GT2, GT3 and GT5 Timing Gear Creator

[looo]

the basic shapes of gt2, gt3, gt5 are now available from the gear-workbench. I used the geometry from the sketch, but I am still not sure if this is the correct way.

[ulrich1a]

the basic shapes of gt2, gt3, gt5 are now available from the gear-workbench. I used the geometry from the sketch, but I am still not sure if this is the correct way.

I tried to understand the how these belts and corresponding pulleys are defined. I think, there is a need to a clear naming of the applied profile. I found profile names GT and GT2 and GT3. GT is a abbreviation of "Gates tooth". The number is added to account for different generations of the profile and has nothing to do with the pitch size.

See here for history of profile GT: https://ww2.gates.com/IF/facts/documents/Gf000212.pdf
This seems to be the related patent for the GT-profile: https://pdfpiw.uspto.gov/.piw?PageNum=0 ... %2F4605389
It corresponds the introduction date of the GT-profile, as stated in this document on page A6: http://dpk3n3gg92jwt.cloudfront.net/gat ... ergrip.pdf
Gates uses different names like Polychain and Power grip together with "GT", which I do not understand.

Here is a reference to the name GT3: http://www.gatesunitta.com/en/powergrip-gt3-

GT-profiles are available in pitch sizes of 2, 3 an 5 mm. The script unfortunately mixes up profile generations names with the pitch width.

The above patent uses a different profile as five connected circular arcs. Unfortunately the dimensions are given in ranges only. So it will be difficult to generate the actual sold profile from the published information.
A table for profile interchangebility is given here: https://www.pfeiferindustries.com/docum ... ility).pdf

From my view there is at least a requirement to more accurate state which type of profile is generated with the script. It may be not of type "GT".

Ulrich

[looo]

the basic shapes of gt2, gt3, gt5 are now available from the gear-workbench. I used the geometry from the sketch, but I am still not sure if this is the correct way.

I tried to understand the how these belts and corresponding pulleys are defined. I think, there is a need to a clear naming of the applied profile. I found profile names GT and GT2 and GT3. GT is a abbreviation of "Gates tooth". The number is added to account for different generations of the profile and has nothing to do with the pitch size.

See here for history of profile GT: https://ww2.gates.com/IF/facts/documents/Gf000212.pdf
This seems to be the related patent for the GT-profile: https://pdfpiw.uspto.gov/.piw?PageNum=0 ... %2F4605389
It corresponds the introduction date of the GT-profile, as stated in this document on page A6: http://dpk3n3gg92jwt.cloudfront.net/gat ... ergrip.pdf
Gates uses different names like Polychain and Power grip together with "GT", which I do not understand.

Here is a reference to the name GT3: http://www.gatesunitta.com/en/powergrip-gt3-

GT-profiles are available in pitch sizes of 2, 3 an 5 mm. The script unfortunately mixes up profile generations names with the pitch width.

The above patent uses a different profile as five connected circular arcs. Unfortunately the dimensions are given in ranges only. So it will be difficult to generate the actual sold profile from the published information.
A table for profile interchangebility is given here: https://www.pfeiferindustries.com/docum ... ility).pdf

From my view there is at least a requirement to more accurate state which type of profile is generated with the script. It may be not of type "GT".

Ulrich

Thanks for the research. I simple used the script and the drawing of the belts as references.

In the origin script the different gears are defined via different parameters. This image already shows that gt2 is not a pitch=2 gates tooth.

