[klaas123]

NOT possible. No network link out there.

Cheers
Roger

Use your smartphone as WiFi access point?

[RCaffin]

Use your smartphone as WiFi access point?

So far I have managed to avoid buying (and paying for) a smart phone. I hope to stay that way.
Workshop is a closed galvanised iron shed. Could be difficult anyhow.

Cheers
Roger

[johnsattuk]

So far I have managed to avoid buying (and paying for) a smart phone. I hope to stay that way.
Cheers
Roger

Me too :cheers:

[louieatienza]

I know, I know.

That said, if I can get the equation for the impellor surface, I should be able to program it parametrically.
Might take a while tho ...

Cheers
Roger

Somehow I highly doubt that an impeller manufacturer would share that information, though there are quite a few models on the 'net to use. Even if you found a simpler design, positioned it in 3D CAM , and indexed around to machine the blade, it would be impressive. Or you can draw the blade profile in 2D and "fake" it by wrapping the Y axis around... a blisk would be a cool thing to machine that you could probably do with Y axis wrapping (or just program in 2.5D, then set your A axis as Y, change the steps to match the circumference of your blank...)

[RCaffin]

Hi Louie

> there are quite a few models on the 'net to use.
I looked. There are, aren't there?
Yes, it could be done if you have the mathematical model, and NOT by using an STL model. STL is good for scanned objects and similar, but many engineering objects are better programmed from a functional perspective.

Oh well, I have to go and sort out why part A is not matching part B properly. 1st guess is an error in one of the programs :-)

Cheers
Roger

[louieatienza]

Hi Louie

> there are quite a few models on the 'net to use.
I looked. There are, aren't there?
Yes, it could be done if you have the mathematical model, and NOT by using an STL model. STL is good for scanned objects and similar, but many engineering objects are better programmed from a functional perspective.

Oh well, I have to go and sort out why part A is not matching part B properly. 1st guess is an error in one of the programs :-)

Cheers
Roger

I'm not a fan of STLs either... Unless I have a RepRap printer I guess. I have found a few models in IGS, STEP, and SLDPRT, so I guess you need to look harder LOL!

A set of hrlical gears would be equally impressive, or even a hypoid or spiroid gearset.

[mcphill]

Here's a nice SolidWorks model:

https://grabcad.com/library/turbocharger-compressor-1

Attachment 276304

As for internet access, how far is "out there"? Would a 200' CAT5 cable get there? How about a Yagi antenna? http://www.instructables.com/id/Easy...-Yagi-Antenna/

[RCaffin]

> Would a 200' CAT5 cable get there?
Maybe, maybe not. Close. (It's a farm you see.)

> How about a Yagi antenna?
Not through galvanised iron.

Basically, I have no incentive to hook my CNC machine to the net (and a lot of incentive to keep it off). I do half my CAD work inside, and then fine tune at the CNC machine. A USB sneaker net is just fine. I repeat, I do not need the cloud. It offers me nothing.

Cheers
Roger

[daniellyall]

most programs will be cloud based soon you could have fusion inside and fusion on your machine computer up to you I will put fusion on my machine computer and see what happens after the time limit is up 30 days or 20 days what ever it is.

I like to do the same thing fix a problem at the machine and not have to go back inside I do it with V carve pro

[RCaffin]

A big unknown is - do the vendors care about us smaller customers, or are they entirely focused on the big customers? It's a worry.

Cheers
Roger

[louieatienza]

A big unknown is - do the vendors care about us smaller customers, or are they entirely focused on the big customers? It's a worry.

Cheers
Roger

Well, what do you want for free? It never truly is, you pay for it by being tethered like an umbilical cord. I'd be afraid to read their terms and conditions, only God knows what you sign away when you click the "agree" box!

[daniellyall]

its about the same as any autodesk product I think with fusion they are trying to corner the hobby market what they could do if they got the drawing part of fusion better, there are something`s v carve pro is better at doing than fusion but the cam side is almost the same as HSM whats the big plus for me and where most of the data is kept its in amazons server farm where autodesk cant even get at it as it under your password.

[Eldon_Joh]

RCaffin do you have any thoughts on the double enveloping worm gear.

I have some thoughts on how to make one, but it would take a lot of time, though it seems to me this should be pretty easy.

The thoughts I have.. Take an off the shelf worm drive and taper lock a plate on the output shaft.
Braze an acme insert or grind an HHS lathe bit to your profile of choice and fit it in a slot milled radially in the plate.
(For a two start worm gear you need two inserts, 180 degrees apart on the plate)
Bolt a plate across the top of the lathe bit to hold it in place, and install a 10-32 or similar screw behind the lathe bit to permit you to incrimentally push it away from the axis of rotation in arbitrarly fine increments.

Find two spur or helical gears with sufficient center distance. (Not critial diameter.)
Install one spur or helical gear on the worm drive screw. (does not have to be the same, ratio can be changed here)

the second spur or helical gear must be ridgidly fixed to the worm screw stock to be cut. (cut it to fit the general profile first on any lathe to reduce the amount of stock to remove)
it should be clear that you can move the gear closer to, or away from the axis of the worm gear in the gear box, but this will change the height, meaning you will have to slide the plate holding your cutting bit up and down on the gear box output shaft, anyhow, this is relatively trivial, and you can fit shims under the lathe bit to change its height.--the cutting bit should be in the same plane as the axis of the screw to be cut.

