TurboCAD Forums

The Ultimate Resource for TurboCAD Knowledge

Register
 
Interested in some really terrific mobile apps? Visit www.turboapps.com for details.

Gears
Read 6497 times
* May 12, 2015, 09:13:13 PM
Hi,

I am considering an upgrade from my 11.2 Pro to the latest Pro (perpetual license) and TC offered me a pretty good deal. The only reason I'm considering the upgrade is to help me complete a project that involves gear sets that consist of external tooth and internal tooth gears. My 11.2 Pro does a great job of generating external tooth spur gears that I export in .dxf format to a waterjet vendor for some of the prototype gears I need.

My question is:
Has anyone used Turbocad to create accurate internal tooth gears (not 'sort of look-like' gears )?  I'm not really sure, but I think both the internal tooth gear and the mating external tooth gear are a 'pair' at design time.

Any help would be appreciated.

gmeast

Logged


May 12, 2015, 11:26:54 PM
#1
Have a look here

http://forums.turbocad.com/index.php/topic,16537.msg97443.html#msg97443

Or perhaps read the answers before asking the same question again.

Best Regards
Bob
« Last Edit: May 12, 2015, 11:28:54 PM by Bobich »

Logged
TC 14 Deluxe, TC 19 Pro
Windows 7, 64bit 8GB Ram


* May 13, 2015, 07:32:01 AM
#2
Hi,

I am considering an upgrade from my 11.2 Pro to the latest Pro (perpetual license) and TC offered me a pretty good deal. The only reason I'm considering the upgrade is to help me complete a project that involves gear sets that consist of external tooth and internal tooth gears. My 11.2 Pro does a great job of generating external tooth spur gears that I export in .dxf format to a waterjet vendor for some of the prototype gears I need.

My question is:
Has anyone used Turbocad to create accurate internal tooth gears (not 'sort of look-like' gears )?  I'm not really sure, but I think both the internal tooth gear and the mating external tooth gear are a 'pair' at design time.

Any help would be appreciated.

gmeast

The TurboCAD gear profile is really a sort of look-like gear, consisting of lines and arcs but no true involute curves -- but it's close enough for many low speed applications.

Also, be aware that the gear tool in v2015 can't produce more than 100 teeth. I don't know what the limit is in v11.2, but a moment ago in that version the tool drew for me a gear with 10,000 teeth!

Henry H
« Last Edit: May 13, 2015, 07:38:27 AM by Henry Hubich »

Logged


* May 13, 2015, 02:42:41 PM
#3
Have a look here

http://forums.turbocad.com/index.php/topic,16537.msg97443.html#msg97443

Or perhaps read the answers before asking the same question again.

Best Regards
Bob

Searching for answers isn't always that easy in these forums IOW search parameters don't get you where you want to go every time.

I did, however, find a true on-line generator. It does a good job but doesn't seem to be complete because it can only generate Draft output and not Normal or High rendering and doesn't do the apex of a tooth and the trough clearances properly for internal spur gears ... but it's a start:
http://hessmer.org/gears/InvoluteSpurGearBuilder.html

Logged


* May 14, 2015, 03:59:44 PM
#4
Hi gmeast,

There are many reasons to upgrade TurboCAD but the accuracy of the gear contour tool isn’t one of them.

The gear contour tool profile in version 19 and earlier is more accurate than later versions. Drawing a gear of say 23 teeth, 2DP and 20º pressure angle and explode the result allows examination of the lines making up the flank profile. Ignore the line adjacent to the root fillet as this does not form part of the involute profile. The start of the next line is on the base circle. There can be no involute profile inside the base circle.

In version 19 there are 10 lines making up the involute profile. The one nearest the base circle is the shortest and each successive line is noticeably longer than the previous with the longest line of all at the tip of the tooth. This suggests that the calculation of points on the involute are calculated using increments of 1/10 of the angle subtended by a line from the gear centre to the origin of the involute and a line from the gear centre to the point of tangency with the base circle of a line to the tip of the teeth. This is the most accurate method because the curvature of the involute is greater near the base circle and this methods provides more points in that area. Of course, smaller increments and more points will give greater accuracy still.

