, placed as it was to match the image above…and behold it matches, well after I rotate the thread 100 degs or so ! hows that, the sloping end does give feather out. :grinning:

Not contemplated the rear end yet.

what do you see as a consequence and bettered by Jim trim to cutter head ?

I see the thread is skimmed a bit on its tip.

TRIM PLANAR the ends then cap ends
Boolean Difference the end


I have asked you to prove the videos on making a real thread in Rhino exist, those that you complain I ignored, the 10,000 etc etc.
and despite me asking you now for proof, you havent linked to a single one.

Lack of videos is why I started this thread for folk like myself wishing to see a video of how its done in Rhino, and I have an essay in reply yet nothing in it showing what I apparently have overlooked in videos available ‘out there’.

so please prove what you are saying, that there is a video on making a real thread in Rhino…
here is the link to that video by CAD CAM CNC.

But no one has linked to such and this thread initially asked for such so I could sit and study it.
I can see the shape, I just wanted Rhino technique on best making a real thread.

I get an essay which says one thing then counteracts it, I am not here for such, I just want folk showing how its done, folk like Tom, Jim and Jeremy are doing so.

At least my posts on how I am achieving it are visual tutorials, something for interested folk to follow.

Can we please keep this thread to how its done PLEASE.
and the alleged thousands of links to others doing same on youtube using Rhino and real Pitch TPI and diameters.

Here is file with saved textures, helixes deleted, and image downsized massiveley. now under 20Mb.

1 hrs work getting it postable, zip made no difference.
Thread make4t recopy again remove threads_dims.3dm (7.9 MB)


1 Like

You’d have to ‘save textures’ if you want others to see the picture:

In real life it’s actually better if this adjacent thread doesn’t get thinned out – imo:

That’s why the whole ‘cone’ theory isn’t the best – imo.

You would probably have to work with ‘threaded rod’ and shortening bolts hundreds or even thousands of times to come to that realization – imo.

The key is to ramp the initial thread without ruining the adjacent threads.

The reason is because the adjacent thread will get thinned out, leading to a sharp edge that can catch and cross-thread the mating piece.

It’s all about probability.

The mating piece should have similar corelative geometry to complement said trade-secret.

If I get time I’ll think about it. Trust me I do think about it. But these things take time.

I agree, a better video can get made.

lol, takes time. I’ll think about it. I’d have to review all the ones that exist first.

You have to dig. The internet isn’t as good as it used to be. Sometimes they’re hidden and shadowbanned, etc.

Oh so then you agree. Affirmative.

Yeah but they’re hard to follow, and They’re not compiled into one thread with one title, and takes more than 10 minutes to study.

Thousands of links is alot. The hours of study are more important than the numbers of links.

Yes, Pitch, and Diameters – those are good for GD&T’s, exactly what one needs to make threads.

now done 1hrs fiddling to redo image then deleted helixes, .

see last post.


when adding an image to a .3dm file , you can save-as and check the ‘save texture’ box during the save, and others can see that picture you placed in the .3dm file – I think.

otherwise it’s blank, and adds an unnecessary entity into the file and adds confusion.

would be better to see ones ‘lined up’ in the file.

a layout with GD&T specs could help too.

and its masive with one small picture, 27mb posting disallowed.
I have had to sit here when tons of things to do, trying to get it under 20Mb, redo imagem, delete out helixes,

good job I am not posting the project file, has 40 or more images, !

There should be a download link option , as 20Mb is mad.

havent time for all this, had hoped to finish that thread today, 12 hrs and still doing it, mostly typing replies on this stuff when all I want is help and methods.

Now to retry and do trim and avoid clipping nearby threads, which prove boolean difference is not the way.

when done I will try and post a storyboard on method !

I like to have something summarised that folk can use. But a video on youtube is best place folk look.

Please everyone realise the object is NOT STANDARD, its not a bolt, or a nut, and just beyond the threads is something too delicate for sticking in a lathe.

I am 99% focused on creating it in Rhino as a visual, it MIGHT get machined. I want Rhino methods on using a chart and making the thread and I am almost there, just a Jim trim and a back end Jeremy surface thing to try.

Lets see someone take my actual file and do a Jim Trim and do that rear end match my photo and dims in first file.


1 Like

I’m compiling some notes, encase I get time to add more to this matter:

I understand. Yes that’s very frustrating. It would be easier to post videos here too without going public on youtube etc.

I have seen the first vid,
as have others, its not good… awful music, turn off speaker, he has swept the ends in, bad practice it is said by folk, then unlike tom_p safe way do half a helix, he sweeps along curve.
He doesnt show making of the profile, then he Boolean Diffs the nut when in fact there is tolerance to add in.

Not helpful, also making the end of the bolt IS WRONG as Jeremy points out the angled milling makes the filter out,.

2nd one , again speaker off not helful when one needs to listen to whats being said, but I dont speal jingly wingly, and what is that, a triangle…ITS MILES FROM WHAT IS NEEDED.

as I say no vids out there on serious thread making.


1 Like

Excellent. It looks like the pictures came through.


Hi Folks, well I have now used the Jim Trim method,
extended the cone, capped it where the true end is.
trimmed away the threads (mine were yet to be joined, so one at a time)
then used threads to trim the cone.

For some reason that cone trim I cant do on Jims. cant see why, he has joined his thread sweep, should make no difference.

I was to post the .3dm file but its now 70Mb.

I save small and add texture, now its 2Mb.

thread_end Jim Rhino thread wont trim red my trim matches.3dm (2.0 MB)

Hope its viewable.
I see the clipped thread (in red) is also there in Jims, so not sure how to avoid that.

How does one avoid the thread on the previous ‘lap’ getting clipped ?

perhaps retract the cone a bit, its 52deg and steeper than a fictitious 45deg.


1 Like

However, its ok at that end, now to do the 0.05 stem at other end.
Follow Jeremy suggestion.
and create the stem with cone taken from front end.
Trim the thread, get some minute slivers showing through I cant trim away.
Place onto blue layer ‘bits’ !
then try to trim cone but cannot do so.

one trim away from completion and yet again thwarted.

entire day on this pesky thread, in more ways than one !

I wonder if Solids better and Boolean Difference.

we have inability to trim cone on Jims, yet I managed it, and inability to trim cone on rear of mine.

some minute slivers wouldnt trim but I placed them on a blue layer, sort of swept them under the carpet !

but the thread looks a little kinky there.

unable to trim cone rear end.3dm (6.9 MB)


I fail to understand why when you are given a method that works and that you understand how to use, but you refuse to use it and instead find some other method that fails and insist on using that. That doesn’t make any sense to me.

It looks to me that you have all the time in the world and are just experimenting with different methods. Trimming with the individual surfaces in the threads might work if that object was composed of surfaces that are accurately connected, but yours aren’t very accurate. The connections between threads are just barely within tolerance and your tolerance is very loose. You might try modeling the threads more accurately and then the trimming and joining as individual surfaces might work.

1 Like

I would have to reveal more about the trade-secret if that’s what you’re referring to. While, I’m sure others know of this and might be something open for debate. So, user preference maybe.