Anyhow it should be simple to see that the plate holding the cutting bit will rotate in sync with the worm screw to be cut. you will have to start and stop the machine to incriment the radial position of the cutting bit each time it makes a pass.

once the worm screw is cut, you reverse the process to cut the worm gear.

the worm screw for a double enveloping worm gear is relatively large diameter, so i don't see how it would be a problem to reverse the process litterally..
however, you could cut relief slots in the worm screw, then harden it and use the usual proceess to hob the worm screw.

[RCaffin]

Hi Eldon

I am not sure I followed your description sufficiently well. I rely a lot on visual input. Would you have a sketch diagram?

But standing back a long way from the problem, it seems to me that any gear set which has one metal surface sliding over another is going to have wear, backlash, or both. That is independent of how the gears are made.

This of course suggests that making the wheel out of a suitable engineering plastic might help, because these can take a lot more sliding without wearing. One could run that against a hardened worm of whatever sort. However, plastics are far more compliant or bendy. You might get rid of the backlash but end up with a lot of compliance.

The harmonic drive with a crossed roller baring is ALL rolling elements. There are no sliding ones.

I haven't answered your question, I know. I need a diagram.

Cheers
Roger

[louieatienza]

Hi Eldon

I am not sure I followed your description sufficiently well. I rely a lot on visual input. Would you have a sketch diagram?

But standing back a long way from the problem, it seems to me that any gear set which has one metal surface sliding over another is going to have wear, backlash, or both. That is independent of how the gears are made.

This of course suggests that making the wheel out of a suitable engineering plastic might help, because these can take a lot more sliding without wearing. One could run that against a hardened worm of whatever sort. However, plastics are far more compliant or bendy. You might get rid of the backlash but end up with a lot of compliance.

The harmonic drive with a crossed roller baring is ALL rolling elements. There are no sliding ones.

I haven't answered your question, I know. I need a diagram.

Cheers
Roger

Check out ConeDrive...

[RCaffin]

Hi Louie

> Check out ConeDrive...
I did.

Forgive me, but all I got was a huge amount of marketing spin and zero technical data at all. A 1900s company which assumes the customer is ignorant and should just hand over his money and not ask questions.
OK, maybe that's an exaggeration, but apart from 1 page marketing glossies their web site told me NOTHING, except that they do nice paint jobs on the boxes. Dinosaur stuff.

As far as I can see, their much-vaunted double enveloping worm drives are nothing more than shaped worms. Been there, seen that, lots of times before. And the backlash is much worse. You could probably do better with a nylon or acetal worm wheel.

Ah - except for the fact that they are now offering a clone of the Harmonic Drive system. Ho hum. I guess the patents have expired.

Cheers
Roger

[Eldon_Joh]

drawing a diagram would be rather difficult.

so imagine you're cutting a worm screw on a lathe. many of them are just acme thread.

for a "cone" drive instead your cutting tool has to rotate with the worm screw, synchronized with the spindle.
this means your tool post has to be gear driven, and rotate, to cause the cutting tool to travel around in a circle(on the same radius as the worm gear you intend to use or manufacture), rather than sliding the carriage, you rotate the tool post into and sweep it out of the worm screw.

i think the cone drive tooth profile is sufficiently close to a standard worm gear you could use a stock gear, grinding the cutting tool to fit the gear.
the only difference is the literature i read specified that the worm gear has flat sides rather than an involute profile.

[RCaffin]

Hi Eldon

OK, I have sorted out the double enveloping bit. I understand what you mean by rotating the tool too (I think).
I am not a skilled mechanical old-school machinist. I would do it on a CNC lathe or with a rotary table instead.

BUT - you would still have backlash, just the same as with a plain straight worm. For metal to metal gears, you MUST have clearance. Manufacturing details and design details don't make any difference to that. (You might get a higher load rating of course.)

My 2c.
Cheers
Roger

[louieatienza]

Hi Louie

> Check out ConeDrive...
I did.

Forgive me, but all I got was a huge amount of marketing spin and zero technical data at all. A 1900s company which assumes the customer is ignorant and should just hand over his money and not ask questions.
OK, maybe that's an exaggeration, but apart from 1 page marketing glossies their web site told me NOTHING, except that they do nice paint jobs on the boxes. Dinosaur stuff.

As far as I can see, their much-vaunted double enveloping worm drives are nothing more than shaped worms. Been there, seen that, lots of times before. And the backlash is much worse. You could probably do better with a nylon or acetal worm wheel.

Ah - except for the fact that they are now offering a clone of the Harmonic Drive system. Ho hum. I guess the patents have expired.

Cheers
Roger

Interestingly, Harmonic Drive has a planetary gearhead with a flex ring gear, which is designed to take up slack in the system. Cool idea, don't have one.

Wondering if you had experienced the "torque ripple" effect on your drive? I haven't tested mine under power or speed, but it does seem to move pretty smooth turning by hand.

Judging by the ConeDrive's construction, they're probably geared (no pun intended) towards higher torque applications. They do have a version with a large through bore which would be really nice for a rotary axis.

[RCaffin]

Wondering if you had experienced the "torque ripple" effect on your drive? I haven't tested mine under power or speed, but it does seem to move pretty smooth turning by hand.

Torque ripple - no. But I don't think that I would see it in my sort of 4th-axis machining operations anyhow. Position ripple - I don't think so either, at least not to any significant level. The hysteresis curves I showed in the final part were taken at single step increments, with 1000 steps per degree. There was nothing really visible in the results.

Judging by the ConeDrive's construction, they're probably geared (no pun intended) towards higher torque applications. They do have a version with a large through bore which would be really nice for a rotary axis.

Yeah, high torque applications are a bit different. Very often they go only in one direction in actual use. Probably the wrong thing for a 4th axis.

Cheers
Roger