Repeating the exercise with version 20, 21, 2015 (multiple explodes required) reveals there are 5 lines of approximately equal length making up the profile. This suggests that the calculation of points on the involute are calculated using increments of 1/5 of the difference in radius between the base diameter and tip diameter. Even ignoring the fact that there are fewer increments, this approach provides fewer points where the curvature is greatest, and therefore, the profile accuracy must be worse than earlier versions (see attachment).

If the gear contour tool were to used a spline through points approach, then both curves would have much greater accuracy.

Present in v19 and earlier, but removed from v20 onwards, is an option to use polylines with arcs to represent the involute. This uses a three point arc alternative to an involute curve. The three points used to define the radius are at the base circle, on or near the reference circle and tip circle. This is the least accurate of the three methods because it flattens the involute curvature near the base circle and exaggerates curvature near the tip.

The question we have to ask ourselves is, does it matter? Generally, it does not. The gear contour tool is not there to design gear teeth. It does not have that capacity. Rather it is there to provide illustrative spur gear contours which, if you don’t use numbers of teeth in the undercut zone or care little about root fillets, it does adequately. Compliance with gear tolerance standards doesn’t come into it.

See attachment

David

Logged


* May 14, 2015, 04:58:32 PM
#5
Hi all,

No one has answered my original question.

My original question  was:
"Has anyone used Turbocad to create accurate internal tooth gears ... ?"

TC11.2 doesn't do Internal Tooth Spur Gears ... or I don't know how to use the tool. It does, however, do a great job on external tooth spur gears.

Again: Has anyone used Turbocad to create accurate internal tooth gears?

Thanks,

gmeast

Logged


* May 14, 2015, 06:51:05 PM
#6
Hi all,

No one has answered my original question.

My original question  was:
"Has anyone used Turbocad to create accurate internal tooth gears ... ?"

TC11.2 doesn't do Internal Tooth Spur Gears ... or I don't know how to use the tool. It does, however, do a great job on external tooth spur gears.

Again: Has anyone used Turbocad to create accurate internal tooth gears?

Thanks,

gmeast

How accurate? What is the intended application for the internal gear(s)?

Henry H

Logged


* May 14, 2015, 08:43:26 PM
#7
V2015 draws one circular arc to represent the entire involute curve of a tooth flank. (Note that exploding an arc will always reduce it to a set of line segments.) The resulting accuracy is p-poor, although it could be good enough for some real-world applications. Attached screenshot shows a TurboCAD v2015 gear profile (green) compared to a spline drawn through six carefully plotted points (white crosses) on a true involute curve.

Henry H

Logged


* May 14, 2015, 10:00:01 PM
#8
Thanks for your input Henry Hubich. Again, has anyone drawn an annulus using Turbocad (any version)? ... a simple question yet to be answered.

As an example:
The following link takes you to an online generator that creates involute tooth profile data for any involute gear - internal or external teeth - and allows you to download a .DXF file. The output can have 3 resolutions ... Draft, Normal and High. The problem with this generator is that it does not output a High resolution .DXF files for internal tooth spur gears, so it's of little use to me because I need to send the file to my waterjet vendor so he can cut my prototype gears. My Turbocad 11.2 creates great external tooth spur gear profiles that run great in the real world even though they are generated as polylines and arcs ... amazingly well running gears. 

http://hessmer.org/gears/InvoluteSpurGearBuilder.html

Again, any help would be appreciated,

gmeast


Logged


* May 14, 2015, 11:12:42 PM
#9
The simple answer is that TC doesn't have a tool to generate any internal gear, in any version.  If anyone has created such a gear in TC, it was done manually.  Nobody's claiming to have done so, so the default answer is no.