Code: Select all

# GT2 Gear Data 
		self.gt2 = {'pitch':2.0,
			'u':0.254,
			'h':0.75,
			'H':1.38,
			'r0':0.555,
			'r1':1.0,
			'rs':0.15,
			'offset':0.40,
			'name':'GT2',
			'group':'GT'
			}
		# GT3 Gear Data
		self.gt3 = {'pitch':3.0,
			'u':0.381,
			'h':1.14,
			'H':2.40,
			'r0':0.85,
			'r1':1.52,
			'rs':0.25,
			'offset':0.61,
			'name':'GT3',
			'group':'GT'
			}
		# GT5 Gear Data
		self.gt5 = {'pitch':5.0,
			'u':0.5715,
			'h':1.93,
			'H':3.81,
			'r0':1.44,
			'r1':2.57,
			'rs':0.416,
			'offset':1.03,
			'name':'GT5',
			'group':'GT'
			}

So this is not only a change of the pitch. Anyway, I think we need to do some more research on this kind of gears to get them right. Normally someone who needs a good solution creates an issue on github and things get optimized. But it's important to have something available to get the mass-optimization started.
Looking at this image it seems like there are a lot of different types:

[ulrich1a]

Anyway, I think we need to do some more research on this kind of gears to get them right.

Yes, here is a start: https://law.resource.org/pub/in/bis/S01 ... 7.2009.pdf
But the corresponding ISO Standard seems to be retracted.
The following document provides on page 36 a list of standards for belt drives: http://dcc.optibelt.com/fileadmin/inhal ... bon-TM.pdf

The US-standard "RMA IP-27" seems to be related to the current script. Just search for this standard. It seems to have all needed data. What is missing is the correspondence between the standardized profiles G, H, R, S and J and the names of the suitable belts on the market.

Ulrich

[SO_yeah]

Hi all!
I made a 64 teeth GT2 pulley using the v1.4 macro (I also checked with the Gear workbench that I was getting the same timinggear shape) but my 3D printed results slip 1 teeth every 32 teeth.

This is a +3% error but I checked other dimensions of my 3D prints and discrepancies are much below 1%.

are you also experiencing this?

GT2 64T

IMG20210529092717.PNG (319.74 KiB) Viewed 3297 times

[SO_yeah]

I got a pretty decent result by editing some parameters of the Macro ('u', 'h', 'offset'). it was all based on trial and error attempting to reduce the circumference by 3% but preserving the tooth profile.

Code: Select all

		# GT2 Gear Data 
		self.gt2 = {'pitch':2,
			'u':0.8,
			'h':0.74,
			'H':1.38,
			'r0':0.555,
			'r1':1.0,
			'rs':0.15,
			'offset':0.43,
			'name':'GT2',
			'group':'GT'
			}

It feels like it got too small, maybe 'u'=0.7 would have been better.
hope this can be useful to you!

IMG20210604084217.jpg (298.93 KiB) Viewed 3205 times

[jimmihenry]

Yesterday i made this... (big image @ the bottom). Now i wanted to check if my creation is true according to the hardware, by chance and found this topic here .
Searching the internet for some dimensions i found a 20 teeth GT2 pulley, with a diameter (D) of 12,22mm. It was a huge difference to my calculation that was
D 12,732395447351626861510701069801. Yes it is long number with a significant difference to 12,22.

My calculations are bollocks! I am about 0,50799544735162686151070106980115mm off what do i miss? Thank you in advanced.

Never mind got it. "big image" updated.

[kelfox]

In advance, thank you very much for this great post and excuse my ignorance because my doubt is the following: When placing the code in the python panel in freecad it throws me the following error.
.
.
.
... # -----------------------------------
... class GTWnd(object):
File "<input>", line 11
class GTWnd(object):
^
SyntaxError: invalid syntax

I have only used freecad for a few jobs. I'm still learning the software and although I have some knowledge of C programming, I don't have much idea about python. I will be grateful to get your help.
Greetings.

[Syres]

When placing the code in the python panel in freecad it throws me the following error.

That's because a macro such as this isn't designed to be used through the Python Console. You should use the Macro menu>Macros>Create button
Give it a suitable name and copy the text into the Editor. As it's been a while since version 1.4 you may find it's not at it's best visibility wise using a recent Stylesheet but still functions from my testing.

[switt]

Very nice. This newbie has been playing with v1.4 to build a 100 tooth GT2. It's 10mm thick so am putting in holes to save filament. I see that v1.4 was created 4/30/19 and some code tweeks above. Should I be using a newer version or tweeking code before slicing and printing?