Logged


* May 15, 2015, 05:01:47 AM
#10
The simple answer is that TC doesn't have a tool to generate any internal gear, in any version.  If anyone has created such a gear in TC, it was done manually.  Nobody's claiming to have done so, so the default answer is no.

Hello Murry Dickinson,

Thank you so very much for your reply. In speaking with Turbocad sales, the individual I spoke with assured me tha I can accurately draw "... any kind of gear I can imagine." Obviously that was just a sales pitch. I guess Turbocad is best suited for architectural design, but mechanical engineering - not so much.

Thanks again. You have saved me about $500.00,

gmeast

Logged


* May 15, 2015, 06:46:59 AM
#11
re: …any kind of gear I can imagine.

I think that statement is basically true, if you draw the teeth individually. What they don't supply is a tool to do this by entering a few numbers. They need to upgrade their Gear Contour tool to encompass the other types of gears.

Logged
John R.

V17—V21, 2015—2019
Designer, Deluxe, (Professional, Expert, Basic), Platinum
RedSDK enabled
Windows 10 Pro (1803), 64-bit


* May 15, 2015, 07:54:01 AM
#12
In TurboCAD, you can produce an internal gear that's approximately as accurate as TCad's spur gears, and I'd expect it to work about as well in the real world. The procedure is to subtract from the blank a spur gear with the same pitch and number of teeth as the required internal gear and then increase both the inside diameter and the outside diameter to prevent interference at the tips. My edition of Machinery's Handbook recommends that the ID be calculated by subtracting 2 from the number of teeth and dividing the result by the diametral pitch. (Subtract a cylinder of this diameter from the annulus.) I'd say that increasing the OD by the same amount is reasonable, but this isn't quite as simple as adjusting the ID. One approach is to work on a single tooth, extending the curve of the flanks outward the required distance, then radial-copy the modified profile.

...We should bear in mind that while involute-profile gearing is pretty nearly ideal, both theoretically and in practice, other tooth shapes have been used successfully where smoothness and/or strength aren't critical. Familiar applications include old-fashioned grist mills in which the gear teeth were simple wooden pegs, and the hand-cranked egg beater, which used stamped and bent sheet-metal teeth.

Henry H

EDIT 16May: I just tried my own suggestion for creating a 48-tooth 4DP internal gear and mated it with a 12-tooth v11 spur. Moved the spur .020" away from its nominal position to provide a little backlash (to compensate for fabrication inaccuracy), and rolled the two together. Found substantial interference and saw no obvious quick-and-easy way to fix it. (See attachment.)

« Last Edit: May 16, 2015, 07:46:07 PM by Henry Hubich »

Logged


* May 15, 2015, 09:43:05 AM
#13
re: …any kind of gear I can imagine.

I think that statement is basically true, if you draw the teeth individually. What they don't supply is a tool to do this by entering a few numbers. They need to upgrade their Gear Contour tool to encompass the other types of gears.

Hi John R,

I agree. TC's gear profile tool in my 11.2 acts like a generator in that (like you said) "... by entering a few numbers."

The generator I dropped a link to earlier generates an internal tooth gear when you enter a (-) plus the number of teeth IOW say, -44 teeth as the tooth count. But it also requires you to also declare the size of the mating gear. So there, you would enter two tooth values, -44 for the annulus and maybe +15 for the mating gear and of course the other stuff like pressure angle, pitch etc. But as I said, the output file is Draft resolution.

Just for fun, I entered a (-) number of teeth in my TC's tool and it looks like someone started to address this in the software because it does generate a ring with a hidden line for the PD when you are dragging it onto the workspace, but when you click the mouse button it doesn't print anything.

gmeast

Logged


* May 15, 2015, 12:03:01 PM
#14
Just a question, as I know nothing of gear profiles. But if the link you stated can generate the proper internal tooth and save as a DXF, can't you then apply the proper materials and parameters to render it in TC? Even if the DXF file is 2D, you can give it a thickness and apply a material for rendering.

Logged
TC 21 Deluxe, TC 20 Platinum, TC 2015 Platinum, TC 2016 Platinum, TC 2017 Platinum
i7- 3770 Ivy Bridge 3.4 GHz, 16G Crucial Ballistic, ASRock Extreme 4, EVGA 1060 SSC, Windows 10 Pro 64 Bit


* May 15, 2015, 02:28:29 PM
#15
Here's a comparison of a v11.2 gear profile, using the "Simple Polyline" option (Cyan), compared to a carefully plotted true involute curve (White). The vertices of the v11 Polyline appear to lie precisely on the true involute. For this 1 DP tooth, the maximum deviation of the Polyline from the true involute looiks to be about .003".

The v11 "Polyline with Arcs" option is no more accurate than the v2015 approximation, illustrated in my earlier post.

The v11 tool seems to have essentially no practical limit to the number of teeth; it produces a 10,000-tooth gear without a hiccup.

Henry H

Logged


* May 15, 2015, 05:27:18 PM
#16
Again: Has anyone used Turbocad to create accurate internal tooth gears?

In my opinion, if I may, a gear can be whatever you want it to be.  If in something like aerospace, exactness is the rule but for a lot of applications a gear can be mated however it needs to be to function.
Here is a link to a gear company that designs gears to suit:  www.kleissgears.com
P.S.  Gears are like threads,  They'll work even if the specs aren't exact.
« Last Edit: May 16, 2015, 08:37:51 AM by John B. »

Logged
John B.
TCv2019 Platinum
Windows 10 Pro 64 Bit
1 GB Hard Drive 16 GB RAM
AMD Radeon RX 580 Series


* May 16, 2015, 04:14:03 PM
#17
Hi gmeast,

Let’s see if we can move this topic on a little.

Don’t get fixated on a method of manufacture that seems to offer a low cost solution to your design problem. The most common (and where necessary the most accurate) methods of gear manufacture do not require a gear tooth profile to follow.

Your design and method of manufacture will depend on the application. Is it a 50 kW drive for a machine tool with high acceleration torque, frequent reversals a life of 10 years with a 70% duty cycle? Or is it a gearbox for a hoist system with intermittent operation for lifting engines out of vehicles up to a capacity of say 5000 lbs? Or is it for a windmill driven animated garden ornament?

Water jet cut gears will be unsuitable for the first, probably unsuitable for the second but may be ok for the third.

Internal gear systems are more difficult to design than external systems. The combination of tooth numbers is critical. The avoidance of tip and root interference is necessary. In order to overcome involute interference the pinion and annulus are usually corrected. This means that the teeth are cut on blanks slightly larger than standard. This in turn means that the glimmer of hope offered by Henry Hubich of subtracting an equivalent external gear from the annulus won’t work because the gear contour tool cannot draw corrected gears. No, you can’t scale them because the profile is altered.

Forgive me if I am wrong but it looks to me that you don’t have the necessary knowledge to design these gears. Therefore, my advice to you is to contact a recognised gear manufacturer in the jobbing/prototyping area of activity who will more than likely design the gears for you and make machine cut prototypes at modest cost.

David
« Last Edit: May 16, 2015, 04:27:09 PM by David »

Logged


* May 16, 2015, 07:39:46 PM
#18
"The avoidance of tip and root interference is necessary. In order to overcome involute interference the pinion and annulus are usually corrected. This means that the teeth are cut on blanks slightly larger than standard. This in turn means that the glimmer of hope offered by Henry Hubich of subtracting an equivalent external gear from the annulus won’t work because the gear contour tool cannot draw corrected gears."

Yup, I have to agree. Please see the amendment to my Reply #12.

Henry H
« Last Edit: May 16, 2015, 07:41:19 PM by Henry Hubich »

Logged